From ac26848e0fcea310549fcc1c4200000c24b82781 Mon Sep 17 00:00:00 2001 From: vorakl Date: Wed, 21 Feb 2024 20:39:24 -0800 Subject: [PATCH] New content --- 404.html | 2 +- allarticles/index.html | 2 +- articles/2019/06/index.html | 2 +- articles/2019/07/index.html | 2 +- articles/2019/08/index.html | 2 +- articles/2019/12/index.html | 2 +- articles/2019/index.html | 2 +- articles/2020/01/index.html | 2 +- articles/2020/04/index.html | 2 +- articles/2020/05/index.html | 2 +- articles/2020/index.html | 2 +- articles/2024/01/index.html | 2 +- articles/2024/02/index.html | 2 +- articles/2024/index.html | 2 +- articles/algorithm/index.html | 2 +- articles/availability/index.html | 2 +- articles/base94/index.html | 2 +- articles/bash-sort/index.html | 2 +- articles/canonical/index.html | 2 +- articles/char-converter/index.html | 2 +- articles/cs-vs-it/index.html | 2 +- articles/csp/index.html | 2 +- articles/data-structure/index.html | 2 +- articles/dreyfus/index.html | 2 +- articles/engineering/index.html | 2 +- articles/goto/index.html | 2 +- articles/index.html | 2 +- articles/learning/index.html | 2 +- articles/maslow/index.html | 2 +- articles/numbers/index.html | 2 +- articles/py-params/index.html | 2 +- articles/remove-webpage-google/index.html | 2 +- articles/smart/index.html | 2 +- articles/sq3r/index.html | 2 +- articles/stream-encoding/index.html | 2 +- articles/turing/index.html | 2 +- articles/udp-link/index.html | 2 +- authors/index.html | 2 +- authors/vorakl/index.html | 2 +- authors/vorakl/p2/index.html | 2 +- authors/vorakl/p3/index.html | 2 +- categories/article/index.html | 2 +- categories/howto/index.html | 2 +- categories/index.html | 2 +- categories/note/index.html | 2 +- categories/note/p2/index.html | 2 +- categories/tutorial/index.html | 2 +- index.html | 2 +- news/index.html | 2 +- news/p2/index.html | 2 +- news/p3/index.html | 2 +- pages/about/index.html | 2 +- pages/contacts/index.html | 2 +- pages/feedback/index.html | 2 +- pages/index.html | 2 +- pages/projects/index.html | 2 +- pages/reading/index.html | 2 +- tags/ai/index.html | 2 +- tags/bash/index.html | 2 +- tags/binary-to-text/index.html | 2 +- tags/cs/index.html | 2 +- tags/encoding/index.html | 2 +- tags/html/index.html | 2 +- tags/http/index.html | 2 +- tags/index.html | 2 +- tags/it/index.html | 2 +- tags/learning/index.html | 2 +- tags/management/index.html | 2 +- tags/math/index.html | 2 +- tags/mindmap/index.html | 2 +- tags/networking/index.html | 2 +- tags/programming/index.html | 2 +- tags/psychology/index.html | 2 +- tags/python/index.html | 2 +- tags/sre/index.html | 2 +- tags/tools/index.html | 2 +- tags/web/index.html | 2 +- theme/bootstrap-pygments.bundle.min.css | 78 ++++++++++++++++++++++- theme/pygments-tango.css | 77 ++++++++++++++++++++++ 79 files changed, 231 insertions(+), 78 deletions(-) create mode 100644 theme/pygments-tango.css diff --git a/404.html b/404.html index ffcadc8f..f6ec22ed 100644 --- a/404.html +++ b/404.html @@ -1 +1 @@ - Not Found :: Vorakl's notes
Vorakl's notes

Not Found

The requested item could not be located.

Perhaps you might want to check all posts or information pages.

\ No newline at end of file + Not Found :: Vorakl's notes
Vorakl's notes

Not Found

The requested item could not be located.

Perhaps you might want to check all posts or information pages.

\ No newline at end of file diff --git a/allarticles/index.html b/allarticles/index.html index 8a351cb6..91d1d468 100644 --- a/allarticles/index.html +++ b/allarticles/index.html @@ -1 +1 @@ - All posts :: Vorakl's notes
Vorakl's notes

All posts

2024-02-20 How to sort arrays natively in Bash
2024-02-18 Availability calculation in "nines" notation
2024-02-17 A little mess with function parameters in Python
2024-01-16 Using udp-link to enhance TCP connections stability
2020-05-13 The zoo of binary-to-text encoding schemes
2020-04-18 Convert binary data to a text with the lowest overhead
2020-01-18 My notes for the "Pragmatic Thinking and Learning" book
2019-12-20 Computer Science vs Information Technology
2019-12-19 Who is an engineer
2019-12-15 Algorithm is...
2019-12-15 Turing: thesis, machine, completeness
2019-08-21 Organizing Unstructured Data
2019-08-16 Number Classification
2019-08-08 A converter of a character's case and Ascii codes
2019-07-22 Constraint Satisfaction Problem (CSP)
2019-07-18 How to remove a webpage from the Google index
2019-07-02 How to redirect a static website on the Github Pages
2019-06-26 Managing your plans in the S.M.A.R.T. way
2019-06-21 Dreyfus model of skill acquisition
2019-06-15 Maslow's hierarchy of needs
2019-06-12 Structured Programming Paradigm
2019-06-10 SQ3R
\ No newline at end of file + All posts :: Vorakl's notes
Vorakl's notes

All posts

2024-02-20 How to sort arrays natively in Bash
2024-02-18 Availability calculation in "nines" notation
2024-02-17 A little mess with function parameters in Python
2024-01-16 Using udp-link to enhance TCP connections stability
2020-05-13 The zoo of binary-to-text encoding schemes
2020-04-18 Convert binary data to a text with the lowest overhead
2020-01-18 My notes for the "Pragmatic Thinking and Learning" book
2019-12-20 Computer Science vs Information Technology
2019-12-19 Who is an engineer
2019-12-15 Algorithm is...
2019-12-15 Turing: thesis, machine, completeness
2019-08-21 Organizing Unstructured Data
2019-08-16 Number Classification
2019-08-08 A converter of a character's case and Ascii codes
2019-07-22 Constraint Satisfaction Problem (CSP)
2019-07-18 How to remove a webpage from the Google index
2019-07-02 How to redirect a static website on the Github Pages
2019-06-26 Managing your plans in the S.M.A.R.T. way
2019-06-21 Dreyfus model of skill acquisition
2019-06-15 Maslow's hierarchy of needs
2019-06-12 Structured Programming Paradigm
2019-06-10 SQ3R
\ No newline at end of file diff --git a/articles/2019/06/index.html b/articles/2019/06/index.html index 4406a25c..84899455 100644 --- a/articles/2019/06/index.html +++ b/articles/2019/06/index.html @@ -1 +1 @@ - June 2019 :: Vorakl's notes
Vorakl's notes

June 2019

2019-06-26 Managing your plans in the S.M.A.R.T. way
2019-06-21 Dreyfus model of skill acquisition
2019-06-15 Maslow's hierarchy of needs
2019-06-12 Structured Programming Paradigm
2019-06-10 SQ3R
\ No newline at end of file + June 2019 :: Vorakl's notes
Vorakl's notes

June 2019

2019-06-26 Managing your plans in the S.M.A.R.T. way
2019-06-21 Dreyfus model of skill acquisition
2019-06-15 Maslow's hierarchy of needs
2019-06-12 Structured Programming Paradigm
2019-06-10 SQ3R
\ No newline at end of file diff --git a/articles/2019/07/index.html b/articles/2019/07/index.html index 69b9279b..e0623e32 100644 --- a/articles/2019/07/index.html +++ b/articles/2019/07/index.html @@ -1 +1 @@ - July 2019 :: Vorakl's notes
Vorakl's notes

July 2019

2019-07-22 Constraint Satisfaction Problem (CSP)
2019-07-18 How to remove a webpage from the Google index
2019-07-02 How to redirect a static website on the Github Pages
\ No newline at end of file + July 2019 :: Vorakl's notes
Vorakl's notes

July 2019

2019-07-22 Constraint Satisfaction Problem (CSP)
2019-07-18 How to remove a webpage from the Google index
2019-07-02 How to redirect a static website on the Github Pages
\ No newline at end of file diff --git a/articles/2019/08/index.html b/articles/2019/08/index.html index b6c9c0d3..de4e4753 100644 --- a/articles/2019/08/index.html +++ b/articles/2019/08/index.html @@ -1 +1 @@ - August 2019 :: Vorakl's notes
Vorakl's notes

August 2019

2019-08-21 Organizing Unstructured Data
2019-08-16 Number Classification
2019-08-08 A converter of a character's case and Ascii codes
\ No newline at end of file + August 2019 :: Vorakl's notes
Vorakl's notes

August 2019

2019-08-21 Organizing Unstructured Data
2019-08-16 Number Classification
2019-08-08 A converter of a character's case and Ascii codes
\ No newline at end of file diff --git a/articles/2019/12/index.html b/articles/2019/12/index.html index 6a2d811a..596c3cbd 100644 --- a/articles/2019/12/index.html +++ b/articles/2019/12/index.html @@ -1 +1 @@ - December 2019 :: Vorakl's notes
Vorakl's notes

December 2019

2019-12-20 Computer Science vs Information Technology
2019-12-19 Who is an engineer
2019-12-15 Algorithm is...
2019-12-15 Turing: thesis, machine, completeness
\ No newline at end of file + December 2019 :: Vorakl's notes
Vorakl's notes

December 2019

2019-12-20 Computer Science vs Information Technology
2019-12-19 Who is an engineer
2019-12-15 Algorithm is...
2019-12-15 Turing: thesis, machine, completeness
\ No newline at end of file diff --git a/articles/2019/index.html b/articles/2019/index.html index 5c8366ee..da5906b4 100644 --- a/articles/2019/index.html +++ b/articles/2019/index.html @@ -1 +1 @@ - 2019 :: Vorakl's notes
Vorakl's notes

2019

2019-12-20 Computer Science vs Information Technology
2019-12-19 Who is an engineer
2019-12-15 Algorithm is...
2019-12-15 Turing: thesis, machine, completeness
2019-08-21 Organizing Unstructured Data
2019-08-16 Number Classification
2019-08-08 A converter of a character's case and Ascii codes
2019-07-22 Constraint Satisfaction Problem (CSP)
2019-07-18 How to remove a webpage from the Google index
2019-07-02 How to redirect a static website on the Github Pages
2019-06-26 Managing your plans in the S.M.A.R.T. way
2019-06-21 Dreyfus model of skill acquisition
2019-06-15 Maslow's hierarchy of needs
2019-06-12 Structured Programming Paradigm
2019-06-10 SQ3R
\ No newline at end of file + 2019 :: Vorakl's notes
Vorakl's notes

2019

2019-12-20 Computer Science vs Information Technology
2019-12-19 Who is an engineer
2019-12-15 Algorithm is...
2019-12-15 Turing: thesis, machine, completeness
2019-08-21 Organizing Unstructured Data
2019-08-16 Number Classification
2019-08-08 A converter of a character's case and Ascii codes
2019-07-22 Constraint Satisfaction Problem (CSP)
2019-07-18 How to remove a webpage from the Google index
2019-07-02 How to redirect a static website on the Github Pages
2019-06-26 Managing your plans in the S.M.A.R.T. way
2019-06-21 Dreyfus model of skill acquisition
2019-06-15 Maslow's hierarchy of needs
2019-06-12 Structured Programming Paradigm
2019-06-10 SQ3R
\ No newline at end of file diff --git a/articles/2020/01/index.html b/articles/2020/01/index.html index ab719a88..403b6667 100644 --- a/articles/2020/01/index.html +++ b/articles/2020/01/index.html @@ -1 +1 @@ - January 2020 :: Vorakl's notes
Vorakl's notes

January 2020

2020-01-18 My notes for the "Pragmatic Thinking and Learning" book
\ No newline at end of file + January 2020 :: Vorakl's notes
Vorakl's notes

January 2020

2020-01-18 My notes for the "Pragmatic Thinking and Learning" book
\ No newline at end of file diff --git a/articles/2020/04/index.html b/articles/2020/04/index.html index 9da920d6..45ce4761 100644 --- a/articles/2020/04/index.html +++ b/articles/2020/04/index.html @@ -1 +1 @@ - April 2020 :: Vorakl's notes
Vorakl's notes

April 2020

2020-04-18 Convert binary data to a text with the lowest overhead
\ No newline at end of file + April 2020 :: Vorakl's notes
Vorakl's notes

April 2020

2020-04-18 Convert binary data to a text with the lowest overhead
\ No newline at end of file diff --git a/articles/2020/05/index.html b/articles/2020/05/index.html index d2f8d496..c8f1346c 100644 --- a/articles/2020/05/index.html +++ b/articles/2020/05/index.html @@ -1 +1 @@ - May 2020 :: Vorakl's notes
Vorakl's notes

May 2020

2020-05-13 The zoo of binary-to-text encoding schemes
\ No newline at end of file + May 2020 :: Vorakl's notes
Vorakl's notes

May 2020

2020-05-13 The zoo of binary-to-text encoding schemes
\ No newline at end of file diff --git a/articles/2020/index.html b/articles/2020/index.html index ffc7a053..53c420e6 100644 --- a/articles/2020/index.html +++ b/articles/2020/index.html @@ -1 +1 @@ - 2020 :: Vorakl's notes
Vorakl's notes

2020

2020-05-13 The zoo of binary-to-text encoding schemes
2020-04-18 Convert binary data to a text with the lowest overhead
2020-01-18 My notes for the "Pragmatic Thinking and Learning" book
\ No newline at end of file + 2020 :: Vorakl's notes
Vorakl's notes

2020

2020-05-13 The zoo of binary-to-text encoding schemes
2020-04-18 Convert binary data to a text with the lowest overhead
2020-01-18 My notes for the "Pragmatic Thinking and Learning" book
\ No newline at end of file diff --git a/articles/2024/01/index.html b/articles/2024/01/index.html index 9fd30c35..7e08ddf9 100644 --- a/articles/2024/01/index.html +++ b/articles/2024/01/index.html @@ -1 +1 @@ - January 2024 :: Vorakl's notes
Vorakl's notes

January 2024

2024-01-16 Using udp-link to enhance TCP connections stability
\ No newline at end of file + January 2024 :: Vorakl's notes
Vorakl's notes

January 2024

2024-01-16 Using udp-link to enhance TCP connections stability
\ No newline at end of file diff --git a/articles/2024/02/index.html b/articles/2024/02/index.html index 87b963a6..1c3a9d7d 100644 --- a/articles/2024/02/index.html +++ b/articles/2024/02/index.html @@ -1 +1 @@ - February 2024 :: Vorakl's notes
Vorakl's notes

February 2024

2024-02-20 How to sort arrays natively in Bash
2024-02-18 Availability calculation in "nines" notation
2024-02-17 A little mess with function parameters in Python
\ No newline at end of file + February 2024 :: Vorakl's notes
Vorakl's notes

February 2024

2024-02-20 How to sort arrays natively in Bash
2024-02-18 Availability calculation in "nines" notation
2024-02-17 A little mess with function parameters in Python
\ No newline at end of file diff --git a/articles/2024/index.html b/articles/2024/index.html index bb538189..747074b2 100644 --- a/articles/2024/index.html +++ b/articles/2024/index.html @@ -1 +1 @@ - 2024 :: Vorakl's notes
Vorakl's notes

2024

2024-02-20 How to sort arrays natively in Bash
2024-02-18 Availability calculation in "nines" notation
2024-02-17 A little mess with function parameters in Python
2024-01-16 Using udp-link to enhance TCP connections stability
\ No newline at end of file + 2024 :: Vorakl's notes
Vorakl's notes

2024

2024-02-20 How to sort arrays natively in Bash
2024-02-18 Availability calculation in "nines" notation
2024-02-17 A little mess with function parameters in Python
2024-01-16 Using udp-link to enhance TCP connections stability
\ No newline at end of file diff --git a/articles/algorithm/index.html b/articles/algorithm/index.html index e1e4b3a7..7bc479e5 100644 --- a/articles/algorithm/index.html +++ b/articles/algorithm/index.html @@ -1 +1 @@ - Algorithm is... :: Vorakl's notes
Vorakl's notes

Algorithm is...

on 2019-12-15 in note about cs mindmap ~2 min read

Common properties of algorithms

Despite the obvious expectation to find some sort of a definition of the term "Algorithm" here, I have to disappoint you, as there isn't any general or well-accepted definition. But, it's not a unique situation! Take mathematics, for example. Although there are plenty of different "definitions" that can be found in the literature, they all are just oversimplified attempts to explain what an algorithm really means.


In general, an algorithm is a way of describing the logic. And that's why it's so hard to cover all possible forms of it in terms of common rules or definitions. Most prominent mathematicians began seriously thinking about computability and what can be computed at the beginning of the 20th century. But it was so hard to generalize all the cases that eventually they had to limit the consideration by functions defined only on the set of Natural numbers.


The most famous works were done by Alan Turing (related to algorithms) and Alonzo Church (related to computable functions). Alan Turing came up with the thesis which basically says, that if a function is computable then it has an algorithm, and if so, then it can be implemented on the Turing machine (TM). In other words, Turing's thesis makes it clear what can be computed and what is needed to get computed.


Turing machine is an abstract system that has a finite set of states and symbols, a few certain operations, and an endless tape (consisted of cells). The behavior of a TM is controlled by a program that defines a state transition and a next tape movement depending on a symbol that was read. Although, there is no a real-world analog of the TM as it is unlikely possible to have infinite memory. So, to get it more realistic, for a real analog of TM, it means two things:

  1. to have enough memory, at least, as much as needed (analog of the tape)
  2. to have a conditional branching, some sort of if/else and goto statements (analog of state transitions)

All algorithms share the same properties:

  • deterministic (produces the same result for the same input)
  • discrete (works with discrete data, like texts, integers, rational numbers)
  • finite (represented by a finite text)

Turing completeness
Found a bug or typo? Please, send me feedback or submit a PR on Github.
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
\ No newline at end of file + Algorithm is... :: Vorakl's notes
Vorakl's notes

Algorithm is...

on 2019-12-15 in note about cs mindmap ~2 min read

Common properties of algorithms

Despite the obvious expectation to find some sort of a definition of the term "Algorithm" here, I have to disappoint you, as there isn't any general or well-accepted definition. But, it's not a unique situation! Take mathematics, for example. Although there are plenty of different "definitions" that can be found in the literature, they all are just oversimplified attempts to explain what an algorithm really means.


In general, an algorithm is a way of describing the logic. And that's why it's so hard to cover all possible forms of it in terms of common rules or definitions. Most prominent mathematicians began seriously thinking about computability and what can be computed at the beginning of the 20th century. But it was so hard to generalize all the cases that eventually they had to limit the consideration by functions defined only on the set of Natural numbers.


The most famous works were done by Alan Turing (related to algorithms) and Alonzo Church (related to computable functions). Alan Turing came up with the thesis which basically says, that if a function is computable then it has an algorithm, and if so, then it can be implemented on the Turing machine (TM). In other words, Turing's thesis makes it clear what can be computed and what is needed to get computed.


Turing machine is an abstract system that has a finite set of states and symbols, a few certain operations, and an endless tape (consisted of cells). The behavior of a TM is controlled by a program that defines a state transition and a next tape movement depending on a symbol that was read. Although, there is no a real-world analog of the TM as it is unlikely possible to have infinite memory. So, to get it more realistic, for a real analog of TM, it means two things:

  1. to have enough memory, at least, as much as needed (analog of the tape)
  2. to have a conditional branching, some sort of if/else and goto statements (analog of state transitions)

All algorithms share the same properties:

  • deterministic (produces the same result for the same input)
  • discrete (works with discrete data, like texts, integers, rational numbers)
  • finite (represented by a finite text)

Turing completeness
Found a bug or typo? Please, send me feedback or submit a PR on Github.
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
\ No newline at end of file diff --git a/articles/availability/index.html b/articles/availability/index.html index 9030b69e..82a7d34a 100644 --- a/articles/availability/index.html +++ b/articles/availability/index.html @@ -1 +1 @@ - Availability calculation in "nines" notation :: Vorakl's notes
Vorakl's notes

Availability calculation in "nines" notation

on 2024-02-18 in article about sre ~4 min read

Estimating one of SRE's most common SLO

TLDR: quick summary of the article


The rapidly growing interest in clouds, distributed systems, microservice architecture, and service-oriented applications has led to the emergence of a new branch of computer systems engineering - Site Reliability Engineering (SRE). One of the primary goals of the SRE is to ensure that a service meets certain requirements for production readiness. Services are generally considered to be production when they can be trusted and relied upon. A service provider and the customers, who usually pay for a service, document a common understanding of trust in a Service Level Agreement (SLA). It contains all expectations in the form of Service Level Objectives (SLO) and penalties if these expectations are not met. SLOs are performance and availability goals for a production service, defined on an annual time scale. These are the system characteristics that are both the most valuable to customers and worth committing to keep them within the defined expectations. SLOs are carefully quantified using Service Level Indicators (SLI). SLIs are chosen specifically for SLOs as a measurable form of some properties. It can be a metric or a value derived from logs. SLIs are typically sampled over a much shorter periods of time, from tens of seconds to a few minutes, and then a mean or an average distribution is applied to obtain a value.


Site Reliability Engineers, in turn, are responsible for ensuring that production services meet all target SLOs defined in the SLA. They do this by focusing on the reliability through a set of practices that are more or less standardized across the industry. Some of the most common practices include:

  • Continuous monitoring of availability and performance characteristics;
  • Troubleshooting failures and eliminating degradation issues;
  • Improving overall stability and scalability through automation to keep all key metrics within expected ranges;
  • Preparing for disaster recovery through continuous stress testing using the error budget, an agreed upon timeframe in which a service can be degraded or unavailable.

Performance SLOs are important goals, but they are only important if a service is available. Availability is so important that it's sometimes mistakenly considered the only SLA component. Finding the right SLI to measure availability can be challenging. It's service-specific and depends on a variety of factors, such as the underlying infrastructure, architecture, etc. In SLO form, availability is expressed as a percentage in what is called "nines" notation. For example, in the clouds, the most common availability SLO is 99.9%, which is called "3-nines". However, you are unlikely to find it higher than 99.999%, or "5-nines". The actual availability of a service in percent is basically calculated as the ratio of the time a service is available to the total uptime (which includes downtime) over the past year.


It is interesting that people who use the nines notation are actually referring to the time when a service is a sort of allowed to be down. This downtime, which is literally allowed by the SLA, forms what is called the error budget. While targeting 100% availability is hardly feasible, it turns out that from a practical point of view, it is more beneficial to commit to a lower availability. Even if all technical possibilities exist to provide more "nines". At certain levels, services with a higher availability will not be noticed by the majority of customers, so it's probably not worth the effort. However, having some error budget opens the doors to experimentation and less stressful deployments of new product features.


It is also useful to know how to estimate a potential downtime, the amount of time when your system may to be out of service. To do this, remember that the availability SLO is defined for a one-year period. Therefore, 60s by 60m by 24h by 365d gives us 31536000 seconds of a total uptime. Then, if the availability is "five-nines" (99.999%), then the downtime is 0.001%, or 31536000 * 0.00001 = 315.36 sec, which is about 5.256 minutes per year that the service can be down. A similar calculation for "three-nines" (99.9%) availability shows that the service can be down for 31536000 * 0.001 = 31536 seconds, or 525.6 minutes, or 8.76 hours per year.


Summary

  • Site Reliability Engineering (SRE) focuses on ensuring production services meet requirements for production readiness and can be trusted and relied upon.
  • A Service Level Agreement (SLA) contains expectations in the form of Service Level Objectives (SLOs) and penalties if not met.
  • SLOs define annual performance and availability goals for production services.
  • Service Level Indicators (SLIs) are metrics chosen to measure SLOs, sampled over short periods like seconds to minutes.
  • SREs ensure services meet SLOs through standardized practices like monitoring, emergency response, and capacity planning.
  • Availability is the most important SLA component and is expressed as percentages or "nines" denoting hours of annual downtime allowed.
  • The 99.9% availability SLO allows 8.76 hours of annual downtime while 99.999% allows 5.256 minutes.
  • Allowing some downtime forms an "error budget" even if 100% uptime is technically possible.
  • Higher availability beyond a certain level may not be noticeable to most customers.
  • Calculating allowed downtime involves determining the total seconds in a year and applying the percentage downtime allowed.
Found a bug or typo? Please, send me feedback or submit a PR on Github.
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
\ No newline at end of file + Availability calculation in "nines" notation :: Vorakl's notes
Vorakl's notes

Availability calculation in "nines" notation

on 2024-02-18 in article about sre ~4 min read

Estimating one of SRE's most common SLO

TLDR: quick summary of the article


The rapidly growing interest in clouds, distributed systems, microservice architecture, and service-oriented applications has led to the emergence of a new branch of computer systems engineering - Site Reliability Engineering (SRE). One of the primary goals of the SRE is to ensure that a service meets certain requirements for production readiness. Services are generally considered to be production when they can be trusted and relied upon. A service provider and the customers, who usually pay for a service, document a common understanding of trust in a Service Level Agreement (SLA). It contains all expectations in the form of Service Level Objectives (SLO) and penalties if these expectations are not met. SLOs are performance and availability goals for a production service, defined on an annual time scale. These are the system characteristics that are both the most valuable to customers and worth committing to keep them within the defined expectations. SLOs are carefully quantified using Service Level Indicators (SLI). SLIs are chosen specifically for SLOs as a measurable form of some properties. It can be a metric or a value derived from logs. SLIs are typically sampled over a much shorter periods of time, from tens of seconds to a few minutes, and then a mean or an average distribution is applied to obtain a value.


Site Reliability Engineers, in turn, are responsible for ensuring that production services meet all target SLOs defined in the SLA. They do this by focusing on the reliability through a set of practices that are more or less standardized across the industry. Some of the most common practices include:

  • Continuous monitoring of availability and performance characteristics;
  • Troubleshooting failures and eliminating degradation issues;
  • Improving overall stability and scalability through automation to keep all key metrics within expected ranges;
  • Preparing for disaster recovery through continuous stress testing using the error budget, an agreed upon timeframe in which a service can be degraded or unavailable.

Performance SLOs are important goals, but they are only important if a service is available. Availability is so important that it's sometimes mistakenly considered the only SLA component. Finding the right SLI to measure availability can be challenging. It's service-specific and depends on a variety of factors, such as the underlying infrastructure, architecture, etc. In SLO form, availability is expressed as a percentage in what is called "nines" notation. For example, in the clouds, the most common availability SLO is 99.9%, which is called "3-nines". However, you are unlikely to find it higher than 99.999%, or "5-nines". The actual availability of a service in percent is basically calculated as the ratio of the time a service is available to the total uptime (which includes downtime) over the past year.


It is interesting that people who use the nines notation are actually referring to the time when a service is a sort of allowed to be down. This downtime, which is literally allowed by the SLA, forms what is called the error budget. While targeting 100% availability is hardly feasible, it turns out that from a practical point of view, it is more beneficial to commit to a lower availability. Even if all technical possibilities exist to provide more "nines". At certain levels, services with a higher availability will not be noticed by the majority of customers, so it's probably not worth the effort. However, having some error budget opens the doors to experimentation and less stressful deployments of new product features.


It is also useful to know how to estimate a potential downtime, the amount of time when your system may to be out of service. To do this, remember that the availability SLO is defined for a one-year period. Therefore, 60s by 60m by 24h by 365d gives us 31536000 seconds of a total uptime. Then, if the availability is "five-nines" (99.999%), then the downtime is 0.001%, or 31536000 * 0.00001 = 315.36 sec, which is about 5.256 minutes per year that the service can be down. A similar calculation for "three-nines" (99.9%) availability shows that the service can be down for 31536000 * 0.001 = 31536 seconds, or 525.6 minutes, or 8.76 hours per year.


Summary

  • Site Reliability Engineering (SRE) focuses on ensuring production services meet requirements for production readiness and can be trusted and relied upon.
  • A Service Level Agreement (SLA) contains expectations in the form of Service Level Objectives (SLOs) and penalties if not met.
  • SLOs define annual performance and availability goals for production services.
  • Service Level Indicators (SLIs) are metrics chosen to measure SLOs, sampled over short periods like seconds to minutes.
  • SREs ensure services meet SLOs through standardized practices like monitoring, emergency response, and capacity planning.
  • Availability is the most important SLA component and is expressed as percentages or "nines" denoting hours of annual downtime allowed.
  • The 99.9% availability SLO allows 8.76 hours of annual downtime while 99.999% allows 5.256 minutes.
  • Allowing some downtime forms an "error budget" even if 100% uptime is technically possible.
  • Higher availability beyond a certain level may not be noticeable to most customers.
  • Calculating allowed downtime involves determining the total seconds in a year and applying the percentage downtime allowed.
Found a bug or typo? Please, send me feedback or submit a PR on Github.
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
\ No newline at end of file diff --git a/articles/base94/index.html b/articles/base94/index.html index 6ecf7799..a126deb2 100644 --- a/articles/base94/index.html +++ b/articles/base94/index.html @@ -1,3 +1,3 @@ - Convert binary data to a text with the lowest overhead :: Vorakl's notes
Vorakl's notes

Convert binary data to a text with the lowest overhead

on 2020-04-18 in article about cs programming binary-to-text encoding ~5 min read

A binary-to-text encoding with any radix from 2 to 94

This article is about binary/text converters, the most popular implementations, and a non-standard approach that uses place-based single number encoding by representing a file as a big number and then converting it to another big number with any non-256 (1 byte/8 bit) radix. To make it practical, it makes sense to limit a radix (base) to 94 for matching numbers to all possible printable symbols within the 7-bit ASCII table. It is probably a theoretical prototype and has a purely academic flavor, as the time and space complexities make it applicable only to small files (up to a few tens of kilobytes), although it allows one to choose any base with no dependencies on powers of two, e.g. 7 or 77.

Background

The main purpose of such converters is to convert a binary file represented by 256 different symbols (radix-256, 1 byte, 2^8) into a form suitable for transmission over a channel with a limited range of supported symbols. A good example is any text-based network protocol, such as HTTP (before ver. 2) or SMTP, where all transmitted binary data must be reversibly converted to a pure text form without control symbols. As you know, ASCII codes from 0 to 31 are considered control characters, and therefore they will definitely be lost during transmission over any logical channel that doesn't allow endpoints to transmit full 8-bit bytes (binary) with codes from 0 to 255. This limits the number of allowed symbols to less than 224 (256-32), but in fact it's limited only by the first 128 standardized symbols in the ASCII table, and even more.


The standard solution today is the Base64 algorithm defined in RFC 4648 (easy reading). It also describes Base32 and Base16 as possible variants. The key point here is that they all share the same property of being powers of two. The wider the range of supported symbols (codes), the more space-efficient the result of the conversion. It will be bigger anyway, the question is how much bigger. For example, Base64 encoding gives about 33% larger output, because 3 input bytes (8 valued bits) are translated into 4 output bytes (6 valued bits, 2^6=64). So the ratio is always 4/3, i.e. the output is larger by 1/3 or 33.(3)%. Practically speaking, Base32 is very inefficient because it means translating 5 input (8 valued bits) bytes to 8 output (5 valued bits, 2^5=32) bytes and the ratio is 8/5, i.e. the output is larger by 3/5 or 60%. In this context, it is hard to consider any kind of efficiency of Base16, since its output size is larger by 100% (each byte of 8 valued bits is represented by two bytes of 4 valued bits, also known as nibbles, 2^4=16). It is not even a translation, just a representation of an 8-bit byte in hexadecimal.


If you're curious how these input/output byte ratios were calculated for the Base64/32/16 encodings, the answer is LCM (Least Common Multiple). Let's calculate it ourselves, and for that we need another function, the GCD (Greatest Common Divisor)


  1. Base64 (Input: 8 bits, Output: 6 bits):
    • LCM(8, 6) = 8*6/GCD(8,6) = 24 bit
    • Input: 24 / 8 = 3 bytes
    • Output: 24 / 6 = 4 bytes
    • Ratio (Output/Input): 4/3
  2. Base32 (Input: 8 bits, Output: 5 bits):
    • LCM(8, 5) = 8*5/GCD(8,5) = 40 bit
    • Input: 40 / 8 = 5 bytes
    • Output: 40 / 5 = 8 bytes
    • Ratio (Output/Input): 8/5
  3. Base16 (Input: 8 bits, Output: 4 bits):
    • LCM(8, 4) = 8*4/GCD(8,4) = 8 bit
    • Input: 8 / 8 = 1 byte
    • Output: 8 / 4 = 2 bytes
    • Ratio (Output/Input): 2/1

What's the point?

The point is this. What if a channel is only able to transmit a few different symbols, like 9 or 17. That is, we have a file represented by a 256-symbol alphabet (a normal 8-bit byte), we are not limited by computing power or memory constraints on either side, but we are able to send only 7 different symbols instead of 256? Base64/32/16 are no solution here. Then Base7 is the only possible output format.


Another example, what if the amount of data transmitted is a concern for a channel? Base64, as it has been shown, increases the data by 33% no matter what is transmitted, always. Base94, for example, only increases the output by 22%.


It may seem that Base94 is not the limit. If the first 32 ASCII codes are control characters, and there are 256 codes in total, what stops you from using an alphabet of 256 - 32 = 224 symbols? There is a reason. Not all of the 224 ASCII codes are printable characters or have a standard representation. In general, only 7 bits (0..127) are standardized, and the rest (128..255) is used for the variety of locales, e.g. Koi8-R, Windows-1251, etc. This means that only 128 - 32 = 96 are available in the standardized range. In addition, the ASCII code 32 is the space character, and 127 doesn't have a visible character either. So 96 - 2 gives us the 94 printable characters that have the same association with their codes on most machines.

Solution

This solution is quite simple, but this simplicity also imposes a significant computational constraint. The entire input file can be treated as a big number with a base of 256. It could easily be a really big number, requiring thousands of bits. Then all we have to do is convert that big number to another base. That's it. And Python3 makes it even easier! Normally, conversions between different bases are done via an intermediate base10. The good news is that Python3 has built-in support for big number calculations (it is built into int). The int class has a method that reads any number of bytes and automatically represents them as a large Base10 number with a desired endian. So essentially all of this complexity can be implemented in just two lines of code, which is pretty amazing!

with open('inpit_file', 'rb') as f:
+ Convert binary data to a text with the lowest overhead :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Convert binary data to a text with the lowest overhead
 
on 2020-04-18 in article about cs programming binary-to-text encoding ~5 min read 
 
A binary-to-text encoding with any radix from 2 to 94
 
 
This article is about binary/text converters, the most popular implementations, and a non-standard approach that uses place-based single number encoding by representing a file as a big number and then converting it to another big number with any non-256 (1 byte/8 bit) radix. To make it practical, it makes sense to limit a radix (base) to 94 for matching numbers to all possible printable symbols within the 7-bit ASCII table. It is probably a theoretical prototype and has a purely academic flavor, as the time and space complexities make it applicable only to small files (up to a few tens of kilobytes), although it allows one to choose any base with no dependencies on powers of two, e.g. 7 or 77.
 
 
Background
 
The main purpose of such converters is to convert a binary file represented by 256 different symbols (radix-256, 1 byte, 2^8) into a form suitable for transmission over a channel with a limited range of supported symbols. A good example is any text-based network protocol, such as HTTP (before ver. 2) or SMTP, where all transmitted binary data must be reversibly converted to a pure text form without control symbols. As you know, ASCII codes from 0 to 31 are considered control characters, and therefore they will definitely be lost during transmission over any logical channel that doesn't allow endpoints to transmit full 8-bit bytes (binary) with codes from 0 to 255. This limits the number of allowed symbols to less than 224 (256-32), but in fact it's limited only by the first 128 standardized symbols in the ASCII table, and even more.
 
 

 
 
The standard solution today is the Base64 algorithm defined in RFC 4648 (easy reading). It also describes Base32 and Base16 as possible variants. The key point here is that they all share the same property of being powers of two. The wider the range of supported symbols (codes), the more space-efficient the result of the conversion. It will be bigger anyway, the question is how much bigger. For example, Base64 encoding gives about 33% larger output, because 3 input bytes (8 valued bits) are translated into 4 output bytes (6 valued bits, 2^6=64). So the ratio is always 4/3, i.e. the output is larger by 1/3 or 33.(3)%. Practically speaking, Base32 is very inefficient because it means translating 5 input (8 valued bits) bytes to 8 output (5 valued bits, 2^5=32) bytes and the ratio is 8/5, i.e. the output is larger by 3/5 or 60%. In this context, it is hard to consider any kind of efficiency of Base16, since its output size is larger by 100% (each byte of 8 valued bits is represented by two bytes of 4 valued bits, also known as nibbles, 2^4=16). It is not even a translation, just a representation of an 8-bit byte in hexadecimal.
 
 

 
 
If you're curious how these input/output byte ratios were calculated for the Base64/32/16 encodings, the answer is LCM (Least Common Multiple). Let's calculate it ourselves, and for that we need another function, the GCD (Greatest Common Divisor)
 
 

 
 
     
  1.  
    Base64 (Input: 8 bits, Output: 6 bits):
     
       
    • LCM(8, 6) = 8*6/GCD(8,6) = 24 bit
      •  
    • Input: 24 / 8 = 3 bytes
      •  
    • Output: 24 / 6 = 4 bytes
      •  
    • Ratio (Output/Input): 4/3
      •  
     
     
     
    2.  
  2.  
    Base32 (Input: 8 bits, Output: 5 bits):
     
       
    • LCM(8, 5) = 8*5/GCD(8,5) = 40 bit
      •  
    • Input: 40 / 8 = 5 bytes
      •  
    • Output: 40 / 5 = 8 bytes
      •  
    • Ratio (Output/Input): 8/5
      •  
     
     
     
    3.  
  3.  
    Base16 (Input: 8 bits, Output: 4 bits):
     
       
    • LCM(8, 4) = 8*4/GCD(8,4) = 8 bit
      •  
    • Input: 8 / 8 = 1 byte
      •  
    • Output: 8 / 4 = 2 bytes
      •  
    • Ratio (Output/Input): 2/1
      •  
     
     
     
    4.  
 
 
 
What's the point?
 
The point is this. What if a channel is only able to transmit a few different symbols, like 9 or 17. That is, we have a file represented by a 256-symbol alphabet (a normal 8-bit byte), we are not limited by computing power or memory constraints on either side, but we are able to send only 7 different symbols instead of 256? Base64/32/16 are no solution here. Then Base7 is the only possible output format.
 
 

 
 
Another example, what if the amount of data transmitted is a concern for a channel? Base64, as it has been shown, increases the data by 33% no matter what is transmitted, always. Base94, for example, only increases the output by 22%.
 
 

 
 
It may seem that Base94 is not the limit. If the first 32 ASCII codes are control characters, and there are 256 codes in total, what stops you from using an alphabet of 256 - 32 = 224 symbols? There is a reason. Not all of the 224 ASCII codes are printable characters or have a standard representation. In general, only 7 bits (0..127) are standardized, and the rest (128..255) is used for the variety of locales, e.g. Koi8-R, Windows-1251, etc. This means that only 128 - 32 = 96 are available in the standardized range. In addition, the ASCII code 32 is the space character, and 127 doesn't have a visible character either. So 96 - 2 gives us the 94 printable characters that have the same association with their codes on most machines.
 
 
 
Solution
 
This solution is quite simple, but this simplicity also imposes a significant computational constraint. The entire input file can be treated as a big number with a base of 256. It could easily be a really big number, requiring thousands of bits. Then all we have to do is convert that big number to another base. That's it. And Python3 makes it even easier! Normally, conversions between different bases are done via an intermediate base10. The good news is that Python3 has built-in support for big number calculations (it is built into int). The int class has a method that reads any number of bytes and automatically represents them as a large Base10 number with a desired endian. So essentially all of this complexity can be implemented in just two lines of code, which is pretty amazing!
 
with open('inpit_file', 'rb') as f:
     in_data = int.from_bytes(f.read(), 'big')
 
 
where in_data is our large Base10 number. This is only two lines, but this is where most of the math happens and most of the time is spent. So now convert it to any other base, as you'd normally do with normal small decimal numbers.
  
 
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/bash-sort/index.html b/articles/bash-sort/index.html
index 9300a2b6..6e74609e 100644
--- a/articles/bash-sort/index.html
+++ b/articles/bash-sort/index.html
@@ -1,4 +1,4 @@
- How to sort arrays natively in Bash :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
How to sort arrays natively in Bash
 
on 2024-02-20 in howto about bash programming ~4 min read 
 
Sorting arrays in pure Bash with the asort built-in command
 
 
TLDR: quick summary of the article
 
 

 
 
What would you do if, while implementing some solution in Bash, you suddenly needed to have an array in a sorted order? You might think of the sort tool from the coreutils package. Or you might even think that it's probably a good time to switch to Python or some other language? But it turns out that Bash supports sorting arrays natively! All you need is the asort built-in command. However, it is often not loaded by default, or even packaged on many modern Linux distributions. In this article I'll show you how to build and install Bash with all loadable modules from source, load them, and start writing faster, more advanced Bash scripts with less use of external commands.
 
 

 
 
First of all, check your Bash version. Version 5.2-release is the target of this article:
 
echo ${BASH_VERSION}
+ How to sort arrays natively in Bash :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
How to sort arrays natively in Bash
 
on 2024-02-20 in howto about bash programming ~4 min read 
 
Sorting arrays in pure Bash with the asort built-in command
 
 
TLDR: quick summary of the article
 
 

 
 
What would you do if, while implementing some solution in Bash, you suddenly needed to have an array in a sorted order? You might think of the sort tool from the coreutils package. Or you might even think that it's probably a good time to switch to Python or some other language? But it turns out that Bash supports sorting arrays natively! All you need is the asort built-in command. However, it is often not loaded by default, or even packaged on many modern Linux distributions. In this article I'll show you how to build and install Bash with all loadable modules from source, load them, and start writing faster, more advanced Bash scripts with less use of external commands.
 
 

 
 
First of all, check your Bash version. Version 5.2-release is the target of this article:
 
echo ${BASH_VERSION}
 
 
 

 
 
The built-in loadable modules are loaded with the enable command. Bash expects to find loadable modules in one of the paths specified in the BASH_LOADABLES_PATH environment variable, which is a colon-separated list of directories. Setting this variable and enabling all the necessary commands can be done, for example, with .bashrc. If you are currently running a pre-installed Bash, check that the asort command is not loaded and it cannot be loaded due to its absence:
 
enable -p | grep asort || { enable -f asort asort && enable -p | grep asort; }
 
 
 

 
 
If you see "enable asort" on the screen then the asort builtin is loaded and you can start using it, for example, by checking its help message:
 
asort --help
 
 
 

 
 
Otherwise, let's build it from source. First of all, clone the project's official git repository and enter its directory:
 
git clone https://git.savannah.gnu.org/git/bash.git && cd bash
diff --git a/articles/canonical/index.html b/articles/canonical/index.html
index a89e8691..c2648474 100644
--- a/articles/canonical/index.html
+++ b/articles/canonical/index.html
@@ -1,4 +1,4 @@
- How to redirect a static website on the Github Pages :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
How to redirect a static website on the Github Pages
 
on 2019-07-02 in howto about it web html http ~4 min read 
 
The use case for a Temporary Redirect and the Canonical Link Element
 
 
I run a few static websites for my private projects on the Github Pages. I'm absolutely happy with the service, as it supports custom domains, automatically redirects to HTTPS, and transparently installs SSL certificates (with automatic issuing via Let's Encrypt). It is very fast (thanks to Fastly's content delivery network) and is extremely reliable (I haven't had any issues for years). Taking into account the fact that I get all of this for free, it perfectly matches my needs at the moment. It has, however, one important limitation: because it serves static websites only, this means no query parameters, no dynamic content generated on the server side, no options for injecting any server-side configuration (e.g., .htaccess), and the only things I can push to the website's root directory are static assets (e.g., HTML, CSS, JS, JPEG, etc.). In general, this is not a big issue. There are a lot of the open source static site generators available, such as Jekyll, which is available by default the dashboard, and Pelican, which I prefer in most cases. Nevertheless, when you need to implement something that is traditionally solved on the server side, a whole new level of challenge begins.
 
 

 
 
For example, I recently had to change a custom domain name for one of my websites. Keeping the old one was ridiculously expensive, and I wasn't willing to continue wasting money. I found a cheaper alternative and immediately faced a bigger problem: all the search engines have the old name in their indexes. Updating indexes takes time, and until that happens, I would have to redirect all requests to the new location. Ideally, I would redirect each indexed resource to the equivalent on the new site, but at minimum, I needed to redirect requests to the new start page. I had access to the old domain name for enough time, and therefore, I could run the site separately on both domain names at the same time.
 
 

 
 
There is one proper solution to this situation that should be used whenever possible: Permanent redirect, or the 301 Moved Permanently status code, is the way to redirect pages implemented in the HTTP protocol. The only issue is that it's supposed to happen on the server side within a server response's HTTP header. But the only solution I could implement resides on a client side; that is, either HTML code or JavaScript. I didn't consider the JS variant because I didn't want to rely on the script's support in web browsers. Once I defined the task, I recalled a solution: the HTML <meta> tag <meta http-equiv> with the 'refresh' HTTP header. Although it can be used to ask browsers to reload a page or jump to another URL after a specified number of seconds, after some research, I learned it is more complicated than I thought with some interesting facts and details.
 
 
The solution
 
TL;DR (for anyone who isn't interested in all the details): In brief, this solution configures two repositories to serve as static websites with custom domain names. On the site with the old domain, I reconstructed the website's entire directory structure and put the following index.html in each of them (including the root):
 
<!DOCTYPE HTML>
+ How to redirect a static website on the Github Pages :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
How to redirect a static website on the Github Pages
 
on 2019-07-02 in howto about it web html http ~4 min read 
 
The use case for a Temporary Redirect and the Canonical Link Element
 
 
I run a few static websites for my private projects on the Github Pages. I'm absolutely happy with the service, as it supports custom domains, automatically redirects to HTTPS, and transparently installs SSL certificates (with automatic issuing via Let's Encrypt). It is very fast (thanks to Fastly's content delivery network) and is extremely reliable (I haven't had any issues for years). Taking into account the fact that I get all of this for free, it perfectly matches my needs at the moment. It has, however, one important limitation: because it serves static websites only, this means no query parameters, no dynamic content generated on the server side, no options for injecting any server-side configuration (e.g., .htaccess), and the only things I can push to the website's root directory are static assets (e.g., HTML, CSS, JS, JPEG, etc.). In general, this is not a big issue. There are a lot of the open source static site generators available, such as Jekyll, which is available by default the dashboard, and Pelican, which I prefer in most cases. Nevertheless, when you need to implement something that is traditionally solved on the server side, a whole new level of challenge begins.
 
 

 
 
For example, I recently had to change a custom domain name for one of my websites. Keeping the old one was ridiculously expensive, and I wasn't willing to continue wasting money. I found a cheaper alternative and immediately faced a bigger problem: all the search engines have the old name in their indexes. Updating indexes takes time, and until that happens, I would have to redirect all requests to the new location. Ideally, I would redirect each indexed resource to the equivalent on the new site, but at minimum, I needed to redirect requests to the new start page. I had access to the old domain name for enough time, and therefore, I could run the site separately on both domain names at the same time.
 
 

 
 
There is one proper solution to this situation that should be used whenever possible: Permanent redirect, or the 301 Moved Permanently status code, is the way to redirect pages implemented in the HTTP protocol. The only issue is that it's supposed to happen on the server side within a server response's HTTP header. But the only solution I could implement resides on a client side; that is, either HTML code or JavaScript. I didn't consider the JS variant because I didn't want to rely on the script's support in web browsers. Once I defined the task, I recalled a solution: the HTML <meta> tag <meta http-equiv> with the 'refresh' HTTP header. Although it can be used to ask browsers to reload a page or jump to another URL after a specified number of seconds, after some research, I learned it is more complicated than I thought with some interesting facts and details.
 
 
The solution
 
TL;DR (for anyone who isn't interested in all the details): In brief, this solution configures two repositories to serve as static websites with custom domain names. On the site with the old domain, I reconstructed the website's entire directory structure and put the following index.html in each of them (including the root):
 
<!DOCTYPE HTML>
 <html lang="en">
     <head>
         <meta charset="utf-8">
diff --git a/articles/char-converter/index.html b/articles/char-converter/index.html
index 080a09df..faeebc5a 100644
--- a/articles/char-converter/index.html
+++ b/articles/char-converter/index.html
@@ -1,4 +1,4 @@
- A converter of a character's case and Ascii codes :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
A converter of a character's case and Ascii codes
 
on 2019-08-08 in article about cs programming python encoding ~5 min read 
 
An example of using the Constraint Programming for calculating multiple but linked results
 
 
The constraint programming paradigm is effectively applied for solving a group of problems which can be translated to variables and constraints or represented as a mathematic equation, and so related to the CSP. Using declarative programming style it describes a general model with certain properties. In contrast to the imperative style, it doesn't tell how to achieve something, but rather what to achieve. Instead of defining a set of instructions with only one obvious way for computing values, the constraint programming declares relationships between variables within constraints. A final model makes it possible to compute the values of variables regardless of direction or changes. Thus, any change of the value of one variable affects the whole system (all other variables) and to satisfy defined constraints it leads to recomputing the other values.
 
 

 
 
Let's take, for example, Pythagoras' theorem: a² + b² = c². The constraint is represented by this equation, which has three variables (a, b, and c), and each has a domain (non-negative). Using the imperative programming style, to compute any of these variables having other two, we would need to create three different functions (because each variable is computed by a different equation):
 
     
  • c = √(a² + b²)
    •  
  • a = √(c² - b²)
    •  
  • b = √(c² - a²)
    •  
 
These functions satisfy the main constraint and to check domains, each function should validate the input. Moreover, at least one more function would be needed for choosing an appropriate function accordingly to provided variables. This is one of possible solutions:
 
def pythagoras(*, a=None, b=None, c=None):
+ A converter of a character's case and Ascii codes :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
A converter of a character's case and Ascii codes
 
on 2019-08-08 in article about cs programming python encoding ~5 min read 
 
An example of using the Constraint Programming for calculating multiple but linked results
 
 
The constraint programming paradigm is effectively applied for solving a group of problems which can be translated to variables and constraints or represented as a mathematic equation, and so related to the CSP. Using declarative programming style it describes a general model with certain properties. In contrast to the imperative style, it doesn't tell how to achieve something, but rather what to achieve. Instead of defining a set of instructions with only one obvious way for computing values, the constraint programming declares relationships between variables within constraints. A final model makes it possible to compute the values of variables regardless of direction or changes. Thus, any change of the value of one variable affects the whole system (all other variables) and to satisfy defined constraints it leads to recomputing the other values.
 
 

 
 
Let's take, for example, Pythagoras' theorem: a² + b² = c². The constraint is represented by this equation, which has three variables (a, b, and c), and each has a domain (non-negative). Using the imperative programming style, to compute any of these variables having other two, we would need to create three different functions (because each variable is computed by a different equation):
 
     
  • c = √(a² + b²)
    •  
  • a = √(c² - b²)
    •  
  • b = √(c² - a²)
    •  
 
These functions satisfy the main constraint and to check domains, each function should validate the input. Moreover, at least one more function would be needed for choosing an appropriate function accordingly to provided variables. This is one of possible solutions:
 
def pythagoras(*, a=None, b=None, c=None):
     ''' Computes a side of a right triangle '''
 
     # Validate
diff --git a/articles/cs-vs-it/index.html b/articles/cs-vs-it/index.html
index 92d3645f..0bf434d5 100644
--- a/articles/cs-vs-it/index.html
+++ b/articles/cs-vs-it/index.html
@@ -1 +1 @@
- Computer Science vs Information Technology :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Computer Science vs Information Technology
 
on 2019-12-20 in note about cs it mindmap ~2 min read 
 
Differences between two computer-related studies
 
 
If you ever thought about getting a computer-related (graduated) education, you probably came across a variety of similar disciplines, more or less connected to each other, but grouped under two major fields of study: Computer Science (CS) and Information Technology (IT). The latest one sometimes comes in a broader meaning - Information Communications Technology (ICT), and Computer Science, in turn, is highly linked to Electrical Engineering. But what exactly makes them all different?
 
 

 
 
Briefly, Computer Science creates computer software technologies, Electrical Engineering creates hardware to run this software in an efficient way, while Information Technology uses them later to create Information Systems for storing, processing and transmitting data.
 
 

 
 
CS is a study of using computation and computer systems for solving real-world problems. Dealing mostly with software, the study includes the theory of computation and computer architecture, design, development, and application of software systems. The most common problems are organized in groups in particular areas, such as Distributed Systems, Artificial Intelligence, Data Science, Programming Languages and Compilers, Algorithms and Data Structures, etc. Summarizing, CS mainly focuses on finding answers to the following questions (by John DeNero, cs61a):
 
     
  • which real-world problems can be solved using computation
    •  
  • how to solve these problems
    •  
  • how to solve them efficiently
    •  
 
 

 
 
The fact that CS is all about software, makes it tightly coupled to Electrical Engineering that deals with hardware and focuses on designing computer systems and electronic devices for running software in the most efficient way.
 
 

 
 
Unlike CS, IT is a study of using computers to design, build, and operate Information Systems which are used for storing and processing information (data). ICT extends it by applying telecommunications for receiving and transmitting data. It is crucial to notice, that IT apply existing technologies (e.g. hardware, operating systems, systems software, middleware applications, databases, networks) for creating Information Systems. Hence, IT professionals are users of technologies and utilize existing solutions (hardware and software) to create larger systems for solving a specific business need.
 
 

 
 Turing completeness 
 

 
 Turing completeness  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ Computer Science vs Information Technology :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Computer Science vs Information Technology
 
on 2019-12-20 in note about cs it mindmap ~2 min read 
 
Differences between two computer-related studies
 
 
If you ever thought about getting a computer-related (graduated) education, you probably came across a variety of similar disciplines, more or less connected to each other, but grouped under two major fields of study: Computer Science (CS) and Information Technology (IT). The latest one sometimes comes in a broader meaning - Information Communications Technology (ICT), and Computer Science, in turn, is highly linked to Electrical Engineering. But what exactly makes them all different?
 
 

 
 
Briefly, Computer Science creates computer software technologies, Electrical Engineering creates hardware to run this software in an efficient way, while Information Technology uses them later to create Information Systems for storing, processing and transmitting data.
 
 

 
 
CS is a study of using computation and computer systems for solving real-world problems. Dealing mostly with software, the study includes the theory of computation and computer architecture, design, development, and application of software systems. The most common problems are organized in groups in particular areas, such as Distributed Systems, Artificial Intelligence, Data Science, Programming Languages and Compilers, Algorithms and Data Structures, etc. Summarizing, CS mainly focuses on finding answers to the following questions (by John DeNero, cs61a):
 
     
  • which real-world problems can be solved using computation
    •  
  • how to solve these problems
    •  
  • how to solve them efficiently
    •  
 
 

 
 
The fact that CS is all about software, makes it tightly coupled to Electrical Engineering that deals with hardware and focuses on designing computer systems and electronic devices for running software in the most efficient way.
 
 

 
 
Unlike CS, IT is a study of using computers to design, build, and operate Information Systems which are used for storing and processing information (data). ICT extends it by applying telecommunications for receiving and transmitting data. It is crucial to notice, that IT apply existing technologies (e.g. hardware, operating systems, systems software, middleware applications, databases, networks) for creating Information Systems. Hence, IT professionals are users of technologies and utilize existing solutions (hardware and software) to create larger systems for solving a specific business need.
 
 

 
 Turing completeness 
 

 
 Turing completeness  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/csp/index.html b/articles/csp/index.html
index 89136337..72df6d09 100644
--- a/articles/csp/index.html
+++ b/articles/csp/index.html
@@ -1 +1 @@
- Constraint Satisfaction Problem (CSP) :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Constraint Satisfaction Problem (CSP)
 
on 2019-07-22 in note about cs programming ai ~2 min read 
 
A mathematical question that is defined by variables, domains, and constraints
 
 
Constraint Satisfaction Problem (CSP) is a class of problems that can be used to represent a large set of real-world problems. In particular, it is widely used in Artificial Intelligent (AI) as finding a solution for a formulated CSP may be used in decision making. There are a few adjacent areas in terms that for solving problems they all involve constraints: SAT (Boolean satisfiability problem), and the SMT (satisfiability modulo theories).
 
 

 
 
Generally speaking, the complexity of finding a solution for CSP is NP-Complete (takes exponential time), because a solution can be guessed and verified relatively fast (in polynomial time), and the SAT problem (NP-Hard) can be translated into a CSP problem. But, it also means, there is no known polynomial-time solution. Thus, the development of efficient algorithms and techniques for solving CSPs is crucial and appears as a subject in many scientific pieces of research.
 
 

 
 
The simplest kind of CSPs are defined by a set of discrete variables (e.g. X, Y), finite non-empty domains (e.g. 0<X<=4, Y<10), and a set of constraints (e.g. Y=X^2, X<>3) which involve some of the variables. There are distinguished two related terms: the Possible World (or the Complete Assignment) of a CSP is an assignment of values to all variables and the Model (or the Solution to a CSP) is a possible world that satisfies all the constraints.
 
 

 
 
Within the topic, there is a programming paradigm - Constraint Programming. It allows building a Constraint Based System where relations between variables are stated in a form of constraints. Hence, this defines certain specifics:
 
     
  • the paradigm doesn't specify a sequence of steps to execute for finding a solution, but rather declares the solution's properties and desired result. This makes the paradigm a sort of Declarative Programming
    •  
  • it's usually characterized by non-directional computation when to satisfy constraints, computations are propagated throughout a system accordingly to changed conditions or variables' values.
    •  
 
The example of using this paradigm can be seen in another my article "A converter of a character's case and Ascii codes".
 
 

 
 
A CSP can be effectively applied in a number of areas like mappings, assignments, planning and scheduling, games (e.g. sudoku), solving system of equations, etc. There are also a number of software frameworks which provide CSP solvers, like python-constraint and Google OR-Tools, just to name a few.
  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ Constraint Satisfaction Problem (CSP) :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Constraint Satisfaction Problem (CSP)
 
on 2019-07-22 in note about cs programming ai ~2 min read 
 
A mathematical question that is defined by variables, domains, and constraints
 
 
Constraint Satisfaction Problem (CSP) is a class of problems that can be used to represent a large set of real-world problems. In particular, it is widely used in Artificial Intelligent (AI) as finding a solution for a formulated CSP may be used in decision making. There are a few adjacent areas in terms that for solving problems they all involve constraints: SAT (Boolean satisfiability problem), and the SMT (satisfiability modulo theories).
 
 

 
 
Generally speaking, the complexity of finding a solution for CSP is NP-Complete (takes exponential time), because a solution can be guessed and verified relatively fast (in polynomial time), and the SAT problem (NP-Hard) can be translated into a CSP problem. But, it also means, there is no known polynomial-time solution. Thus, the development of efficient algorithms and techniques for solving CSPs is crucial and appears as a subject in many scientific pieces of research.
 
 

 
 
The simplest kind of CSPs are defined by a set of discrete variables (e.g. X, Y), finite non-empty domains (e.g. 0<X<=4, Y<10), and a set of constraints (e.g. Y=X^2, X<>3) which involve some of the variables. There are distinguished two related terms: the Possible World (or the Complete Assignment) of a CSP is an assignment of values to all variables and the Model (or the Solution to a CSP) is a possible world that satisfies all the constraints.
 
 

 
 
Within the topic, there is a programming paradigm - Constraint Programming. It allows building a Constraint Based System where relations between variables are stated in a form of constraints. Hence, this defines certain specifics:
 
     
  • the paradigm doesn't specify a sequence of steps to execute for finding a solution, but rather declares the solution's properties and desired result. This makes the paradigm a sort of Declarative Programming
    •  
  • it's usually characterized by non-directional computation when to satisfy constraints, computations are propagated throughout a system accordingly to changed conditions or variables' values.
    •  
 
The example of using this paradigm can be seen in another my article "A converter of a character's case and Ascii codes".
 
 

 
 
A CSP can be effectively applied in a number of areas like mappings, assignments, planning and scheduling, games (e.g. sudoku), solving system of equations, etc. There are also a number of software frameworks which provide CSP solvers, like python-constraint and Google OR-Tools, just to name a few.
  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/data-structure/index.html b/articles/data-structure/index.html
index b25b4042..2f26dcd9 100644
--- a/articles/data-structure/index.html
+++ b/articles/data-structure/index.html
@@ -1 +1 @@
- Organizing Unstructured Data :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Organizing Unstructured Data
 
on 2019-08-21 in note about cs programming ~5 min read 
 
Managing data complexity using types, structures, ADTs, and objects
 
 
 
Topics
  
 

 
 
The main, if not the only, purpose of a computer is to compute information. It doesn't always have to be a computation of mathematical formulas. In general, it is a transformation of one piece of information into another. Computers only work with information that can be represented as discrete data. The input and output of a computer engine are always natural numbers or text (a sequence of symbols from a dictionary that correspond to certain natural numbers).
 computation diagram 
As long as data is unstructured, it's hard to make some sense of it. But once data is given a structured form, it becomes meaningful and suitable for further transformation.
 
 

 
 
 
 
Type
 
The simplest form of data organization is Type. In general, a data type defines a set of values with certain properties. It usually defines a size in bytes. A primitive data type is an ordered set of bytes. When a variable of a primitive data type has only one value (holds only one piece of information), it's called a scalar and a type - scalar data type. Well-known examples are integer, float, pointer, and char. A collection of primitive (scalar) data types is called an aggregate data type, and it allows multiple values to be stored. This can be a homogeneous collection, where all elements are of the same type, such as an array, a string, or a file. Or it can be heterogeneous, where elements are of different types, such as a structure or a class. The main property is an ordered set of bytes. The internal organization is simple, straightforward, and all actions (e.g. reading or modifying) are performed directly on the data, according to a hardware architecture that defines the byte order in memory (little-/big-endian).
 
 

 
 
 
 
Data Structure
 
The next level of data abstraction is called Data Structure. It brings more complexity, but also more flexibility to make the right choice between access speed, ability to grow, modification speed, etc. Internally, it's represented by a collection of the scalar or aggregate data types. The main focus is on the details of the internal organization and a set of rules to control that organization. There are two types of data structures that result from a difference in the memory allocation of the underlying elements:
 
     
  • Array Data Structures (static), based on physically contiguous elements in memory, with no gaps between them;
    •  
  • Linked Data Structures (dynamic), based on elements, dynamically allocated in memory and linked in a linear structure using pointers (usually, one or two)
    •  
 
Well-known examples are linked list, hash (dictionary), set, list. These data structures are defined only by their physical organization in memory and a set of rules for data modifications that are performed directly. All internal implementation details are open. The actions performed on the data structures (add, remove, update, etc.) and the ways in which they are used can vary.
 
 

 
 
 
 
Abstract Data Type (ADT)
 
A higher level of data abstraction is represented by an Abstract Data Type (ADT), which shifts the focus from "how to store data" to "how to work with data". An ADT represents a logical organization, defined mainly by a list of predefined operations (functions) for manipulating data and controlling its consistency. Internally, data can be stored in any data structure or combination thereof. However, these internals are hidden and should not be directly accessible. All interactions with data are done through an interface (operations exposed to users). Most of ADTs share a common set of primitive operations, such as
 
     
  • create - a constructor of a new instance
    •  
  • destroy - a destructor of an existing instance
    •  
  • add, get - the set-get functions for adding and removing elements of an instance
    •  
  • is_empty, size - useful functions for managing existing data in an instance
    •  
 
The most common examples of ADTs are stack and queue. Both of these ADTs can be implemented using either array or linked data structures, and both have specific rules for adding and removing elements. All of these specifics are abstracted as functions, which in turn, perform appropriate actions on internal data. Dividing an ADT into operations and data structures creates an abstraction barrier that allows you to maintain a solid interface with the flexibility to change internals without side effects on the code using that ADT.
 
 

 
 
 
 
Object
 
A more comprehensive way of abstracting data is represented by Objects. An object can be thought of as a container for a piece of data that has certain properties. Similar to the ADT, this data is not directly accessible (known as encapsulation or isolation), but instead each object has a set of tightly bound methods that can be applied to operate on its data to produce an expected behavior for that object (known as polymorphism). All such methods are really just functions collected under a class. However, they become methods when called to operate on a particular object. Methods can also be inherited from another class, which is called a superclass. Unlike an ADT, an object doesn't represent a particular type of data, but rather a set of attributes, and it behaves as it should when its methods are invoked. Attributes are nothing more than variables of any type (including ADTs). Formally speaking, classes act as specifications of all of the object's attributes and the methods that can be invoked to deal with those attributes.
 
 

 
 
The Object-Oriented Programming (OOP) paradigm uses objects as the central elements of a program design. At program runtime, each object exists as an instance of a class. The class, in turn, plays a dual role: it defines the behavior (through a set of methods) of all objects instantiated from it, and it declares a prototype of data that will carry some state within the object once it's instantiated. As long as the state is isolated (incapsulated) in the objects, access to that state is organized by communication between the objects via message passing. It's usually implemented by calling a method of an object, which is equivalent to "passing" a message to that object.
 
 

 
 
This behavior is completely different from the Structured Programming Paradigm, which instead of maintaining a collection of interacting objects with an an embedded state, relies on dividing of a project's code into a sequence of mostly independent tasks (functions) that operate with an externally (to them) stored state.
 
 

 
 Data Organization  
 
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ Organizing Unstructured Data :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Organizing Unstructured Data
 
on 2019-08-21 in note about cs programming ~5 min read 
 
Managing data complexity using types, structures, ADTs, and objects
 
 
 
Topics
  
 

 
 
The main, if not the only, purpose of a computer is to compute information. It doesn't always have to be a computation of mathematical formulas. In general, it is a transformation of one piece of information into another. Computers only work with information that can be represented as discrete data. The input and output of a computer engine are always natural numbers or text (a sequence of symbols from a dictionary that correspond to certain natural numbers).
 computation diagram 
As long as data is unstructured, it's hard to make some sense of it. But once data is given a structured form, it becomes meaningful and suitable for further transformation.
 
 

 
 
 
 
Type
 
The simplest form of data organization is Type. In general, a data type defines a set of values with certain properties. It usually defines a size in bytes. A primitive data type is an ordered set of bytes. When a variable of a primitive data type has only one value (holds only one piece of information), it's called a scalar and a type - scalar data type. Well-known examples are integer, float, pointer, and char. A collection of primitive (scalar) data types is called an aggregate data type, and it allows multiple values to be stored. This can be a homogeneous collection, where all elements are of the same type, such as an array, a string, or a file. Or it can be heterogeneous, where elements are of different types, such as a structure or a class. The main property is an ordered set of bytes. The internal organization is simple, straightforward, and all actions (e.g. reading or modifying) are performed directly on the data, according to a hardware architecture that defines the byte order in memory (little-/big-endian).
 
 

 
 
 
 
Data Structure
 
The next level of data abstraction is called Data Structure. It brings more complexity, but also more flexibility to make the right choice between access speed, ability to grow, modification speed, etc. Internally, it's represented by a collection of the scalar or aggregate data types. The main focus is on the details of the internal organization and a set of rules to control that organization. There are two types of data structures that result from a difference in the memory allocation of the underlying elements:
 
     
  • Array Data Structures (static), based on physically contiguous elements in memory, with no gaps between them;
    •  
  • Linked Data Structures (dynamic), based on elements, dynamically allocated in memory and linked in a linear structure using pointers (usually, one or two)
    •  
 
Well-known examples are linked list, hash (dictionary), set, list. These data structures are defined only by their physical organization in memory and a set of rules for data modifications that are performed directly. All internal implementation details are open. The actions performed on the data structures (add, remove, update, etc.) and the ways in which they are used can vary.
 
 

 
 
 
 
Abstract Data Type (ADT)
 
A higher level of data abstraction is represented by an Abstract Data Type (ADT), which shifts the focus from "how to store data" to "how to work with data". An ADT represents a logical organization, defined mainly by a list of predefined operations (functions) for manipulating data and controlling its consistency. Internally, data can be stored in any data structure or combination thereof. However, these internals are hidden and should not be directly accessible. All interactions with data are done through an interface (operations exposed to users). Most of ADTs share a common set of primitive operations, such as
 
     
  • create - a constructor of a new instance
    •  
  • destroy - a destructor of an existing instance
    •  
  • add, get - the set-get functions for adding and removing elements of an instance
    •  
  • is_empty, size - useful functions for managing existing data in an instance
    •  
 
The most common examples of ADTs are stack and queue. Both of these ADTs can be implemented using either array or linked data structures, and both have specific rules for adding and removing elements. All of these specifics are abstracted as functions, which in turn, perform appropriate actions on internal data. Dividing an ADT into operations and data structures creates an abstraction barrier that allows you to maintain a solid interface with the flexibility to change internals without side effects on the code using that ADT.
 
 

 
 
 
 
Object
 
A more comprehensive way of abstracting data is represented by Objects. An object can be thought of as a container for a piece of data that has certain properties. Similar to the ADT, this data is not directly accessible (known as encapsulation or isolation), but instead each object has a set of tightly bound methods that can be applied to operate on its data to produce an expected behavior for that object (known as polymorphism). All such methods are really just functions collected under a class. However, they become methods when called to operate on a particular object. Methods can also be inherited from another class, which is called a superclass. Unlike an ADT, an object doesn't represent a particular type of data, but rather a set of attributes, and it behaves as it should when its methods are invoked. Attributes are nothing more than variables of any type (including ADTs). Formally speaking, classes act as specifications of all of the object's attributes and the methods that can be invoked to deal with those attributes.
 
 

 
 
The Object-Oriented Programming (OOP) paradigm uses objects as the central elements of a program design. At program runtime, each object exists as an instance of a class. The class, in turn, plays a dual role: it defines the behavior (through a set of methods) of all objects instantiated from it, and it declares a prototype of data that will carry some state within the object once it's instantiated. As long as the state is isolated (incapsulated) in the objects, access to that state is organized by communication between the objects via message passing. It's usually implemented by calling a method of an object, which is equivalent to "passing" a message to that object.
 
 

 
 
This behavior is completely different from the Structured Programming Paradigm, which instead of maintaining a collection of interacting objects with an an embedded state, relies on dividing of a project's code into a sequence of mostly independent tasks (functions) that operate with an externally (to them) stored state.
 
 

 
 Data Organization  
 
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/dreyfus/index.html b/articles/dreyfus/index.html
index 93d8b3e3..445614b4 100644
--- a/articles/dreyfus/index.html
+++ b/articles/dreyfus/index.html
@@ -1 +1 @@
- Dreyfus model of skill acquisition :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Dreyfus model of skill acquisition
 
on 2019-06-21 in note about ai learning mindmap ~4 min read 
 
The power of the human mind over the reasoning machines
 
 
There is a special term to describe the intelligence of computers that actually mimics human intelligence only to a certain extent. It is called Artificial Intelligence (AI) to emphasize its synthetic nature. Going back to 1980, the U.S. military had been supporting AI research for more than 20 years, and it was time to ask why almost all efforts to build AI systems capable of "providing expert medical advice, exhibiting common sense, and functioning autonomously in critical military situations" had failed.
 
 

 
 
In February 1980, the Dreyfus brothers (Stuart E. and Hubert L.) presented their report on the research conducted at the University of California, Berkeley with the support of the US Air Force Office of Scientific Research - "A Five-Stage Model of the Mental Activities Involved in Directed Skill Acquisition". This model shows how humans acquire new skills through instruction and experience. After the publication, they continued to work on the model and with some modifications and extensions, the results were published in their book "Mind over Machine. The Power of Human Intuition and Expertise in the Era of the Computer" in 1986. As the brothers said in the book, "our intention is more modest, but more fundamental": "what we can reasonably expect from computer intelligence".
 
 

 
 
A process of skill acquisition typically goes through five developmental stages (novice, advanced beginner, competent, proficient, expert). Each stage characterizes a particular behavior and mental response to a situation. A successful transition from one stage to another is associated with the appropriate transformation of four mental functions: recall, recognition, decision, awareness. Understanding the stages of development is essential for any skill training procedure to facilitate the process of acquiring new skills and advancing to the next stage. For human beings, the entire journey from novice to expert usually takes an average of 10 years. Or, more realistically, 10,000 hours of learning and practice. Some research has also shown that there are only up to 5% of experts on the planet, regardless of their field of expertise.
 
 

 
 
The report also pointed out that "as the student becomes more skilled, he depends less on abstract principles and more on concrete experience" and that only "skill in its minimal form is produced by following abstract formal rules". Thus, the higher levels of performance depend on "everyday, concrete, experience in problem solving". By gaining experience, a student is able to start from scratch by following rules of action in context-free situations. Then, after gaining some experience, a student is able to use guidelines for responding to situational aspects. Further practice leads to being able to use maxims (principles) to determine an appropriate action. All of these three transformations correspond to the first three stages and always represent some kind of analytical decision-making process that is needed to "connect his understanding of the general situation with a specific action".
 
 

 
 
The significant change occurs at Stage 4, when the number of "experienced situations is so vast that normally each specific situation immediately dictates an intuitively appropriate action". The key point here is the shift "from analytical thinking to intuitive response". The highest level of expertise with masterful performance occurs only when the expert "no longer needs principles" and is able to "move almost instantaneously into the production of the appropriate perspective and its associated action". This is one of the most important observations: experts perform beyond the rules, and their performance degrades significantly when they are constrained by any kind of formal rules or processes. In 2008 the Pragmatic Bookshelf published an excellent book "Pragmatic Thinking and Learning" by Andy Hunt. It gives a lot of insight into the human brain and how it works, a number of tips on how to learn more and faster, including a detailed review of the Dreyfus model (ch.2). I also created mind maps for the entire book.
 
 

 
 
Understanding the difference between the human mind and how a reasoning machine "thinks" helps to set realistic expectations for the development of artificial intelligence systems. In trying to define what computers should do, it is first necessary to be clear about what computers can do. In this regard, the computer is an analytical engine, so it can apply rules and make logical inferences. Of course, it does this with extreme speed, high accuracy, and remarkable reproducibility, but it still follows a certain logic. It turns out that precisely this crucial difference between a reasoning machine, which is perhaps perpetually limited, and the intuitive expertise of the human mind, which cannot be reduced to rules, seems to be a good starting point for finding the balance between humans and computers.
 
 

 
 MindMap  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ Dreyfus model of skill acquisition :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Dreyfus model of skill acquisition
 
on 2019-06-21 in note about ai learning mindmap ~4 min read 
 
The power of the human mind over the reasoning machines
 
 
There is a special term to describe the intelligence of computers that actually mimics human intelligence only to a certain extent. It is called Artificial Intelligence (AI) to emphasize its synthetic nature. Going back to 1980, the U.S. military had been supporting AI research for more than 20 years, and it was time to ask why almost all efforts to build AI systems capable of "providing expert medical advice, exhibiting common sense, and functioning autonomously in critical military situations" had failed.
 
 

 
 
In February 1980, the Dreyfus brothers (Stuart E. and Hubert L.) presented their report on the research conducted at the University of California, Berkeley with the support of the US Air Force Office of Scientific Research - "A Five-Stage Model of the Mental Activities Involved in Directed Skill Acquisition". This model shows how humans acquire new skills through instruction and experience. After the publication, they continued to work on the model and with some modifications and extensions, the results were published in their book "Mind over Machine. The Power of Human Intuition and Expertise in the Era of the Computer" in 1986. As the brothers said in the book, "our intention is more modest, but more fundamental": "what we can reasonably expect from computer intelligence".
 
 

 
 
A process of skill acquisition typically goes through five developmental stages (novice, advanced beginner, competent, proficient, expert). Each stage characterizes a particular behavior and mental response to a situation. A successful transition from one stage to another is associated with the appropriate transformation of four mental functions: recall, recognition, decision, awareness. Understanding the stages of development is essential for any skill training procedure to facilitate the process of acquiring new skills and advancing to the next stage. For human beings, the entire journey from novice to expert usually takes an average of 10 years. Or, more realistically, 10,000 hours of learning and practice. Some research has also shown that there are only up to 5% of experts on the planet, regardless of their field of expertise.
 
 

 
 
The report also pointed out that "as the student becomes more skilled, he depends less on abstract principles and more on concrete experience" and that only "skill in its minimal form is produced by following abstract formal rules". Thus, the higher levels of performance depend on "everyday, concrete, experience in problem solving". By gaining experience, a student is able to start from scratch by following rules of action in context-free situations. Then, after gaining some experience, a student is able to use guidelines for responding to situational aspects. Further practice leads to being able to use maxims (principles) to determine an appropriate action. All of these three transformations correspond to the first three stages and always represent some kind of analytical decision-making process that is needed to "connect his understanding of the general situation with a specific action".
 
 

 
 
The significant change occurs at Stage 4, when the number of "experienced situations is so vast that normally each specific situation immediately dictates an intuitively appropriate action". The key point here is the shift "from analytical thinking to intuitive response". The highest level of expertise with masterful performance occurs only when the expert "no longer needs principles" and is able to "move almost instantaneously into the production of the appropriate perspective and its associated action". This is one of the most important observations: experts perform beyond the rules, and their performance degrades significantly when they are constrained by any kind of formal rules or processes. In 2008 the Pragmatic Bookshelf published an excellent book "Pragmatic Thinking and Learning" by Andy Hunt. It gives a lot of insight into the human brain and how it works, a number of tips on how to learn more and faster, including a detailed review of the Dreyfus model (ch.2). I also created mind maps for the entire book.
 
 

 
 
Understanding the difference between the human mind and how a reasoning machine "thinks" helps to set realistic expectations for the development of artificial intelligence systems. In trying to define what computers should do, it is first necessary to be clear about what computers can do. In this regard, the computer is an analytical engine, so it can apply rules and make logical inferences. Of course, it does this with extreme speed, high accuracy, and remarkable reproducibility, but it still follows a certain logic. It turns out that precisely this crucial difference between a reasoning machine, which is perhaps perpetually limited, and the intuitive expertise of the human mind, which cannot be reduced to rules, seems to be a good starting point for finding the balance between humans and computers.
 
 

 
 MindMap  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/engineering/index.html b/articles/engineering/index.html
index 699c4391..7c09d712 100644
--- a/articles/engineering/index.html
+++ b/articles/engineering/index.html
@@ -1 +1 @@
- Who is an engineer :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Who is an engineer
 
on 2019-12-19 in note about mindmap ~1 min read 
 
What's the crucial difference between engineers and scientists
 
 
With the coming of the Industrial Age (approx. 1760-1950), an agricultural society transitioned to an economy, based primarily on massive industrial production. It was the time of the rise of specialized educational centers, where people could get deep knowledge in many different fields of science and became either scientists or engineers.
 
 

 
 
Briefly, the main difference between scientists and engineers is that scientists discover, but engineers invent. That is Engineers, using discoveries of scientists, invent systems, devices, processes, which they design, develop, implement, build, manage, maintain, and improve as different stages of the Engineering process. Engineering is a practical application of scientific knowledge, integrated with business and management. In other words, engineers act as a bridge between science and society by doing inventions for the real world and people.
 
 

 
 
In the modern time of the Information Age, the role of an engineer has been extended by non-technical skills, as a result of the globalization and spreading of trade relationships across the globe. These are skills such as intellectual (communication, foreign languages, critical thinking), management (time management, self-organization, planning), and standards awareness (tech certifications, best practices).
 
 

 
 Engineers 
 

 
 Engineering 
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ Who is an engineer :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Who is an engineer
 
on 2019-12-19 in note about mindmap ~1 min read 
 
What's the crucial difference between engineers and scientists
 
 
With the coming of the Industrial Age (approx. 1760-1950), an agricultural society transitioned to an economy, based primarily on massive industrial production. It was the time of the rise of specialized educational centers, where people could get deep knowledge in many different fields of science and became either scientists or engineers.
 
 

 
 
Briefly, the main difference between scientists and engineers is that scientists discover, but engineers invent. That is Engineers, using discoveries of scientists, invent systems, devices, processes, which they design, develop, implement, build, manage, maintain, and improve as different stages of the Engineering process. Engineering is a practical application of scientific knowledge, integrated with business and management. In other words, engineers act as a bridge between science and society by doing inventions for the real world and people.
 
 

 
 
In the modern time of the Information Age, the role of an engineer has been extended by non-technical skills, as a result of the globalization and spreading of trade relationships across the globe. These are skills such as intellectual (communication, foreign languages, critical thinking), management (time management, self-organization, planning), and standards awareness (tech certifications, best practices).
 
 

 
 Engineers 
 

 
 Engineering 
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/goto/index.html b/articles/goto/index.html
index 57ab25a6..0f18c7e3 100644
--- a/articles/goto/index.html
+++ b/articles/goto/index.html
@@ -1 +1 @@
- Structured Programming Paradigm :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Structured Programming Paradigm
 
on 2019-06-12 in note about cs programming mindmap ~1 min read 
 
What can cause too much use of "goto statements"
 
 
There was a time when computer programs were so long and unstructured that only a few people could logically navigate the source code of huge software projects. With low-level programming languages, programmers used various equivalents of "goto" statements for conditional branching, which often resulted in decreased readability and difficulty maintaining logical context, especially when jumping too far into another subroutine.
 
 

 
 
A few things happened on the way to a solution that eventually appeared in the form of the Structured Programming Paradigm. In 1966, Corrado Böhm and Guiseppe Jacopini proved a theorem that any computer program that can be represented as a flowchart can be rewritten using only 3 control structures (sequence, selection, and iteration).
 
 

 
 
In 1968, Edsger W. Dijkstra published the influential article "Go To Statement Considered Harmful", in which he pointed out that using too many goto statements would make computer programs harder to read and understand. However, his intention was unfortunately misunderstood and misused by the almost complete abandonment of the use of "goto" in high-level programming languages, even at the cost of less readable and vague code.
 
 

 
 
As a result of his work on improving ALGOL, Niklaus Wirth designed a new imperative programming language, Pascal, which was released in 1970. It has been widely used for teaching structured programming design to students for several decades since.
 
 

 
 MindMap  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ Structured Programming Paradigm :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Structured Programming Paradigm
 
on 2019-06-12 in note about cs programming mindmap ~1 min read 
 
What can cause too much use of "goto statements"
 
 
There was a time when computer programs were so long and unstructured that only a few people could logically navigate the source code of huge software projects. With low-level programming languages, programmers used various equivalents of "goto" statements for conditional branching, which often resulted in decreased readability and difficulty maintaining logical context, especially when jumping too far into another subroutine.
 
 

 
 
A few things happened on the way to a solution that eventually appeared in the form of the Structured Programming Paradigm. In 1966, Corrado Böhm and Guiseppe Jacopini proved a theorem that any computer program that can be represented as a flowchart can be rewritten using only 3 control structures (sequence, selection, and iteration).
 
 

 
 
In 1968, Edsger W. Dijkstra published the influential article "Go To Statement Considered Harmful", in which he pointed out that using too many goto statements would make computer programs harder to read and understand. However, his intention was unfortunately misunderstood and misused by the almost complete abandonment of the use of "goto" in high-level programming languages, even at the cost of less readable and vague code.
 
 

 
 
As a result of his work on improving ALGOL, Niklaus Wirth designed a new imperative programming language, Pascal, which was released in 1970. It has been widely used for teaching structured programming design to students for several decades since.
 
 

 
 MindMap  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/index.html b/articles/index.html
index 7b4bc0a6..889e7576 100644
--- a/articles/index.html
+++ b/articles/index.html
@@ -1 +1 @@
- Archive :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Archive
 
 
 
  
 2024 February , January  
 
 2020 May , April , January  
 
 2019 December , August , July , June  
  
 
 
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Archive :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Archive
 
 
 
  
 2024 February , January  
 
 2020 May , April , January  
 
 2019 December , August , July , June  
  
 
 
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/learning/index.html b/articles/learning/index.html
index 5cab2ad0..71cd5920 100644
--- a/articles/learning/index.html
+++ b/articles/learning/index.html
@@ -1 +1 @@
- My notes for the "Pragmatic Thinking and Learning" book :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
My notes for the "Pragmatic Thinking and Learning" book
 
on 2020-01-18 in note about learning mindmap ~1 min read 
 
Notes in the form of mindmaps
 
 
 
New Skill Acquisition
 New skill acquisition 
 

 
 
 
 
Pragmatic Learning Plan
 Pragmatic learning plan 
 

 
 
 
 
Dreyfus model
 Dreyfus model 
 

 
 
 
 
Mastering Knowledge
 Mastering knowledge 
 

 
 
 
 
Gaining Experience
 Gaining experience 
 

 
 
 
 
How to start learning
 How to start learning 
 

 
 
  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ My notes for the "Pragmatic Thinking and Learning" book :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
My notes for the "Pragmatic Thinking and Learning" book
 
on 2020-01-18 in note about learning mindmap ~1 min read 
 
Notes in the form of mindmaps
 
 
 
New Skill Acquisition
 New skill acquisition 
 

 
 
 
 
Pragmatic Learning Plan
 Pragmatic learning plan 
 

 
 
 
 
Dreyfus model
 Dreyfus model 
 

 
 
 
 
Mastering Knowledge
 Mastering knowledge 
 

 
 
 
 
Gaining Experience
 Gaining experience 
 

 
 
 
 
How to start learning
 How to start learning 
 

 
 
  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/maslow/index.html b/articles/maslow/index.html
index ad18c1a1..641b7343 100644
--- a/articles/maslow/index.html
+++ b/articles/maslow/index.html
@@ -1 +1 @@
- Maslow's hierarchy of needs :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Maslow's hierarchy of needs
 
on 2019-06-15 in note about psychology mindmap ~1 min read 
 
This theory describes the stages of growth in humans
 
 
Maslow's hierarchy of needs is a theory that shows a path of psychological development in humans through several hierarchical stages. Each stage characterizes a common set of motivations and needs. The important observation is that a certain motivation belongs to an appropriate stage and in order to occur, all previous stages must be well satisfied. The main goal of the Maslow's theory is to show all prerequisites which a human needs to meet for reaching the top level of psychological development - "self-actualization" - that can be achieved only when all basic and mental needs are fulfilled.
 
 

 
 
Self-actualization is the need that is perceived very specifically and is described as the desire to accomplish everything that one can, to become the most that one can be, to realize one's full potential. Maslow used this term to describe namely a desire, not a driving force. In his later theory, Maslow also noticed that the fullest realization can be found in giving oneself to something beyond oneself.
 
 

 
 
For me personally, this theory played an eye-opening role in understanding what was that implicit force which has been pushing me for many years to learn, to immigrate, and finally realize where I'm heading to.
 
 

 
 MindMap  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ Maslow's hierarchy of needs :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Maslow's hierarchy of needs
 
on 2019-06-15 in note about psychology mindmap ~1 min read 
 
This theory describes the stages of growth in humans
 
 
Maslow's hierarchy of needs is a theory that shows a path of psychological development in humans through several hierarchical stages. Each stage characterizes a common set of motivations and needs. The important observation is that a certain motivation belongs to an appropriate stage and in order to occur, all previous stages must be well satisfied. The main goal of the Maslow's theory is to show all prerequisites which a human needs to meet for reaching the top level of psychological development - "self-actualization" - that can be achieved only when all basic and mental needs are fulfilled.
 
 

 
 
Self-actualization is the need that is perceived very specifically and is described as the desire to accomplish everything that one can, to become the most that one can be, to realize one's full potential. Maslow used this term to describe namely a desire, not a driving force. In his later theory, Maslow also noticed that the fullest realization can be found in giving oneself to something beyond oneself.
 
 

 
 
For me personally, this theory played an eye-opening role in understanding what was that implicit force which has been pushing me for many years to learn, to immigrate, and finally realize where I'm heading to.
 
 

 
 MindMap  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/numbers/index.html b/articles/numbers/index.html
index 325b5f2a..2b885543 100644
--- a/articles/numbers/index.html
+++ b/articles/numbers/index.html
@@ -1 +1 @@
- Number Classification :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Number Classification
 
on 2019-08-16 in note about math cs ~2 min read 
 
All number categories, from Complex to Counting
 
 
Mathematics is unique. The unique science if everyone could agree that it is Science. But, it's also hard to argue that it is not Art. Math is absolutely certain, except the cases when it is not ("as far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality"). Still having no one general definition, math doesn't even bother to have one opinion on such the fundamental building block as Numbers. Nevertheless, math is an important part of almost every field of science, engineering, and human life.
 
 

 
 
Here is the most common and well-accepted number classification tree:
 Number classification 
 

 
 
It also shouldn't be a surprise to find slight distinctions in the meaning of the same essences in Math and Computer Science (CS):
 
     
  • Natural numbers. In Math, they are meant to be Positive Integers (1, 2, 3, ...), but in CS they are non-negative Integers which include Zero (0, 1, 2, 3 ...)
    •  
  • Mantissa. In Math, it is a fractional part of the logarithm. In CS, it is significant digits of a floating-point number (thus, quite often are used other definitions in this case, like significand and coefficient)
    •  
 
 

 
 
There is a quite related topic in terms of the values which a variable can take on. In mathematics, a variable may be two different types: continuous and discrete:
 
     
  • A variable is continuous when it can take on infinitely many, uncountable values. There is always another value in between two others in a non-empty range, no matter how close they are.
    •  
  • A variable is discrete when there is always a positive minimum distance between two values in a non-empty range. The set of numbers is finite or countably infinite (e.g. Natural numbers)
    •  
 
The understanding of the discreteness is crucial in Computer Science as all real-world computers internally work only with discrete data (which makes it challenging to represent Irrational numbers). All existing computability theories (e.g. Turing thesis, Church thesis) are defined on discrete values, and the domain is the set of Natural numbers.
  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ Number Classification :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Number Classification
 
on 2019-08-16 in note about math cs ~2 min read 
 
All number categories, from Complex to Counting
 
 
Mathematics is unique. The unique science if everyone could agree that it is Science. But, it's also hard to argue that it is not Art. Math is absolutely certain, except the cases when it is not ("as far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality"). Still having no one general definition, math doesn't even bother to have one opinion on such the fundamental building block as Numbers. Nevertheless, math is an important part of almost every field of science, engineering, and human life.
 
 

 
 
Here is the most common and well-accepted number classification tree:
 Number classification 
 

 
 
It also shouldn't be a surprise to find slight distinctions in the meaning of the same essences in Math and Computer Science (CS):
 
     
  • Natural numbers. In Math, they are meant to be Positive Integers (1, 2, 3, ...), but in CS they are non-negative Integers which include Zero (0, 1, 2, 3 ...)
    •  
  • Mantissa. In Math, it is a fractional part of the logarithm. In CS, it is significant digits of a floating-point number (thus, quite often are used other definitions in this case, like significand and coefficient)
    •  
 
 

 
 
There is a quite related topic in terms of the values which a variable can take on. In mathematics, a variable may be two different types: continuous and discrete:
 
     
  • A variable is continuous when it can take on infinitely many, uncountable values. There is always another value in between two others in a non-empty range, no matter how close they are.
    •  
  • A variable is discrete when there is always a positive minimum distance between two values in a non-empty range. The set of numbers is finite or countably infinite (e.g. Natural numbers)
    •  
 
The understanding of the discreteness is crucial in Computer Science as all real-world computers internally work only with discrete data (which makes it challenging to represent Irrational numbers). All existing computability theories (e.g. Turing thesis, Church thesis) are defined on discrete values, and the domain is the set of Natural numbers.
  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/py-params/index.html b/articles/py-params/index.html
index fb4ec26a..227954d5 100644
--- a/articles/py-params/index.html
+++ b/articles/py-params/index.html
@@ -1,4 +1,4 @@
- A little mess with function parameters in Python :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
A little mess with function parameters in Python
 
on 2024-02-17 in tutorial about python programming ~8 min read 
 
A variety of ways to define function parameters
 
 
TLDR: quick summary of the article
 
 

 
 
At first glance, Python functions look like those in most other languages, and they behave just as you'd expect. They take arguments, have default values, and can also return a value. This is intentional, of course. But once you dive deeper, you'll see how many specific nuances are hidden internally, providing a programmer with a number of features that make using functions in Python a much more powerful experience. Knowing the differences is critical to understanding why they behave the way they do, so you can get the most out of them.
 
 

 
 
One of the key feature is that functions in Python are objects that are created as soon as they are defined. This allows you to use functions as arguments in other functions or as return values, just like any other Python object. Functions lifetime is different from the execution time, and they exist even after execution has finished. Functions, being objects, also have a set of predefined attributes that can be extended at any time, and their state is maintained outside of the execution. Parameters become local variables, which are completely different entities from function attributes, which exist only at execution time. Default values in the function definition can also be expressions, but they are evaluated only once. Function arguments are always passed by value, but the values they contain are references. This is why they're sometimes called pass-by-object-references. This also means that parameters, like any other variable in Python, are untyped, and contain a copy of a reference to an object. Changing a parameter (a local variable) generally doesn't change an object (passed as an argument) itself, but only stores a reference to another object. However, there is still a way to change an object that is passed as an argument, if it is a mutable object and the change is made directly to it rather than to a variable. For example, updating elements of a list or a dictionary.
 
 

 
 
This tutorial will focus only on parameters, their different types, and various ways to define them. Let's start with the most common: a function definition with 4 parameters (a, b, c, d). No types, just names, with a lifetime during function execution, i.e. they are created on the stack as local variables only during function execution. When the function is called, it gets 4 arguments (w, x, y, z), which are also local variables (live on a stack), but in the calling environment, and contain references to some objects. Python takes these references stored in the arguments (w, x, y, z) and copies them into parameters (a, b, c, d) that live as local variables on a stack in the called environment:
 
 

 
 
def myfunc(a, b, c, d):
+ A little mess with function parameters in Python :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
A little mess with function parameters in Python
 
on 2024-02-17 in tutorial about python programming ~8 min read 
 
A variety of ways to define function parameters
 
 
TLDR: quick summary of the article
 
 

 
 
At first glance, Python functions look like those in most other languages, and they behave just as you'd expect. They take arguments, have default values, and can also return a value. This is intentional, of course. But once you dive deeper, you'll see how many specific nuances are hidden internally, providing a programmer with a number of features that make using functions in Python a much more powerful experience. Knowing the differences is critical to understanding why they behave the way they do, so you can get the most out of them.
 
 

 
 
One of the key feature is that functions in Python are objects that are created as soon as they are defined. This allows you to use functions as arguments in other functions or as return values, just like any other Python object. Functions lifetime is different from the execution time, and they exist even after execution has finished. Functions, being objects, also have a set of predefined attributes that can be extended at any time, and their state is maintained outside of the execution. Parameters become local variables, which are completely different entities from function attributes, which exist only at execution time. Default values in the function definition can also be expressions, but they are evaluated only once. Function arguments are always passed by value, but the values they contain are references. This is why they're sometimes called pass-by-object-references. This also means that parameters, like any other variable in Python, are untyped, and contain a copy of a reference to an object. Changing a parameter (a local variable) generally doesn't change an object (passed as an argument) itself, but only stores a reference to another object. However, there is still a way to change an object that is passed as an argument, if it is a mutable object and the change is made directly to it rather than to a variable. For example, updating elements of a list or a dictionary.
 
 

 
 
This tutorial will focus only on parameters, their different types, and various ways to define them. Let's start with the most common: a function definition with 4 parameters (a, b, c, d). No types, just names, with a lifetime during function execution, i.e. they are created on the stack as local variables only during function execution. When the function is called, it gets 4 arguments (w, x, y, z), which are also local variables (live on a stack), but in the calling environment, and contain references to some objects. Python takes these references stored in the arguments (w, x, y, z) and copies them into parameters (a, b, c, d) that live as local variables on a stack in the called environment:
 
 

 
 
def myfunc(a, b, c, d):
     print(a, b, c, d)
 
 def caller():
diff --git a/articles/remove-webpage-google/index.html b/articles/remove-webpage-google/index.html
index 088ec88f..6d823c34 100644
--- a/articles/remove-webpage-google/index.html
+++ b/articles/remove-webpage-google/index.html
@@ -1,4 +1,4 @@
- How to remove a webpage from the Google index :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
How to remove a webpage from the Google index
 
on 2019-07-18 in howto about it web html ~3 min read 
 
The approach for removing outdated or deleted content from Google search results
 
 
It's important to keep in mind that search engines scan websites on periodic bases and these periods may vary depending on a number of factors. In general, websites' owners don't have full control over the behavior of search engines, but instead, they can define preferences in a form of instructions. Such instructions, for example, allow excluding certain web pages from showing up in search results or preventing search engines from digging into specific paths. There are two ways to declare preferences: tweaking parameters of robots.txt in the root of a website and HTML <meta> tag "robots" in the <head> block of web pages.
 
 

 
 
I've recently needed to move one of my static websites to another domain. It became a complex task as I'm not able to change a server-side configuration, and the redirection of HTTP-requests is only one part of the story. Once all users are being redirected to a new location, I had to initiate and speed up a process of cleaning up the search results from links to my old website.
 
 

 
 
There are basically a few common ways to remove web pages from search indexes:
 
     
  • remove a page completely, so clients will be getting 404 Not Found HTTP response. It is clearly not my case, as the old website responses with valid and existing web pages
    •  
  • restrict access to a page by asking clients to enter credentials. Then, the server will be sending 401 Unauthorized HTTP response. This also won't work for me, as requires changing the configuration on the server-side
    •  
  • add an HTML <meta> tag robots with the value noindex. That's exactly what I needed and can be implemented on the client-side.
    •  
 
The last method allows setting different preferences per page right from the HTML code. That is, search engines must have access to a page to read it and find this instruction. This also means that all web pages with robots meta tag shouldn't be blocked even by a robots.txt file!
 
 

 
 
This solution will show a few steps for removing an entire website from Google's search results.
 
     
  • check robots.txt (if it exists) and be sure that search bots are allowed to go through the site and read all indexed web pages. The file should either be empty or something like this (allows any bots read any webpage on a site):
    •  
 
User-agent: *
+ How to remove a webpage from the Google index :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
How to remove a webpage from the Google index
 
on 2019-07-18 in howto about it web html ~3 min read 
 
The approach for removing outdated or deleted content from Google search results
 
 
It's important to keep in mind that search engines scan websites on periodic bases and these periods may vary depending on a number of factors. In general, websites' owners don't have full control over the behavior of search engines, but instead, they can define preferences in a form of instructions. Such instructions, for example, allow excluding certain web pages from showing up in search results or preventing search engines from digging into specific paths. There are two ways to declare preferences: tweaking parameters of robots.txt in the root of a website and HTML <meta> tag "robots" in the <head> block of web pages.
 
 

 
 
I've recently needed to move one of my static websites to another domain. It became a complex task as I'm not able to change a server-side configuration, and the redirection of HTTP-requests is only one part of the story. Once all users are being redirected to a new location, I had to initiate and speed up a process of cleaning up the search results from links to my old website.
 
 

 
 
There are basically a few common ways to remove web pages from search indexes:
 
     
  • remove a page completely, so clients will be getting 404 Not Found HTTP response. It is clearly not my case, as the old website responses with valid and existing web pages
    •  
  • restrict access to a page by asking clients to enter credentials. Then, the server will be sending 401 Unauthorized HTTP response. This also won't work for me, as requires changing the configuration on the server-side
    •  
  • add an HTML <meta> tag robots with the value noindex. That's exactly what I needed and can be implemented on the client-side.
    •  
 
The last method allows setting different preferences per page right from the HTML code. That is, search engines must have access to a page to read it and find this instruction. This also means that all web pages with robots meta tag shouldn't be blocked even by a robots.txt file!
 
 

 
 
This solution will show a few steps for removing an entire website from Google's search results.
 
     
  • check robots.txt (if it exists) and be sure that search bots are allowed to go through the site and read all indexed web pages. The file should either be empty or something like this (allows any bots read any webpage on a site):
    •  
 
User-agent: *
 Disallow:
 
 
     
  • add robots HTML <meta> tag in the <head> block with "noindex, nofollow" value in each indexed web page:
    •  
 
<meta name="robots" content="noindex, nofollow" />
 
 
     
  • create a sitemap.xml file and define all indexed web pages with the <lastmod> section which points to some recent time. For example:
    •  
 
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
diff --git a/articles/smart/index.html b/articles/smart/index.html
index 0f1dc4cf..f270cfa6 100644
--- a/articles/smart/index.html
+++ b/articles/smart/index.html
@@ -1 +1 @@
- Managing your plans in the S.M.A.R.T. way :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Managing your plans in the S.M.A.R.T. way
 
on 2019-06-26 in note about management mindmap ~3 min read 
 
Reach your goals by setting SMART objectives within the action plan
 
 
The benefit of using SMART criteria in planning has been known for a few last decades. They were so widely applied, integrated as a proven technique in managing organizations' goals and objectives, and improved by so many contributors that currently is quite hard to name the only one certain definition of 'SMART'. For different people, this acronym means different things. Even the terms 'goal' and 'objective', in some cases, have either opposite or the same meaning. Nevertheless, the knowledge of the original ideas helps to get the most from the whole approach.
 
 

 
 
In 1981, the article of George T. Doran "There's a S.M.A.R.T. way to write management's goals and objectives" was published. He reasonably pointed out that despite all the available literature and seminars, "most managers still don't know what objectives are and how they can be written", that "the majority of U.S. corporations don't really have an effective objective setting/planning process", and that "the process of writing objectives is a major source of anxiety that many individuals would like to live without".
 
 

 
 
It was a frustrating reality caused by the lack of proper education across "corporate officers, managers, and supervisors", on all levels. George T. Doran has also mentioned, "objective setting must become a way of life" and suggested an effective way to tackle this problem. When it comes to define goals or write objectives, one has "to think of the acronym SMART" and be clear about a distinction between terms 'goal' and 'objective'.
 
 

 
 
"Goals represent unique executive beliefs and philosophies. They are usually of a form that is continuous and long-term". "Objectives , on the other hand, give quantitative support and expression to managements' beliefs", "enable an organization to focus on problems, and give the company a sense of direction", "a statement of results to be achieved".
 
 

 
 
The acronym itself, in the original form, had the following meaning:
 
     
  • Specific, target a specific area for improvement
    •  
  • Measurable, quantify or at least suggest an indicator of progress
    •  
  • Assignable, specify who will do it
    •  
  • Realistic, state what results can realistically be achieved, given available resources
    •  
  • Time-Related, specify when the result(s) can be achieved
    •  
 
Taking into consideration the fact, that the proposed solution was targeted to organizations, it will not be a surprise that few "letters" changed their meaning to satisfy the needs of smaller groups or individuals. For instance, for personal use, as there are no others to whom it could be assigned, "A" and "R" change their meaning to Achievable (similar to Realistic) and Relevant (appropriate and related to the context).
 
 

 
 
There are two important observations to notice:
 
     
  • the proposed technique doesn't require all objectives to be measurable or quantified. In some situations, it can lead to "lose the benefit of a more abstract objective in order to gain quantification"
    •  
  • it is not required to have all five criteria set. "However, the closer we get to the SMART criteria as a guideline, the smarter our objectives will be".
    •  
 
One way to make goals and objectives working together is to join them in one Action Plan. The action plan makes it possible to reach long-term goals by using short-run objectives in the context of your real situation.
 
 

 
 MindMap  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ Managing your plans in the S.M.A.R.T. way :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Managing your plans in the S.M.A.R.T. way
 
on 2019-06-26 in note about management mindmap ~3 min read 
 
Reach your goals by setting SMART objectives within the action plan
 
 
The benefit of using SMART criteria in planning has been known for a few last decades. They were so widely applied, integrated as a proven technique in managing organizations' goals and objectives, and improved by so many contributors that currently is quite hard to name the only one certain definition of 'SMART'. For different people, this acronym means different things. Even the terms 'goal' and 'objective', in some cases, have either opposite or the same meaning. Nevertheless, the knowledge of the original ideas helps to get the most from the whole approach.
 
 

 
 
In 1981, the article of George T. Doran "There's a S.M.A.R.T. way to write management's goals and objectives" was published. He reasonably pointed out that despite all the available literature and seminars, "most managers still don't know what objectives are and how they can be written", that "the majority of U.S. corporations don't really have an effective objective setting/planning process", and that "the process of writing objectives is a major source of anxiety that many individuals would like to live without".
 
 

 
 
It was a frustrating reality caused by the lack of proper education across "corporate officers, managers, and supervisors", on all levels. George T. Doran has also mentioned, "objective setting must become a way of life" and suggested an effective way to tackle this problem. When it comes to define goals or write objectives, one has "to think of the acronym SMART" and be clear about a distinction between terms 'goal' and 'objective'.
 
 

 
 
"Goals represent unique executive beliefs and philosophies. They are usually of a form that is continuous and long-term". "Objectives , on the other hand, give quantitative support and expression to managements' beliefs", "enable an organization to focus on problems, and give the company a sense of direction", "a statement of results to be achieved".
 
 

 
 
The acronym itself, in the original form, had the following meaning:
 
     
  • Specific, target a specific area for improvement
    •  
  • Measurable, quantify or at least suggest an indicator of progress
    •  
  • Assignable, specify who will do it
    •  
  • Realistic, state what results can realistically be achieved, given available resources
    •  
  • Time-Related, specify when the result(s) can be achieved
    •  
 
Taking into consideration the fact, that the proposed solution was targeted to organizations, it will not be a surprise that few "letters" changed their meaning to satisfy the needs of smaller groups or individuals. For instance, for personal use, as there are no others to whom it could be assigned, "A" and "R" change their meaning to Achievable (similar to Realistic) and Relevant (appropriate and related to the context).
 
 

 
 
There are two important observations to notice:
 
     
  • the proposed technique doesn't require all objectives to be measurable or quantified. In some situations, it can lead to "lose the benefit of a more abstract objective in order to gain quantification"
    •  
  • it is not required to have all five criteria set. "However, the closer we get to the SMART criteria as a guideline, the smarter our objectives will be".
    •  
 
One way to make goals and objectives working together is to join them in one Action Plan. The action plan makes it possible to reach long-term goals by using short-run objectives in the context of your real situation.
 
 

 
 MindMap  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/sq3r/index.html b/articles/sq3r/index.html
index 84d0a53a..e2081869 100644
--- a/articles/sq3r/index.html
+++ b/articles/sq3r/index.html
@@ -1 +1 @@
- SQ3R :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
SQ3R
 
on 2019-06-10 in note about learning mindmap ~1 min read 
 
A reading comprehension method
 
 
The SQ3R is a reading comprehension method of reading textbook materials named for its five steps: Survey, Question, Read, Recite, and Review. It provides an active and fairly efficient approach to acquiring information stored as text. A text form is far from the ideal or best source of meaningful data for our minds, but the SQ3R method makes it possible to remember, understand, and integrate new knowledge with what we already know.
 
 

 
 MindMap  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ SQ3R :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
SQ3R
 
on 2019-06-10 in note about learning mindmap ~1 min read 
 
A reading comprehension method
 
 
The SQ3R is a reading comprehension method of reading textbook materials named for its five steps: Survey, Question, Read, Recite, and Review. It provides an active and fairly efficient approach to acquiring information stored as text. A text form is far from the ideal or best source of meaningful data for our minds, but the SQ3R method makes it possible to remember, understand, and integrate new knowledge with what we already know.
 
 

 
 MindMap  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/stream-encoding/index.html b/articles/stream-encoding/index.html
index 95af95d2..11766500 100644
--- a/articles/stream-encoding/index.html
+++ b/articles/stream-encoding/index.html
@@ -1,4 +1,4 @@
- The zoo of binary-to-text encoding schemes :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
The zoo of binary-to-text encoding schemes
 
on 2020-05-13 in article about cs programming binary-to-text encoding ~15 min read 
 
A stream encoding algorithm with a variable base (16, 32, 36, 64, 58, 85, 94)
 
 
In the previous article, I discussed the use of the positional numeral system for the purpose of binary-to-text translation. That method represents a binary file as a single big number with the radix 256 and then converts this big number to another one with an arbitrary radix (base) in a range from 2 to 94. Although this approach gives the minimum possible size overhead, unfortunately, it also has a number of downsides which make it hardly usable in a real-world situation. In this article, I'll show what is used in practice, which encodings could be found in the wild, and how to build your own encoder.
 
 

 
 
 
What's wrong with the positional single number encoding?
 
The main issue with converting a file as a big number in radix 256 to another big number with a smaller radix is that you need to read the whole file, load it to the memory and build actually that big number from each byte of the file. To construct a number, the Least Significant Byte (LSB), which is the last byte of a file, needs to be read and loaded. Although, there is not always enough memory to load a whole file as well as there is not always the whole file is available at any given time. For instance, if it's being transmitted over a network and only a small amount of bytes from the beginning (from the Most Significant Byte, MSB) has been loaded. This issue is usually addressed by processing a file as a stream of bytes, in chunks, which then are being converted in the same way (by converting a number from one base to another). These chunks are much smaller and, ideally, fit the CPU registers' size (up to 8 bytes). The only question here is how to find the best size and ratio of such chunks (input and output) to keep the size overhead as closely as possible to a minimum available by treating files as big numbers.
 
 

 
 
 
 
What's the essence of a positional numeral system?
 
In the positional numeral systems, everything turns around a radix (base) which shows how many different symbols are used to represent values. The actual glyph doesn't matter. Only their quantity. All these symbols are grouped in an alphabet (a table) where every symbol is defined by its own position, and this position represents its value. As long as counting starts from 0, the maximum symbol's value, in any numeral system, is always radix - 1. For instance, in the numeral system with a radix 10 (Decimal), the maximum value has a symbol '9'. But, for a system with a radix 2 (Binary), the maximum value has a symbol '1'. When symbols from an alphabet appear as a part of a number, they are called digits. A digit's position, in this case, is called index and defines the power of a radix while its value (position in the alphabet) defines a coefficient within the power of that radix.
 
 

 
 
The first crucial conclusion here is that any number, represented in some positional numeral system, gets its meaning only when is known its radix.
 
 

 
 
The second conclusion is not so obvious. Humans in most cases nowadays use the Decimal numeral system. Numbers gain more sense for them when they are represented as Decimal numbers and this is the system that is used the most for calculations. To any symbol in an alphabet is assigned its certain position which is a number with some radix. In most cases, this radix is 10 (Decimal). The Decimal numeral system is a temporary system that is used for converting one numeral system to another. Every time, when a number is defined by a radix, this radix is Decimal, no matter what's the radix of a number. Every time, when there is a need to convert a number X with radix M to a number Y with radix M, both numbers (X and Y) are represented by some certain alphabets (which define symbols with values), but their radixes (M and N) are always represented in Decimal system, thus, Decimal system is used as an intermediate numeral system to which a number X is converted first, and then the intermediate number is converted to a number Y. The intermediate numeral system could have been any radix, but radix 10 is what people use for calculations and that's what can be found in most converters implementations.
 
 

 
 
The third conclusion is even more important. Symbols don't bring any value, only their position in the alphabet. This means we need to know not only an actual number's representation but also its radix and an alphabet - the table that contains symbols assigned to values (position within the table). A good example is an alphabet of 16 symbols for Hexadecimal numbers (radix 16). There are first 10 digits linked to equivalent values, so the symbol '0' is linked to 0, '1' to 1, and so on up to the symbol '9' linked to 9. The rest 6 values (from 10 to 15) linked to English letter symbols (from 'A' to 'F'). And again, these values (positions in the table) are all Decimal numbers (radix 10). By the way, the table could have been different, but that's what is used by convention, so anyone is able to interpret Hexadecimal numbers in the same way.
 
 

 
 
 
 
Where does the overhead come from?
 
Let's take a look at a few examples. This is a number '123' that is represented by three symbols, but until we know a radix, it is not possible to understand its value. If the radix is 10 then it is 'one hundred twenty three' in the Decimal system and it can be calculated by the formula for converting a numeral system with any radix to radix 10 (because all numbers in this formula have radix 10): 1*10^2 + 2*10^1 + 3*10^0 = 123. If the radix is 8, then it is an Octal system and it is constructed as 1*8^2 + 2*8^1 + 3*8^0 which gives us a Decimal number 83. So, '123 base 8' equals to '83 base 10'. It is worth noticing that converting a number to a higher radix leads to lower a number of symbols needed for its representation. The converse is also true. If a number 83 with a radix 10 is converted to a radix 2, it gets a form '1010011'. Notice, the radix is changed from 10 to 2 and the number of symbols changed from 2 to 7! As lower a radix gets, as more symbols appear in representation.
 
 

 
 
Let's get back to binary files. What we can determine as 'symbol representation' or 'digits', 'alphabet', and 'radix' based on a structure of an ordinary file? Any file consists of bytes as it is the minimum addressable group of bits. It cannot be less than 8 bits. So, we can think about a number representation as of some amount of bytes. The chunks can vary from 1 byte to a file's size. For example, if there is only one byte, then the number consists of only one digit. One byte or 8 bits (binary digits with a radix 2) allows one to represent 2^8 = 256 different numbers. That means, we can persist 256 different symbols with their positions to build an alphabet. The good news, such a table has already been standardized many years ago and called ASCII. And the last thing, as the alphabet size is 256 symbols then a radix is also 256. Here is our number: a number of bytes in the chunk that we are going to process are the number of digits, a radix is 256, and the coefficient has a range from 0 to 255. For example, if a group of bytes to read from a stream and process at once consists of 4 bytes (from MSB to LSB): [13, 200, 3, 65] then our number can be represented as a Decimal number (radix 10) as 13*256^3 + 200*256^2 + 3*256^1 + 65*256^0 = 231211841
 
 

 
 
As it was discussed in the previous article, we can use no more than 94 different symbols to reliably represent texts. Thus, the desired radix lies somewhere in the range from 2 to 94. Even 94 is much less than 256, so a number's representation in a new radix is likely to have more symbols. This means, in turn, that the output group will have more bytes as it is a minimum amount of data we can operate on, even if a digit represented by a symbol needs fewer bits. You'll still need to allocate the whole byte for each symbol in the new radix number representation. Some amount of bits in such bytes will never be used. This is the root of inefficiency, and that's why it's highly important to find a good ratio of output to input byte groups. For instance, the most used nowadays Base64 encoding converts binary files to texts by reading 3-bytes groups from the input stream, represents them as a 3-digits number with a radix 256 (log[256^3, 2] = 24 bit), and then converts this number to a 4-digits number with a radix 64 (log[64^4, 2] = 24 bit), which in turn is written to the output stream as a group of 4 bytes. So, the ratio of output to input is 4/3 = 1.333333. In other words, the size overhead is 33.(3)%. There are a few considerations behind the logic of choosing the exact combination of input and output groups for a streaming conversion, which includes a target radix, a desirable/available alphabet, an ability to natively compute on a CPU, etc.
 
 

 
 
 
 
How to calculate a minimal overhead?
 
Let's calculate first, how many digits of a target base (radix) are needed to represent exactly the same number in the initial base. For instance, there is given a number 123 with a radix 10. How many bits (binary digits, a radix 2) are needed to represent the same decimal number? Every digit is a coefficient of power of a base. If it is not enough, one more base is added in power +1 to finally construct a number. Keeping in mind that counting starts from 0, if it's said that to represent some number 8 bit are needed, this means all bases in powers from 0 to 7 with their coefficients have to be summed up. Thus, to find out a number of digits needed to represent the number in some radix, we need to find an exponent, to which a new radix needs to be exponentiated. In our case, for a base-10 number 123, we need to calculate an exponent of a base-2 by using a logarithm function: log[123, 2] = 6.9425145. This means, to represent a number 123 with base 10, a little bit less than 7 bits will be enough. All computer systems operate on a set of natural numbers only. It is not possible to use 6.9425145 bits as this number is an approximated value of needed bits. 6 bits apparently won't be enough (2^6 = 64, which is much less than 123), so the only right approach is always to round up (by calling a ceil function) any non-integer values. Unfortunately, 7 bits are able to represent a bigger number (2^7 = 128) and this again contributes to a final overhead.
 
 

 
 
Let's have a look at the Base64 again. We know already (but not why is that, yet), that this streaming system uses 3 input bytes (a 3-digit number with a base 256) and converts them to a number with a base 64. How many base-64 digits will this number contain? The answer is log[256^3, 64] = 4, four digits, hence 4 symbols from the base64 alphabet.
 
 

 
 
While looking for the good input and output group sizes it's good to know a theoretically possible minimum of the overhead. To find it out, we need to do a similar calculation but take the minimally possible amount of input data, which is one byte (8 bits, decimal 2^8 = 256). For the Base64, it is log[256, 64] = 1.33(3), that is again 33.(3)%. For the Base32 it is log[256, 32] =  1.6, that is 60%. And for the Base16 it is log[256, 16] = 2, that is 100%. Wow! These theoretical numbers are exactly the same as practically used ratios of output bytes to input bytes give. Here are they: for the Base64 it is 4 / 3 = 1.33(3), for the Base32 it is 8 / 5 = 1.6, and for the Base16 it is 2 / 1 = 2. There is one interesting fact, all these three bases (16, 32, 64) have one thing in common - they all are powers of two! This leads us to the conclusion that converting numbers within the "power of two" bases allows one to get the best possible ratio and match precisely an input bits group to an output bits group. Although it is not always desirable or even possible. Sometimes there is a need to use a specific alphabet, e.g. in Base36, or the minimal overhead, e.g. in Base85 or Base94. All these bases are not the "powers of two", so a tradeoff has to be found to minimize the overhead.
 
 

 
 
 
 
How to calculate optimal input and output groups?
 
Alright, we've calculated a number of digits needed to represent some number in another base. But, why is that only a theoretical minimum? Why in practice it would need more? And, why would we still need to find a good ratio of output to input byte groups? To answer these questions, let's have a look at the Base85 encoding. To represent 1 byte (Base256) of information in Base85, it needs log[256, 85] = 1.24816852 digits. But, we can't use 1.248 digits. Only positive whole numbers are available! 1 digit is neither possible (too little). Then, 2 digits are the only way to go. In other words, to represent 1 byte (with a number in Base256), in fact, we'd need 2 bytes (with a number in Base85), where ~75% of space will be wasted, as the ratio is 2/1 = 2 and this is a 100% overhead, instead of a theoretical 24.8%. There is no point to use 1-byte input group and 2-bytes output group. Thus, there should be some good input and output groups so their ratio goes as close as possible to a calculated minimum or even match it!
 
 

 
 
The following approach starts from 1-byte group and using the same formula, every time checks a number of digits in the destination base. if it's not close enough, increments the input group by 1 byte and checks again. You can decide on your own, what is the applicable size of an input group and how close to the whole number up (ceil function) the output group needs to be.
 
 

 
 
This code goes through all bases, from 2 to 94, and prints a first found input/output group that has a delta between the number of digits and its rounded value less or equal 0.1, if any. That is, ceil(x) - x <=0.1. I limited an input group by 20 bytes but in reality, groups larger than 8 bytes (64bit) will require either a more complicated implementation still based on 64bit variable types or the big number mathematics which would bring it back to the solution from the previous article.
 
 

 
 
from math import log, ceil
+ The zoo of binary-to-text encoding schemes :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
The zoo of binary-to-text encoding schemes
 
on 2020-05-13 in article about cs programming binary-to-text encoding ~15 min read 
 
A stream encoding algorithm with a variable base (16, 32, 36, 64, 58, 85, 94)
 
 
In the previous article, I discussed the use of the positional numeral system for the purpose of binary-to-text translation. That method represents a binary file as a single big number with the radix 256 and then converts this big number to another one with an arbitrary radix (base) in a range from 2 to 94. Although this approach gives the minimum possible size overhead, unfortunately, it also has a number of downsides which make it hardly usable in a real-world situation. In this article, I'll show what is used in practice, which encodings could be found in the wild, and how to build your own encoder.
 
 

 
 
 
What's wrong with the positional single number encoding?
 
The main issue with converting a file as a big number in radix 256 to another big number with a smaller radix is that you need to read the whole file, load it to the memory and build actually that big number from each byte of the file. To construct a number, the Least Significant Byte (LSB), which is the last byte of a file, needs to be read and loaded. Although, there is not always enough memory to load a whole file as well as there is not always the whole file is available at any given time. For instance, if it's being transmitted over a network and only a small amount of bytes from the beginning (from the Most Significant Byte, MSB) has been loaded. This issue is usually addressed by processing a file as a stream of bytes, in chunks, which then are being converted in the same way (by converting a number from one base to another). These chunks are much smaller and, ideally, fit the CPU registers' size (up to 8 bytes). The only question here is how to find the best size and ratio of such chunks (input and output) to keep the size overhead as closely as possible to a minimum available by treating files as big numbers.
 
 

 
 
 
 
What's the essence of a positional numeral system?
 
In the positional numeral systems, everything turns around a radix (base) which shows how many different symbols are used to represent values. The actual glyph doesn't matter. Only their quantity. All these symbols are grouped in an alphabet (a table) where every symbol is defined by its own position, and this position represents its value. As long as counting starts from 0, the maximum symbol's value, in any numeral system, is always radix - 1. For instance, in the numeral system with a radix 10 (Decimal), the maximum value has a symbol '9'. But, for a system with a radix 2 (Binary), the maximum value has a symbol '1'. When symbols from an alphabet appear as a part of a number, they are called digits. A digit's position, in this case, is called index and defines the power of a radix while its value (position in the alphabet) defines a coefficient within the power of that radix.
 
 

 
 
The first crucial conclusion here is that any number, represented in some positional numeral system, gets its meaning only when is known its radix.
 
 

 
 
The second conclusion is not so obvious. Humans in most cases nowadays use the Decimal numeral system. Numbers gain more sense for them when they are represented as Decimal numbers and this is the system that is used the most for calculations. To any symbol in an alphabet is assigned its certain position which is a number with some radix. In most cases, this radix is 10 (Decimal). The Decimal numeral system is a temporary system that is used for converting one numeral system to another. Every time, when a number is defined by a radix, this radix is Decimal, no matter what's the radix of a number. Every time, when there is a need to convert a number X with radix M to a number Y with radix M, both numbers (X and Y) are represented by some certain alphabets (which define symbols with values), but their radixes (M and N) are always represented in Decimal system, thus, Decimal system is used as an intermediate numeral system to which a number X is converted first, and then the intermediate number is converted to a number Y. The intermediate numeral system could have been any radix, but radix 10 is what people use for calculations and that's what can be found in most converters implementations.
 
 

 
 
The third conclusion is even more important. Symbols don't bring any value, only their position in the alphabet. This means we need to know not only an actual number's representation but also its radix and an alphabet - the table that contains symbols assigned to values (position within the table). A good example is an alphabet of 16 symbols for Hexadecimal numbers (radix 16). There are first 10 digits linked to equivalent values, so the symbol '0' is linked to 0, '1' to 1, and so on up to the symbol '9' linked to 9. The rest 6 values (from 10 to 15) linked to English letter symbols (from 'A' to 'F'). And again, these values (positions in the table) are all Decimal numbers (radix 10). By the way, the table could have been different, but that's what is used by convention, so anyone is able to interpret Hexadecimal numbers in the same way.
 
 

 
 
 
 
Where does the overhead come from?
 
Let's take a look at a few examples. This is a number '123' that is represented by three symbols, but until we know a radix, it is not possible to understand its value. If the radix is 10 then it is 'one hundred twenty three' in the Decimal system and it can be calculated by the formula for converting a numeral system with any radix to radix 10 (because all numbers in this formula have radix 10): 1*10^2 + 2*10^1 + 3*10^0 = 123. If the radix is 8, then it is an Octal system and it is constructed as 1*8^2 + 2*8^1 + 3*8^0 which gives us a Decimal number 83. So, '123 base 8' equals to '83 base 10'. It is worth noticing that converting a number to a higher radix leads to lower a number of symbols needed for its representation. The converse is also true. If a number 83 with a radix 10 is converted to a radix 2, it gets a form '1010011'. Notice, the radix is changed from 10 to 2 and the number of symbols changed from 2 to 7! As lower a radix gets, as more symbols appear in representation.
 
 

 
 
Let's get back to binary files. What we can determine as 'symbol representation' or 'digits', 'alphabet', and 'radix' based on a structure of an ordinary file? Any file consists of bytes as it is the minimum addressable group of bits. It cannot be less than 8 bits. So, we can think about a number representation as of some amount of bytes. The chunks can vary from 1 byte to a file's size. For example, if there is only one byte, then the number consists of only one digit. One byte or 8 bits (binary digits with a radix 2) allows one to represent 2^8 = 256 different numbers. That means, we can persist 256 different symbols with their positions to build an alphabet. The good news, such a table has already been standardized many years ago and called ASCII. And the last thing, as the alphabet size is 256 symbols then a radix is also 256. Here is our number: a number of bytes in the chunk that we are going to process are the number of digits, a radix is 256, and the coefficient has a range from 0 to 255. For example, if a group of bytes to read from a stream and process at once consists of 4 bytes (from MSB to LSB): [13, 200, 3, 65] then our number can be represented as a Decimal number (radix 10) as 13*256^3 + 200*256^2 + 3*256^1 + 65*256^0 = 231211841
 
 

 
 
As it was discussed in the previous article, we can use no more than 94 different symbols to reliably represent texts. Thus, the desired radix lies somewhere in the range from 2 to 94. Even 94 is much less than 256, so a number's representation in a new radix is likely to have more symbols. This means, in turn, that the output group will have more bytes as it is a minimum amount of data we can operate on, even if a digit represented by a symbol needs fewer bits. You'll still need to allocate the whole byte for each symbol in the new radix number representation. Some amount of bits in such bytes will never be used. This is the root of inefficiency, and that's why it's highly important to find a good ratio of output to input byte groups. For instance, the most used nowadays Base64 encoding converts binary files to texts by reading 3-bytes groups from the input stream, represents them as a 3-digits number with a radix 256 (log[256^3, 2] = 24 bit), and then converts this number to a 4-digits number with a radix 64 (log[64^4, 2] = 24 bit), which in turn is written to the output stream as a group of 4 bytes. So, the ratio of output to input is 4/3 = 1.333333. In other words, the size overhead is 33.(3)%. There are a few considerations behind the logic of choosing the exact combination of input and output groups for a streaming conversion, which includes a target radix, a desirable/available alphabet, an ability to natively compute on a CPU, etc.
 
 

 
 
 
 
How to calculate a minimal overhead?
 
Let's calculate first, how many digits of a target base (radix) are needed to represent exactly the same number in the initial base. For instance, there is given a number 123 with a radix 10. How many bits (binary digits, a radix 2) are needed to represent the same decimal number? Every digit is a coefficient of power of a base. If it is not enough, one more base is added in power +1 to finally construct a number. Keeping in mind that counting starts from 0, if it's said that to represent some number 8 bit are needed, this means all bases in powers from 0 to 7 with their coefficients have to be summed up. Thus, to find out a number of digits needed to represent the number in some radix, we need to find an exponent, to which a new radix needs to be exponentiated. In our case, for a base-10 number 123, we need to calculate an exponent of a base-2 by using a logarithm function: log[123, 2] = 6.9425145. This means, to represent a number 123 with base 10, a little bit less than 7 bits will be enough. All computer systems operate on a set of natural numbers only. It is not possible to use 6.9425145 bits as this number is an approximated value of needed bits. 6 bits apparently won't be enough (2^6 = 64, which is much less than 123), so the only right approach is always to round up (by calling a ceil function) any non-integer values. Unfortunately, 7 bits are able to represent a bigger number (2^7 = 128) and this again contributes to a final overhead.
 
 

 
 
Let's have a look at the Base64 again. We know already (but not why is that, yet), that this streaming system uses 3 input bytes (a 3-digit number with a base 256) and converts them to a number with a base 64. How many base-64 digits will this number contain? The answer is log[256^3, 64] = 4, four digits, hence 4 symbols from the base64 alphabet.
 
 

 
 
While looking for the good input and output group sizes it's good to know a theoretically possible minimum of the overhead. To find it out, we need to do a similar calculation but take the minimally possible amount of input data, which is one byte (8 bits, decimal 2^8 = 256). For the Base64, it is log[256, 64] = 1.33(3), that is again 33.(3)%. For the Base32 it is log[256, 32] =  1.6, that is 60%. And for the Base16 it is log[256, 16] = 2, that is 100%. Wow! These theoretical numbers are exactly the same as practically used ratios of output bytes to input bytes give. Here are they: for the Base64 it is 4 / 3 = 1.33(3), for the Base32 it is 8 / 5 = 1.6, and for the Base16 it is 2 / 1 = 2. There is one interesting fact, all these three bases (16, 32, 64) have one thing in common - they all are powers of two! This leads us to the conclusion that converting numbers within the "power of two" bases allows one to get the best possible ratio and match precisely an input bits group to an output bits group. Although it is not always desirable or even possible. Sometimes there is a need to use a specific alphabet, e.g. in Base36, or the minimal overhead, e.g. in Base85 or Base94. All these bases are not the "powers of two", so a tradeoff has to be found to minimize the overhead.
 
 

 
 
 
 
How to calculate optimal input and output groups?
 
Alright, we've calculated a number of digits needed to represent some number in another base. But, why is that only a theoretical minimum? Why in practice it would need more? And, why would we still need to find a good ratio of output to input byte groups? To answer these questions, let's have a look at the Base85 encoding. To represent 1 byte (Base256) of information in Base85, it needs log[256, 85] = 1.24816852 digits. But, we can't use 1.248 digits. Only positive whole numbers are available! 1 digit is neither possible (too little). Then, 2 digits are the only way to go. In other words, to represent 1 byte (with a number in Base256), in fact, we'd need 2 bytes (with a number in Base85), where ~75% of space will be wasted, as the ratio is 2/1 = 2 and this is a 100% overhead, instead of a theoretical 24.8%. There is no point to use 1-byte input group and 2-bytes output group. Thus, there should be some good input and output groups so their ratio goes as close as possible to a calculated minimum or even match it!
 
 

 
 
The following approach starts from 1-byte group and using the same formula, every time checks a number of digits in the destination base. if it's not close enough, increments the input group by 1 byte and checks again. You can decide on your own, what is the applicable size of an input group and how close to the whole number up (ceil function) the output group needs to be.
 
 

 
 
This code goes through all bases, from 2 to 94, and prints a first found input/output group that has a delta between the number of digits and its rounded value less or equal 0.1, if any. That is, ceil(x) - x <=0.1. I limited an input group by 20 bytes but in reality, groups larger than 8 bytes (64bit) will require either a more complicated implementation still based on 64bit variable types or the big number mathematics which would bring it back to the solution from the previous article.
 
 

 
 
from math import log, ceil
 
 def find_dec_fractions(num):
     for i, k in [(i, log(256**i, num)) for i in range(1,20)]:
diff --git a/articles/turing/index.html b/articles/turing/index.html
index 20dcf749..78872bda 100644
--- a/articles/turing/index.html
+++ b/articles/turing/index.html
@@ -1 +1 @@
- Turing: thesis, machine, completeness :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Turing: thesis, machine, completeness
 
on 2019-12-15 in note about cs mindmap ~1 min read 
 
A formal system in the computability theory
 
 
Alan Turing is one of the pioneers of the computability theory and logic formalization. He came up with the hypothesis of which algorithms can be implemented and computed by machines (Turing's thesis), created an abstract model of such machine (Turing machine), and described absolutely vital abilities of any system for being able to realize any logic that can be computed (Turing completeness).
 
 

 
 
Turing's thesis is only one of the existing formal systems in the computability theory. There are also λ-calculus, Markov algorithms, but they all were implemented on the Turing Machine that is used at this time as a general computational model to classify which real-world systems (mostly programming languages) are able to compute mathematical functions or implement algorithms.
 
 

 
 
All existing computability theories are defined on discrete values, and the domain is the set of Natural numbers.
 
 

 
 
I prepared several mindmaps to summarize basic ideas and statements:
 
     
  • Turing's thesis:
    •  
 Turing's thesis 
 

 
 
     
  • Turing machine:
    •  
 Turing machine 
 

 
 
     
  • Turing completeness:
    •  
 Turing completeness 
 

 
  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
+ Turing: thesis, machine, completeness :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Turing: thesis, machine, completeness
 
on 2019-12-15 in note about cs mindmap ~1 min read 
 
A formal system in the computability theory
 
 
Alan Turing is one of the pioneers of the computability theory and logic formalization. He came up with the hypothesis of which algorithms can be implemented and computed by machines (Turing's thesis), created an abstract model of such machine (Turing machine), and described absolutely vital abilities of any system for being able to realize any logic that can be computed (Turing completeness).
 
 

 
 
Turing's thesis is only one of the existing formal systems in the computability theory. There are also λ-calculus, Markov algorithms, but they all were implemented on the Turing Machine that is used at this time as a general computational model to classify which real-world systems (mostly programming languages) are able to compute mathematical functions or implement algorithms.
 
 

 
 
All existing computability theories are defined on discrete values, and the domain is the set of Natural numbers.
 
 

 
 
I prepared several mindmaps to summarize basic ideas and statements:
 
     
  • Turing's thesis:
    •  
 Turing's thesis 
 

 
 
     
  • Turing machine:
    •  
 Turing machine 
 

 
 
     
  • Turing completeness:
    •  
 Turing completeness 
 

 
  
 
 
 Found a bug or typo? Please, send me feedback or submit a PR on Github. 
 
 
This is my personal blog. All ideas, opinions, examples, and other information that can be found here are my own and belong entirely to me. This is the result of my personal efforts and activities at my free time. It doesn't relate to any professional work I've done and doesn't have correlations with any companies I worked for, I'm currently working, or will work in the future.
 
 
 
 
  
  
\ No newline at end of file
diff --git a/articles/udp-link/index.html b/articles/udp-link/index.html
index f90fae8e..1c509830 100644
--- a/articles/udp-link/index.html
+++ b/articles/udp-link/index.html
@@ -1,4 +1,4 @@
- Using udp-link to enhance TCP connections stability :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Using udp-link to enhance TCP connections stability
 
on 2024-01-16 in article about networking tools ~5 min read 
 
A UDP transport layer implementation for proxying TCP connections
 
 
TLDR: quick summary of the article
 
 

 
 
I recently discovered udp-link, a very useful project for all those guys like me who spend most of their working time in terminals over ssh connections. The tool implements the UDP transport layer, which acts as a proxy for TCP connections. It's designed to be integrated into the OpenSSH configuration. However, with a little trick, it can also be used as a general-purpose TCP-over-UDP proxy. udp-link greatly improves the stability of connections over unreliable networks that experience packet loss and intermittent connectivity. It also includes an IP roaming, which allows TCP connections to remain alive even if an IP address changes.
 
 

 
 
udp-link is written in C by Pavel Gulchuk, who has a lot of experience in running unreliable networks. Despite being a young project, the version v0.4 shows pretty stable results. Once configured, you won't think about it anymore. Unless you're surprised every time when ssh connections don't brake, survive a laptop's sleep mode and connections to different Wi-Fi networks.
 
 

 
 
In the current architecture, the client-side tool takes data on the standard input and sends it to the server side via UDP. The same copy of the tool takes that data from the network on a specific UDP port and sends it to a TCP service (local or remote from a server-side perspective). The destination TCP service and a UDP listening port on the server side can be specified on the client at startup. Otherwise, a TCP connection will be established with 127.0.0.1:22 and a port is randomly chosen from a predefined port range. Note that the server firewall should allow the traffic to this port range on UDP. The TCP service can also reside on a different host, if the server side is used as a jumpbox. I consider it one of the greatest features that udp-link uses a zero server-side configuration, all configuration tweaks happen only on the client side.
 
 

 
 
udp-link on the server side does not run as a daemon or listen on a UDP port all the time. Instead, the client initiates the invocation of the tool on the on the server side in listening mode with a randomly generated key. This key is used to authenticate the client connection. This is done on demand by establishing a normal ssh connection over TCP with the server side, temporarily, just to run the tool in the background. The connection is then closed. This is where a secure client authentication comes into play. udp-link doesn't encrypt the transferred data, which is useful when is used together with ssh because it avoids a double encryption, but needs to be kept that in mind when used with other configurations.
 
 

 
 
To start using udp-link, you need to clone the repository, compile, and install the tool on both sides
 
git clone https://github.com/pgul/udp-link.git
+ Using udp-link to enhance TCP connections stability :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Using udp-link to enhance TCP connections stability
 
on 2024-01-16 in article about networking tools ~5 min read 
 
A UDP transport layer implementation for proxying TCP connections
 
 
TLDR: quick summary of the article
 
 

 
 
I recently discovered udp-link, a very useful project for all those guys like me who spend most of their working time in terminals over ssh connections. The tool implements the UDP transport layer, which acts as a proxy for TCP connections. It's designed to be integrated into the OpenSSH configuration. However, with a little trick, it can also be used as a general-purpose TCP-over-UDP proxy. udp-link greatly improves the stability of connections over unreliable networks that experience packet loss and intermittent connectivity. It also includes an IP roaming, which allows TCP connections to remain alive even if an IP address changes.
 
 

 
 
udp-link is written in C by Pavel Gulchuk, who has a lot of experience in running unreliable networks. Despite being a young project, the version v0.4 shows pretty stable results. Once configured, you won't think about it anymore. Unless you're surprised every time when ssh connections don't brake, survive a laptop's sleep mode and connections to different Wi-Fi networks.
 
 

 
 
In the current architecture, the client-side tool takes data on the standard input and sends it to the server side via UDP. The same copy of the tool takes that data from the network on a specific UDP port and sends it to a TCP service (local or remote from a server-side perspective). The destination TCP service and a UDP listening port on the server side can be specified on the client at startup. Otherwise, a TCP connection will be established with 127.0.0.1:22 and a port is randomly chosen from a predefined port range. Note that the server firewall should allow the traffic to this port range on UDP. The TCP service can also reside on a different host, if the server side is used as a jumpbox. I consider it one of the greatest features that udp-link uses a zero server-side configuration, all configuration tweaks happen only on the client side.
 
 

 
 
udp-link on the server side does not run as a daemon or listen on a UDP port all the time. Instead, the client initiates the invocation of the tool on the on the server side in listening mode with a randomly generated key. This key is used to authenticate the client connection. This is done on demand by establishing a normal ssh connection over TCP with the server side, temporarily, just to run the tool in the background. The connection is then closed. This is where a secure client authentication comes into play. udp-link doesn't encrypt the transferred data, which is useful when is used together with ssh because it avoids a double encryption, but needs to be kept that in mind when used with other configurations.
 
 

 
 
To start using udp-link, you need to clone the repository, compile, and install the tool on both sides
 
git clone https://github.com/pgul/udp-link.git
 cd  udp-link
 make
 sudo make install
diff --git a/authors/index.html b/authors/index.html
index 6af70e12..1464dac4 100644
--- a/authors/index.html
+++ b/authors/index.html
@@ -1 +1 @@
- Authors :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Authors
 
 
 
  
 vorakl 22 articles 
  
 
 
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Authors :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Authors
 
 
 
  
 vorakl 22 articles 
  
 
 
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/authors/vorakl/index.html b/authors/vorakl/index.html
index ac40cabb..54fd3ccb 100644
--- a/authors/vorakl/index.html
+++ b/authors/vorakl/index.html
@@ -1 +1 @@
- Articles written by vorakl :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Articles written by vorakl
 
 
  
 2024-02-20 How to sort arrays natively in Bash 
 
 2024-02-18 Availability calculation in "nines" notation 
 
 2024-02-17 A little mess with function parameters in Python 
 
 2024-01-16 Using udp-link to enhance TCP connections stability 
 
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2020-01-18 My notes for the "Pragmatic Thinking and Learning" book 
 
 2019-12-20 Computer Science vs Information Technology 
 
 2019-12-19 Who is an engineer 
 
 2019-12-15 Algorithm is... 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Articles written by vorakl :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Articles written by vorakl
 
 
  
 2024-02-20 How to sort arrays natively in Bash 
 
 2024-02-18 Availability calculation in "nines" notation 
 
 2024-02-17 A little mess with function parameters in Python 
 
 2024-01-16 Using udp-link to enhance TCP connections stability 
 
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2020-01-18 My notes for the "Pragmatic Thinking and Learning" book 
 
 2019-12-20 Computer Science vs Information Technology 
 
 2019-12-19 Who is an engineer 
 
 2019-12-15 Algorithm is... 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/authors/vorakl/p2/index.html b/authors/vorakl/p2/index.html
index 889932ec..1dd89a1b 100644
--- a/authors/vorakl/p2/index.html
+++ b/authors/vorakl/p2/index.html
@@ -1 +1 @@
- Articles written by vorakl :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Articles written by vorakl
 
 
  
 2019-12-15 Turing: thesis, machine, completeness 
 
 2019-08-21 Organizing Unstructured Data 
 
 2019-08-16 Number Classification 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
 2019-06-26 Managing your plans in the S.M.A.R.T. way 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
 2019-06-15 Maslow's hierarchy of needs 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Articles written by vorakl :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Articles written by vorakl
 
 
  
 2019-12-15 Turing: thesis, machine, completeness 
 
 2019-08-21 Organizing Unstructured Data 
 
 2019-08-16 Number Classification 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
 2019-06-26 Managing your plans in the S.M.A.R.T. way 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
 2019-06-15 Maslow's hierarchy of needs 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/authors/vorakl/p3/index.html b/authors/vorakl/p3/index.html
index 57071914..cff0ca55 100644
--- a/authors/vorakl/p3/index.html
+++ b/authors/vorakl/p3/index.html
@@ -1 +1 @@
- Articles written by vorakl :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Articles written by vorakl
 
 
  
 2019-06-12 Structured Programming Paradigm 
 
 2019-06-10 SQ3R 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Articles written by vorakl :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Articles written by vorakl
 
 
  
 2019-06-12 Structured Programming Paradigm 
 
 2019-06-10 SQ3R 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/categories/article/index.html b/categories/article/index.html
index 2c8fb22d..90ffbf6d 100644
--- a/categories/article/index.html
+++ b/categories/article/index.html
@@ -1 +1 @@
- A full story about some specific topic :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A full story about some specific topic
 
 
  
 2024-02-18 Availability calculation in "nines" notation 
 
 2024-01-16 Using udp-link to enhance TCP connections stability 
 
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ A full story about some specific topic :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A full story about some specific topic
 
 
  
 2024-02-18 Availability calculation in "nines" notation 
 
 2024-01-16 Using udp-link to enhance TCP connections stability 
 
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/categories/howto/index.html b/categories/howto/index.html
index f884754d..0ccbb5b3 100644
--- a/categories/howto/index.html
+++ b/categories/howto/index.html
@@ -1 +1 @@
- A practical guide how to make something :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A practical guide how to make something
 
 
  
 2024-02-20 How to sort arrays natively in Bash 
 
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ A practical guide how to make something :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A practical guide how to make something
 
 
  
 2024-02-20 How to sort arrays natively in Bash 
 
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/categories/index.html b/categories/index.html
index fdfe60e7..19a46fa5 100644
--- a/categories/index.html
+++ b/categories/index.html
@@ -1 +1 @@
- Categories :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Categories
 
 
 
  
 article A full story about some specific topic 
 
 howto A practical guide how to make something 
 
 note A brief record of thoughts 
 
 tutorial A theoretical explanation of a topic with examples 
  
 
 
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Categories :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Categories
 
 
 
  
 article A full story about some specific topic 
 
 howto A practical guide how to make something 
 
 note A brief record of thoughts 
 
 tutorial A theoretical explanation of a topic with examples 
  
 
 
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/categories/note/index.html b/categories/note/index.html
index d55192aa..9eb893af 100644
--- a/categories/note/index.html
+++ b/categories/note/index.html
@@ -1 +1 @@
- A brief record of thoughts :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A brief record of thoughts
 
 
  
 2020-01-18 My notes for the "Pragmatic Thinking and Learning" book 
 
 2019-12-20 Computer Science vs Information Technology 
 
 2019-12-19 Who is an engineer 
 
 2019-12-15 Algorithm is... 
 
 2019-12-15 Turing: thesis, machine, completeness 
 
 2019-08-21 Organizing Unstructured Data 
 
 2019-08-16 Number Classification 
 
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-06-26 Managing your plans in the S.M.A.R.T. way 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ A brief record of thoughts :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A brief record of thoughts
 
 
  
 2020-01-18 My notes for the "Pragmatic Thinking and Learning" book 
 
 2019-12-20 Computer Science vs Information Technology 
 
 2019-12-19 Who is an engineer 
 
 2019-12-15 Algorithm is... 
 
 2019-12-15 Turing: thesis, machine, completeness 
 
 2019-08-21 Organizing Unstructured Data 
 
 2019-08-16 Number Classification 
 
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-06-26 Managing your plans in the S.M.A.R.T. way 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/categories/note/p2/index.html b/categories/note/p2/index.html
index 1a3abb6f..959f59b6 100644
--- a/categories/note/p2/index.html
+++ b/categories/note/p2/index.html
@@ -1 +1 @@
- A brief record of thoughts :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A brief record of thoughts
 
 
  
 2019-06-15 Maslow's hierarchy of needs 
 
 2019-06-12 Structured Programming Paradigm 
 
 2019-06-10 SQ3R 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ A brief record of thoughts :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A brief record of thoughts
 
 
  
 2019-06-15 Maslow's hierarchy of needs 
 
 2019-06-12 Structured Programming Paradigm 
 
 2019-06-10 SQ3R 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/categories/tutorial/index.html b/categories/tutorial/index.html
index 84d71822..ab1c705b 100644
--- a/categories/tutorial/index.html
+++ b/categories/tutorial/index.html
@@ -1 +1 @@
- A theoretical explanation of a topic with examples :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A theoretical explanation of a topic with examples
 
 
  
 2024-02-17 A little mess with function parameters in Python 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ A theoretical explanation of a topic with examples :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A theoretical explanation of a topic with examples
 
 
  
 2024-02-17 A little mess with function parameters in Python 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/index.html b/index.html
index 1b19932f..63ce5bb4 100644
--- a/index.html
+++ b/index.html
@@ -1 +1 @@
- Notes on Systems and Software Engineering :: Vorakl's notes   
\ No newline at end of file
+ Notes on Systems and Software Engineering :: Vorakl's notes   
\ No newline at end of file
diff --git a/news/index.html b/news/index.html
index 8eb6e74d..63af378c 100644
--- a/news/index.html
+++ b/news/index.html
@@ -1 +1 @@
- Notes on Systems and Software Engineering :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Notes on Systems and Software Engineering
 
 
  
 2024-02-20 How to sort arrays natively in Bash 
 
 2024-02-18 Availability calculation in "nines" notation 
 
 2024-02-17 A little mess with function parameters in Python 
 
 2024-01-16 Using udp-link to enhance TCP connections stability 
 
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2020-01-18 My notes for the "Pragmatic Thinking and Learning" book 
 
 2019-12-20 Computer Science vs Information Technology 
 
 2019-12-19 Who is an engineer 
 
 2019-12-15 Algorithm is... 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Notes on Systems and Software Engineering :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Notes on Systems and Software Engineering
 
 
  
 2024-02-20 How to sort arrays natively in Bash 
 
 2024-02-18 Availability calculation in "nines" notation 
 
 2024-02-17 A little mess with function parameters in Python 
 
 2024-01-16 Using udp-link to enhance TCP connections stability 
 
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2020-01-18 My notes for the "Pragmatic Thinking and Learning" book 
 
 2019-12-20 Computer Science vs Information Technology 
 
 2019-12-19 Who is an engineer 
 
 2019-12-15 Algorithm is... 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/news/p2/index.html b/news/p2/index.html
index 561b039b..548eff19 100644
--- a/news/p2/index.html
+++ b/news/p2/index.html
@@ -1 +1 @@
- Notes on Systems and Software Engineering :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Notes on Systems and Software Engineering
 
 
  
 2019-12-15 Turing: thesis, machine, completeness 
 
 2019-08-21 Organizing Unstructured Data 
 
 2019-08-16 Number Classification 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
 2019-06-26 Managing your plans in the S.M.A.R.T. way 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
 2019-06-15 Maslow's hierarchy of needs 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Notes on Systems and Software Engineering :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Notes on Systems and Software Engineering
 
 
  
 2019-12-15 Turing: thesis, machine, completeness 
 
 2019-08-21 Organizing Unstructured Data 
 
 2019-08-16 Number Classification 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
 2019-06-26 Managing your plans in the S.M.A.R.T. way 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
 2019-06-15 Maslow's hierarchy of needs 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/news/p3/index.html b/news/p3/index.html
index 292e1c5e..66ad3561 100644
--- a/news/p3/index.html
+++ b/news/p3/index.html
@@ -1 +1 @@
- Notes on Systems and Software Engineering :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Notes on Systems and Software Engineering
 
 
  
 2019-06-12 Structured Programming Paradigm 
 
 2019-06-10 SQ3R 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Notes on Systems and Software Engineering :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Notes on Systems and Software Engineering
 
 
  
 2019-06-12 Structured Programming Paradigm 
 
 2019-06-10 SQ3R 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/pages/about/index.html b/pages/about/index.html
index 01aba4db..b3428fff 100644
--- a/pages/about/index.html
+++ b/pages/about/index.html
@@ -1 +1 @@
- About me and my blog :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
About me and my blog
 
A brief overview of my career and how I got to where I am today
 
 
Hi! I'm glad you're here, and I hope you find something interesting and useful in this blog. It serves an educational purpose and is a kind of notebook where I document what I've learned in computer systems and software engineering. So you are likely to find articles about computer systems architecture, embedded systems, operating systems, clouds, observability, performance tuning and troubleshooting, programming, automation, debugging, and so on. If you find a mistake or a typo, have an idea to discuss, or just want to tell me something, send me a message via any of the available channels and I'll be happy to chat with you!
 
 

 
 
I currently reside with my family in Seattle, WA. However, prior to moving to the United States, I also lived in Ukraine, Poland, and Germany. I've been working in the field of computer systems engineering for two decades. Although, my formal education is in Telecommunications Engineering which I received at the National Technical University of Ukraine and graduated with a Master's degree in 2004. In my professional life, I have worked for companies in many different industries, including retail, e-commerce, Internet service providers, and multinational high-tech companies such as IBM and Google, to name a few. Throughout my career, I've been designing, building and maintaining computer networks and distributed systems with a focus on automation and site reliability. I have experience with various types of infrastructure, from on-premises to cloud computing.
 
 

 
 
My personal journey with computers began around 1992 when I started taking after-school programming classes at Sphere and had some fun with Basic on a Soviet PDP-11-compatible hardware. I continued writing some simple games in Basic at home when my parents bought me a ZX Spectrum, which was the iconic computer of the time. Later, when I had access to a PC with an Intel Pentium processor, I opened up a whole world of Turbo Pascal 7.0 and desperately tried to understand object-oriented concepts with the Turbo Vision 2.0 library. Last 2 years of high school I was taking extra programming classes and got a certificate of Assistant Programmer in 1998. In the first years of university I suddenly discovered computer networks. During the day I was administering a FidoNet node on the dial-up line at home, but at night I was surfing the Internet. I was so interested in networking and Unix that in my last year at university I applied for the 2-year course of Computer Networks and Systems Administration at computer academy Step, which I finished in 2005 with a diploma with honors.
 
 

 
  
 
 
 
 
  
  
\ No newline at end of file
+ About me and my blog :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
About me and my blog
 
A brief overview of my career and how I got to where I am today
 
 
Hi! I'm glad you're here, and I hope you find something interesting and useful in this blog. It serves an educational purpose and is a kind of notebook where I document what I've learned in computer systems and software engineering. So you are likely to find articles about computer systems architecture, embedded systems, operating systems, clouds, observability, performance tuning and troubleshooting, programming, automation, debugging, and so on. If you find a mistake or a typo, have an idea to discuss, or just want to tell me something, send me a message via any of the available channels and I'll be happy to chat with you!
 
 

 
 
I currently reside with my family in Seattle, WA. However, prior to moving to the United States, I also lived in Ukraine, Poland, and Germany. I've been working in the field of computer systems engineering for two decades. Although, my formal education is in Telecommunications Engineering which I received at the National Technical University of Ukraine and graduated with a Master's degree in 2004. In my professional life, I have worked for companies in many different industries, including retail, e-commerce, Internet service providers, and multinational high-tech companies such as IBM and Google, to name a few. Throughout my career, I've been designing, building and maintaining computer networks and distributed systems with a focus on automation and site reliability. I have experience with various types of infrastructure, from on-premises to cloud computing.
 
 

 
 
My personal journey with computers began around 1992 when I started taking after-school programming classes at Sphere and had some fun with Basic on a Soviet PDP-11-compatible hardware. I continued writing some simple games in Basic at home when my parents bought me a ZX Spectrum, which was the iconic computer of the time. Later, when I had access to a PC with an Intel Pentium processor, I opened up a whole world of Turbo Pascal 7.0 and desperately tried to understand object-oriented concepts with the Turbo Vision 2.0 library. Last 2 years of high school I was taking extra programming classes and got a certificate of Assistant Programmer in 1998. In the first years of university I suddenly discovered computer networks. During the day I was administering a FidoNet node on the dial-up line at home, but at night I was surfing the Internet. I was so interested in networking and Unix that in my last year at university I applied for the 2-year course of Computer Networks and Systems Administration at computer academy Step, which I finished in 2005 with a diploma with honors.
 
 

 
  
 
 
 
 
  
  
\ No newline at end of file
diff --git a/pages/contacts/index.html b/pages/contacts/index.html
index 6216068b..8dd72923 100644
--- a/pages/contacts/index.html
+++ b/pages/contacts/index.html
@@ -1 +1 @@
- Contact information :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Contact information
 
A few ways to reach me out
 
 
  
 
 
 
 
  
  
\ No newline at end of file
+ Contact information :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Contact information
 
A few ways to reach me out
 
 
  
 
 
 
 
  
  
\ No newline at end of file
diff --git a/pages/feedback/index.html b/pages/feedback/index.html
index 89a12aa3..81dcc471 100644
--- a/pages/feedback/index.html
+++ b/pages/feedback/index.html
@@ -1 +1 @@
- Feedback :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Feedback
 
Send me a direct message
 
 
 
 E-mail 
  
 
 Name 
  
 
 Message * 
  
 
 * - required fields
  
 
 
 
 
  
  
\ No newline at end of file
+ Feedback :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Feedback
 
Send me a direct message
 
 
 
 E-mail 
  
 
 Name 
  
 
 Message * 
  
 
 * - required fields
  
 
 
 
 
  
  
\ No newline at end of file
diff --git a/pages/index.html b/pages/index.html
index c45b0866..bfc616a1 100644
--- a/pages/index.html
+++ b/pages/index.html
@@ -1 +1 @@
- Pages :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Pages
 
 
 
  
 about A brief overview of my career and how I got to where I am today 
 
 contacts A few ways to reach me out 
 
 feedback Send me a direct message 
 
 projects Some of my projects which I'd like to show 
 
 reading Books that I've read and recommend 
  
 
 
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Pages :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Pages
 
 
 
  
 about A brief overview of my career and how I got to where I am today 
 
 contacts A few ways to reach me out 
 
 feedback Send me a direct message 
 
 projects Some of my projects which I'd like to show 
 
 reading Books that I've read and recommend 
  
 
 
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/pages/projects/index.html b/pages/projects/index.html
index a52c37a8..15e19c19 100644
--- a/pages/projects/index.html
+++ b/pages/projects/index.html
@@ -1 +1 @@
- My projects :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
My projects
 
Some of my projects which I'd like to show
 
 
     
  •  
    Vorakl's Bash Library
     
    A collection of Bash Modules
     
     
    •  
  •  
    TrivialRC
     
    A minimalistic RC system and process manager for containers and applications
     
     
    •  
  •  
    Aves
     
    A theme for the Pelican (a static site generator)
     
     
    •  
  
 
 
 
 
  
  
\ No newline at end of file
+ My projects :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
My projects
 
Some of my projects which I'd like to show
 
 
     
  •  
    Vorakl's Bash Library
     
    A collection of Bash Modules
     
     
    •  
  •  
    TrivialRC
     
    A minimalistic RC system and process manager for containers and applications
     
     
    •  
  •  
    Aves
     
    A theme for the Pelican (a static site generator)
     
     
    •  
  
 
 
 
 
  
  
\ No newline at end of file
diff --git a/pages/reading/index.html b/pages/reading/index.html
index 8f6fb6f7..a883c943 100644
--- a/pages/reading/index.html
+++ b/pages/reading/index.html
@@ -1 +1 @@
- Reading Recommendations :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Reading Recommendations
 
Books that I've read and recommend
 
 
Even before I start reading a new technical book, I analyze it, read through the content, skim it, and even read a few paragraphs on a random page. It is usually pretty clear within the first few minutes whether a book is worth your time. Reading techniques like SQ3R make all this preparation and reading a thoughtful and organized process. When I read a book from cover to cover, there is no doubt that I found it to be a really good book and I can definitely recommend it.
 
 

 
 
These books are either technical or nearly technical. The order in each category is more or less important, but the categories themselves can be based on current interest.
 
 

 
 
Happy reading!
 
 

 
  
 
Computer Science Fundamentals
 
     
  •  
    Understanding the Digital World, Brian Kernighan, 2017
     
    As the preface perfectly puts it: "What you need to know about computers, the Internet, privacy, and security. Although the main reader of the book is a non-technical person with no previous IT background, everyone can benefit from reading it. The author describes very technical and difficult topics in simple terms and makes understanding the big picture easy and natural.
     
     
    •  
  •  
    [russian] Программирование: введение в профессию, Том 1: азы программирования, А.В.Столяров, 2016
     
    В первый том "вошли избранные сведения из истории вычислительной техники, обсуждение некоторых областей математики, непосредственно используемых программистами (таких как алгебра логики, комбинаторика, позиционные системы счисления), математических основ программирования (теория вычислимости и теория алгоритмов), принципы построения и функционирования вычислительных систем". Это одна из тех не многих книг, которая формирует фундамент, подкреплённый не только теоретическим материалом, но и полезными практическими примерами.
     
     
    •  
 
 

 
 
 
OOP
  
 

 
 
 
 
 
Programming
 
 

 
 
 
BASH
 
     
  •  
    The Unix Programming Environment, Brian W. Kernighan, Rob Pike, 1984
     
    .
     
     
    •  
  •  
    Pro Bash Programming, 2ed, Chris F. A. Johnson, Jayant Varma, 2015
     
    .
     
     
    •  
  •  
    Bash Quick Start Guide, Tom Ryder, 2018
     
    .
     
     
    •  
 
 

 
 
 
 
C / C++
 
     
  •  
    The C Programming Language, 2ed, Brian W. Kernighan, Dennis M. Ritchie, 1988
     
    .
     
     
    •  
  •  
    [russian] Введение в язык С++, 5изд, Столяров А.В., 2020
     
    .
     
     
    •  
 
 

 
 
 
 
SQL
 
     
  •  
    Simply SQL, Rudy Limeback, 2009
     
    .
     
     
    •  
 
 

 
 
 
 
 
Software Development Practice
 
     
  •  
    Understanding Software, Max Kanat-Alexander, 2017
     
    The whole book is pure experience. It's completely practical, and not only gives you an understanding of what it's like to build and maintain enormously large software, but also how to succeed by avoiding pitfalls and doing things right.
     
     
    •  
  •  
    The DevOps Adoption Playbook, Sanjeev Sharma, 2017
     
    In my opinion, one of the most detailed and clear overviews of DevOps practice in software development. Unlike many other books on the subject, the author goes back to the roots, explaining what it was like to develop software long before Agile, what were the main pitfalls of common Agile practice, and what exactly and how was addressed by DevOps.
     
     
    •  
 
 

 
 
 
 
Databases
 
     
  •  
    Database Systems: Design, Implementation, & Management, 13ed, Carlos Coronel, Steven Morris, 2018
     
    .
     
     
    •  
 
 

 
 
 
 
Networks
 
     
  •  
    [russian] Компьютерные сети. Принципы, технологии, протоколы, Олифер В.Г., Олифер Н.А, 2001
     
    .
     
     
    •  
  •  
    High Performance Browser Networking, Ilya Grigorik, 2013
     
    .
     
     
    •  
  •  
    HTTP: The Definitive Guide, David Gourley and Brian Totty, 2002
     
    .
     
     
    •  
 
 

 
 
 
 
Soft Skills
 
     
  •  
    Pragmatic Thinking and Learning, Andy Hunt, 2008
     
    The great example of work that everyone would benefit from, no matter what their profession or hobby, because it provides a lot of insight into how our brains actually work, how we learn and remember, how we think and solve problems. Here are my notes in the form of mind maps.
     
     
    •  
  •  
    Soft Skills: The software developer's life manual, John Z. Sonmez, 2014
     
    This book covers various non-technical but very important aspects of the life of a modern software engineer, such as career, productivity, communication, setting life goals, learning, etc.
     
     
    •  
 
 

 
 
 
 
Miscellaneous
 
     
  •  
    Hackers, Steven Levy, 2010
     
    .
     
     
    •  
  •  
    The Art of Deception: Controlling the Human Element of Security, Kevin D. Mitnick, William L. Simon, 2003
     
    .
     
     
    •  
  •  
    Just for FUN: the story of an accidental revolutionary, Linus Torvalds, David Diamond, 2002
     
    .
     
     
    •  
  •  
    Python Interviews, Mike Driscoll, 2018
     
    A collection of independent interviews with a dozen major figures in the Python community, helping to understand the language's past, how it evolved from the beginning, where it's at now, and where it is going.
     
     
    •  
  
 
 
 
 
 
  
  
\ No newline at end of file
+ Reading Recommendations :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
 
Reading Recommendations
 
Books that I've read and recommend
 
 
Even before I start reading a new technical book, I analyze it, read through the content, skim it, and even read a few paragraphs on a random page. It is usually pretty clear within the first few minutes whether a book is worth your time. Reading techniques like SQ3R make all this preparation and reading a thoughtful and organized process. When I read a book from cover to cover, there is no doubt that I found it to be a really good book and I can definitely recommend it.
 
 

 
 
These books are either technical or nearly technical. The order in each category is more or less important, but the categories themselves can be based on current interest.
 
 

 
 
Happy reading!
 
 

 
  
 
Computer Science Fundamentals
 
     
  •  
    Understanding the Digital World, Brian Kernighan, 2017
     
    As the preface perfectly puts it: "What you need to know about computers, the Internet, privacy, and security. Although the main reader of the book is a non-technical person with no previous IT background, everyone can benefit from reading it. The author describes very technical and difficult topics in simple terms and makes understanding the big picture easy and natural.
     
     
    •  
  •  
    [russian] Программирование: введение в профессию, Том 1: азы программирования, А.В.Столяров, 2016
     
    В первый том "вошли избранные сведения из истории вычислительной техники, обсуждение некоторых областей математики, непосредственно используемых программистами (таких как алгебра логики, комбинаторика, позиционные системы счисления), математических основ программирования (теория вычислимости и теория алгоритмов), принципы построения и функционирования вычислительных систем". Это одна из тех не многих книг, которая формирует фундамент, подкреплённый не только теоретическим материалом, но и полезными практическими примерами.
     
     
    •  
 
 

 
 
 
OOP
  
 

 
 
 
 
 
Programming
 
 

 
 
 
BASH
 
     
  •  
    The Unix Programming Environment, Brian W. Kernighan, Rob Pike, 1984
     
    .
     
     
    •  
  •  
    Pro Bash Programming, 2ed, Chris F. A. Johnson, Jayant Varma, 2015
     
    .
     
     
    •  
  •  
    Bash Quick Start Guide, Tom Ryder, 2018
     
    .
     
     
    •  
 
 

 
 
 
 
C / C++
 
     
  •  
    The C Programming Language, 2ed, Brian W. Kernighan, Dennis M. Ritchie, 1988
     
    .
     
     
    •  
  •  
    [russian] Введение в язык С++, 5изд, Столяров А.В., 2020
     
    .
     
     
    •  
 
 

 
 
 
 
SQL
 
     
  •  
    Simply SQL, Rudy Limeback, 2009
     
    .
     
     
    •  
 
 

 
 
 
 
 
Software Development Practice
 
     
  •  
    Understanding Software, Max Kanat-Alexander, 2017
     
    The whole book is pure experience. It's completely practical, and not only gives you an understanding of what it's like to build and maintain enormously large software, but also how to succeed by avoiding pitfalls and doing things right.
     
     
    •  
  •  
    The DevOps Adoption Playbook, Sanjeev Sharma, 2017
     
    In my opinion, one of the most detailed and clear overviews of DevOps practice in software development. Unlike many other books on the subject, the author goes back to the roots, explaining what it was like to develop software long before Agile, what were the main pitfalls of common Agile practice, and what exactly and how was addressed by DevOps.
     
     
    •  
 
 

 
 
 
 
Databases
 
     
  •  
    Database Systems: Design, Implementation, & Management, 13ed, Carlos Coronel, Steven Morris, 2018
     
    .
     
     
    •  
 
 

 
 
 
 
Networks
 
     
  •  
    [russian] Компьютерные сети. Принципы, технологии, протоколы, Олифер В.Г., Олифер Н.А, 2001
     
    .
     
     
    •  
  •  
    High Performance Browser Networking, Ilya Grigorik, 2013
     
    .
     
     
    •  
  •  
    HTTP: The Definitive Guide, David Gourley and Brian Totty, 2002
     
    .
     
     
    •  
 
 

 
 
 
 
Soft Skills
 
     
  •  
    Pragmatic Thinking and Learning, Andy Hunt, 2008
     
    The great example of work that everyone would benefit from, no matter what their profession or hobby, because it provides a lot of insight into how our brains actually work, how we learn and remember, how we think and solve problems. Here are my notes in the form of mind maps.
     
     
    •  
  •  
    Soft Skills: The software developer's life manual, John Z. Sonmez, 2014
     
    This book covers various non-technical but very important aspects of the life of a modern software engineer, such as career, productivity, communication, setting life goals, learning, etc.
     
     
    •  
 
 

 
 
 
 
Miscellaneous
 
     
  •  
    Hackers, Steven Levy, 2010
     
    .
     
     
    •  
  •  
    The Art of Deception: Controlling the Human Element of Security, Kevin D. Mitnick, William L. Simon, 2003
     
    .
     
     
    •  
  •  
    Just for FUN: the story of an accidental revolutionary, Linus Torvalds, David Diamond, 2002
     
    .
     
     
    •  
  •  
    Python Interviews, Mike Driscoll, 2018
     
    A collection of independent interviews with a dozen major figures in the Python community, helping to understand the language's past, how it evolved from the beginning, where it's at now, and where it is going.
     
     
    •  
  
 
 
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/ai/index.html b/tags/ai/index.html
index d9cc0a9f..cbb3ce7a 100644
--- a/tags/ai/index.html
+++ b/tags/ai/index.html
@@ -1 +1 @@
- Artificial Intelligence :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Artificial Intelligence
 
 
  
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Artificial Intelligence :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Artificial Intelligence
 
 
  
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/bash/index.html b/tags/bash/index.html
index 2093600e..ab112ff4 100644
--- a/tags/bash/index.html
+++ b/tags/bash/index.html
@@ -1 +1 @@
- A programming language :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A programming language
 
 
  
 2024-02-20 How to sort arrays natively in Bash 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ A programming language :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A programming language
 
 
  
 2024-02-20 How to sort arrays natively in Bash 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/binary-to-text/index.html b/tags/binary-to-text/index.html
index 0cabffef..02f790b2 100644
--- a/tags/binary-to-text/index.html
+++ b/tags/binary-to-text/index.html
@@ -1 +1 @@
- Translation between binary data and text :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Translation between binary data and text
 
 
  
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Translation between binary data and text :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Translation between binary data and text
 
 
  
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/cs/index.html b/tags/cs/index.html
index 44981dbc..aa5165d4 100644
--- a/tags/cs/index.html
+++ b/tags/cs/index.html
@@ -1 +1 @@
- Computer Science :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Computer Science
 
 
  
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2019-12-20 Computer Science vs Information Technology 
 
 2019-12-15 Algorithm is... 
 
 2019-12-15 Turing: thesis, machine, completeness 
 
 2019-08-21 Organizing Unstructured Data 
 
 2019-08-16 Number Classification 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-06-12 Structured Programming Paradigm 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Computer Science :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Computer Science
 
 
  
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2019-12-20 Computer Science vs Information Technology 
 
 2019-12-15 Algorithm is... 
 
 2019-12-15 Turing: thesis, machine, completeness 
 
 2019-08-21 Organizing Unstructured Data 
 
 2019-08-16 Number Classification 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-06-12 Structured Programming Paradigm 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/encoding/index.html b/tags/encoding/index.html
index 40dd6f32..6996b3ea 100644
--- a/tags/encoding/index.html
+++ b/tags/encoding/index.html
@@ -1 +1 @@
- Transformation one form of data to anoher :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Transformation one form of data to anoher
 
 
  
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Transformation one form of data to anoher :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Transformation one form of data to anoher
 
 
  
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/html/index.html b/tags/html/index.html
index 97670e08..982a5a87 100644
--- a/tags/html/index.html
+++ b/tags/html/index.html
@@ -1 +1 @@
- The markup language for Web pages :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
The markup language for Web pages
 
 
  
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ The markup language for Web pages :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
The markup language for Web pages
 
 
  
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/http/index.html b/tags/http/index.html
index 46e6c950..66c03a3e 100644
--- a/tags/http/index.html
+++ b/tags/http/index.html
@@ -1 +1 @@
- The HyperText Transfer Protocol :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
The HyperText Transfer Protocol
 
 
  
 2019-07-02 How to redirect a static website on the Github Pages 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ The HyperText Transfer Protocol :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
The HyperText Transfer Protocol
 
 
  
 2019-07-02 How to redirect a static website on the Github Pages 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/index.html b/tags/index.html
index f8742cc3..a1405e06 100644
--- a/tags/index.html
+++ b/tags/index.html
@@ -1 +1 @@
- Tags :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Tags
 
 
 
  
 bash A programming language 
 
 programming Everything related to creating computer programs 
 
 sre Site Reliability Engineering Practice 
 
 python A programming language 
 
 networking Computer Networks and Protocols 
 
 tools A variety of useful tools 
 
 cs Computer Science 
 
 binary-to-text Translation between binary data and text 
 
 encoding Transformation one form of data to anoher 
 
 learning Materials about acquiring new skills 
 
 mindmap Notes in hierarchically linked diagrams 
 
 it Information Technology 
 
 math Topics related to mathematics 
 
 ai Artificial Intelligence 
 
 web The World Wide Web (hypertext system) 
 
 html The markup language for Web pages 
 
 http The HyperText Transfer Protocol 
 
 management Management, Planning, Leading, etc 
 
 psychology All about the human mind and behavior 
  
 
 
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Tags :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Tags
 
 
 
  
 bash A programming language 
 
 programming Everything related to creating computer programs 
 
 sre Site Reliability Engineering Practice 
 
 python A programming language 
 
 networking Computer Networks and Protocols 
 
 tools A variety of useful tools 
 
 cs Computer Science 
 
 binary-to-text Translation between binary data and text 
 
 encoding Transformation one form of data to anoher 
 
 learning Materials about acquiring new skills 
 
 mindmap Notes in hierarchically linked diagrams 
 
 it Information Technology 
 
 math Topics related to mathematics 
 
 ai Artificial Intelligence 
 
 web The World Wide Web (hypertext system) 
 
 html The markup language for Web pages 
 
 http The HyperText Transfer Protocol 
 
 management Management, Planning, Leading, etc 
 
 psychology All about the human mind and behavior 
  
 
 
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/it/index.html b/tags/it/index.html
index 2209db9d..1d86647c 100644
--- a/tags/it/index.html
+++ b/tags/it/index.html
@@ -1 +1 @@
- Information Technology :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Information Technology
 
 
  
 2019-12-20 Computer Science vs Information Technology 
 
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Information Technology :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Information Technology
 
 
  
 2019-12-20 Computer Science vs Information Technology 
 
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/learning/index.html b/tags/learning/index.html
index 190db041..bb10de0c 100644
--- a/tags/learning/index.html
+++ b/tags/learning/index.html
@@ -1 +1 @@
- Materials about acquiring new skills :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Materials about acquiring new skills
 
 
  
 2020-01-18 My notes for the "Pragmatic Thinking and Learning" book 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
 2019-06-10 SQ3R 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Materials about acquiring new skills :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Materials about acquiring new skills
 
 
  
 2020-01-18 My notes for the "Pragmatic Thinking and Learning" book 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
 2019-06-10 SQ3R 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/management/index.html b/tags/management/index.html
index 2fb009b0..0b0272cc 100644
--- a/tags/management/index.html
+++ b/tags/management/index.html
@@ -1 +1 @@
- Management, Planning, Leading, etc :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Management, Planning, Leading, etc
 
 
  
 2019-06-26 Managing your plans in the S.M.A.R.T. way 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Management, Planning, Leading, etc :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Management, Planning, Leading, etc
 
 
  
 2019-06-26 Managing your plans in the S.M.A.R.T. way 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/math/index.html b/tags/math/index.html
index e59f4578..0d2c9d46 100644
--- a/tags/math/index.html
+++ b/tags/math/index.html
@@ -1 +1 @@
- Topics related to mathematics :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Topics related to mathematics
 
 
  
 2019-08-16 Number Classification 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Topics related to mathematics :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Topics related to mathematics
 
 
  
 2019-08-16 Number Classification 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/mindmap/index.html b/tags/mindmap/index.html
index 4766fa9a..d03777d8 100644
--- a/tags/mindmap/index.html
+++ b/tags/mindmap/index.html
@@ -1 +1 @@
- Notes in hierarchically linked diagrams :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Notes in hierarchically linked diagrams
 
 
  
 2020-01-18 My notes for the "Pragmatic Thinking and Learning" book 
 
 2019-12-20 Computer Science vs Information Technology 
 
 2019-12-19 Who is an engineer 
 
 2019-12-15 Algorithm is... 
 
 2019-12-15 Turing: thesis, machine, completeness 
 
 2019-06-26 Managing your plans in the S.M.A.R.T. way 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
 2019-06-15 Maslow's hierarchy of needs 
 
 2019-06-12 Structured Programming Paradigm 
 
 2019-06-10 SQ3R 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Notes in hierarchically linked diagrams :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Notes in hierarchically linked diagrams
 
 
  
 2020-01-18 My notes for the "Pragmatic Thinking and Learning" book 
 
 2019-12-20 Computer Science vs Information Technology 
 
 2019-12-19 Who is an engineer 
 
 2019-12-15 Algorithm is... 
 
 2019-12-15 Turing: thesis, machine, completeness 
 
 2019-06-26 Managing your plans in the S.M.A.R.T. way 
 
 2019-06-21 Dreyfus model of skill acquisition 
 
 2019-06-15 Maslow's hierarchy of needs 
 
 2019-06-12 Structured Programming Paradigm 
 
 2019-06-10 SQ3R 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/networking/index.html b/tags/networking/index.html
index 7fe96ff7..c0e41fba 100644
--- a/tags/networking/index.html
+++ b/tags/networking/index.html
@@ -1 +1 @@
- Computer Networks and Protocols :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Computer Networks and Protocols
 
 
  
 2024-01-16 Using udp-link to enhance TCP connections stability 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Computer Networks and Protocols :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Computer Networks and Protocols
 
 
  
 2024-01-16 Using udp-link to enhance TCP connections stability 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/programming/index.html b/tags/programming/index.html
index 3cd70364..58cdf090 100644
--- a/tags/programming/index.html
+++ b/tags/programming/index.html
@@ -1 +1 @@
- Everything related to creating computer programs :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Everything related to creating computer programs
 
 
  
 2024-02-20 How to sort arrays natively in Bash 
 
 2024-02-17 A little mess with function parameters in Python 
 
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2019-08-21 Organizing Unstructured Data 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-06-12 Structured Programming Paradigm 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Everything related to creating computer programs :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Everything related to creating computer programs
 
 
  
 2024-02-20 How to sort arrays natively in Bash 
 
 2024-02-17 A little mess with function parameters in Python 
 
 2020-05-13 The zoo of binary-to-text encoding schemes 
 
 2020-04-18 Convert binary data to a text with the lowest overhead 
 
 2019-08-21 Organizing Unstructured Data 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
 2019-07-22 Constraint Satisfaction Problem (CSP) 
 
 2019-06-12 Structured Programming Paradigm 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/psychology/index.html b/tags/psychology/index.html
index 3942cd1e..4d443645 100644
--- a/tags/psychology/index.html
+++ b/tags/psychology/index.html
@@ -1 +1 @@
- All about the human mind and behavior :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
All about the human mind and behavior
 
 
  
 2019-06-15 Maslow's hierarchy of needs 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ All about the human mind and behavior :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
All about the human mind and behavior
 
 
  
 2019-06-15 Maslow's hierarchy of needs 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/python/index.html b/tags/python/index.html
index 358668ab..420a69b1 100644
--- a/tags/python/index.html
+++ b/tags/python/index.html
@@ -1 +1 @@
- A programming language :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A programming language
 
 
  
 2024-02-17 A little mess with function parameters in Python 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ A programming language :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A programming language
 
 
  
 2024-02-17 A little mess with function parameters in Python 
 
 2019-08-08 A converter of a character's case and Ascii codes 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/sre/index.html b/tags/sre/index.html
index 396083b8..f860ee67 100644
--- a/tags/sre/index.html
+++ b/tags/sre/index.html
@@ -1 +1 @@
- Site Reliability Engineering Practice :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Site Reliability Engineering Practice
 
 
  
 2024-02-18 Availability calculation in "nines" notation 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ Site Reliability Engineering Practice :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
Site Reliability Engineering Practice
 
 
  
 2024-02-18 Availability calculation in "nines" notation 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/tools/index.html b/tags/tools/index.html
index aecd8140..17973ada 100644
--- a/tags/tools/index.html
+++ b/tags/tools/index.html
@@ -1 +1 @@
- A variety of useful tools :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A variety of useful tools
 
 
  
 2024-01-16 Using udp-link to enhance TCP connections stability 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ A variety of useful tools :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
A variety of useful tools
 
 
  
 2024-01-16 Using udp-link to enhance TCP connections stability 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/tags/web/index.html b/tags/web/index.html
index 665a63b5..e5dd22c8 100644
--- a/tags/web/index.html
+++ b/tags/web/index.html
@@ -1 +1 @@
- The World Wide Web (hypertext system) :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
The World Wide Web (hypertext system)
 
 
  
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
+ The World Wide Web (hypertext system) :: Vorakl's notes  
 
 
 Vorakl's notes 
  
 
 
 
 
 
 
 
 
 
The World Wide Web (hypertext system)
 
 
  
 2019-07-18 How to remove a webpage from the Google index 
 
 2019-07-02 How to redirect a static website on the Github Pages 
 
  
 
 
  
 
 
 
  
  
\ No newline at end of file
diff --git a/theme/bootstrap-pygments.bundle.min.css b/theme/bootstrap-pygments.bundle.min.css
index 73c83f97..e01af670 100644
--- a/theme/bootstrap-pygments.bundle.min.css
+++ b/theme/bootstrap-pygments.bundle.min.css
@@ -4,4 +4,80 @@ article,aside,details,figcaption,figure,footer,header,hgroup,nav,section{display
 @media (max-width:480px){.nav-collapse{-webkit-transform:translate3d(0, 0, 0);} .page-header h1 small{display:block;line-height:18px;} input[class*="span"],select[class*="span"],textarea[class*="span"],.uneditable-input{display:block;width:100%;height:28px;-webkit-box-sizing:border-box;-moz-box-sizing:border-box;-ms-box-sizing:border-box;box-sizing:border-box;} .input-prepend input[class*="span"],.input-append input[class*="span"]{width:auto;} input[type="checkbox"],input[type="radio"]{border:1px solid #ccc;} .form-horizontal .control-group>label{float:none;width:auto;padding-top:0;text-align:left;} .form-horizontal .controls{margin-left:0;} .form-horizontal .control-list{padding-top:0;} .form-horizontal .form-actions{padding-left:10px;padding-right:10px;} .modal{position:absolute;top:10px;left:10px;right:10px;width:auto;margin:0;}.modal.fade.in{top:auto;} .modal-header .close{padding:10px;margin:-10px;} .carousel-caption{position:static;}}@media (max-width:768px){.container{width:auto;padding:0 20px;} .row-fluid{width:100%;} .row{margin-left:0;} .row>[class*="span"],.row-fluid>[class*="span"]{float:none;display:block;width:auto;margin:0;}}@media (min-width:768px) and (max-width:980px){.row{margin-left:-20px;*zoom:1;}.row:before,.row:after{display:table;content:"";} .row:after{clear:both;} [class*="span"]{float:left;margin-left:20px;} .span1{width:42px;} .span2{width:104px;} .span3{width:166px;} .span4{width:228px;} .span5{width:290px;} .span6{width:352px;} .span7{width:414px;} .span8{width:476px;} .span9{width:538px;} .span10{width:600px;} .span11{width:662px;} .span12,.container{width:724px;} .offset1{margin-left:82px;} .offset2{margin-left:144px;} .offset3{margin-left:206px;} .offset4{margin-left:268px;} .offset5{margin-left:330px;} .offset6{margin-left:392px;} .offset7{margin-left:454px;} .offset8{margin-left:516px;} .offset9{margin-left:578px;} .offset10{margin-left:640px;} .offset11{margin-left:702px;} .row-fluid{width:100%;*zoom:1;}.row-fluid:before,.row-fluid:after{display:table;content:"";} .row-fluid:after{clear:both;} .row-fluid>[class*="span"]{float:left;margin-left:2.762430939%;} .row-fluid>[class*="span"]:first-child{margin-left:0;} .row-fluid .span1{width:5.801104972%;} .row-fluid .span2{width:14.364640883%;} .row-fluid .span3{width:22.928176794%;} .row-fluid .span4{width:31.491712705%;} .row-fluid .span5{width:40.055248616%;} .row-fluid .span6{width:48.618784527%;} .row-fluid .span7{width:57.182320438000005%;} .row-fluid .span8{width:65.74585634900001%;} .row-fluid .span9{width:74.30939226%;} .row-fluid .span10{width:82.87292817100001%;} .row-fluid .span11{width:91.436464082%;} .row-fluid .span12{width:99.999999993%;} input.span1,textarea.span1,.uneditable-input.span1{width:32px;} input.span2,textarea.span2,.uneditable-input.span2{width:94px;} input.span3,textarea.span3,.uneditable-input.span3{width:156px;} input.span4,textarea.span4,.uneditable-input.span4{width:218px;} input.span5,textarea.span5,.uneditable-input.span5{width:280px;} input.span6,textarea.span6,.uneditable-input.span6{width:342px;} input.span7,textarea.span7,.uneditable-input.span7{width:404px;} input.span8,textarea.span8,.uneditable-input.span8{width:466px;} input.span9,textarea.span9,.uneditable-input.span9{width:528px;} input.span10,textarea.span10,.uneditable-input.span10{width:590px;} input.span11,textarea.span11,.uneditable-input.span11{width:652px;} input.span12,textarea.span12,.uneditable-input.span12{width:714px;}}@media (max-width:675px){body{padding-top:0;} .navbar-fixed-top{position:static;margin-bottom:18px;} .navbar-fixed-top .navbar-inner{padding:5px;} .navbar .container{width:auto;padding:0;} .navbar .brand{padding-left:10px;padding-right:10px;margin:0 0 0 -5px;} .navbar .nav-collapse{clear:left;} .navbar .nav{float:none;margin:0 0 9px;} .navbar .nav>li{float:none;} .navbar .nav>li>a{margin-bottom:2px;} .navbar .nav>.divider-vertical{display:none;} .navbar .nav>li>a,.navbar .dropdown-menu a{padding:6px 15px;font-weight:bold;color:#999999;-webkit-border-radius:3px;-moz-border-radius:3px;border-radius:3px;} .navbar .dropdown-menu li+li a{margin-bottom:2px;} .navbar .nav>li>a:hover,.navbar .dropdown-menu a:hover{background-color:#222222;} .navbar .dropdown-menu{position:static;top:auto;left:auto;float:none;display:block;max-width:none;margin:0 15px;padding:0;background-color:transparent;border:none;-webkit-border-radius:0;-moz-border-radius:0;border-radius:0;-webkit-box-shadow:none;-moz-box-shadow:none;box-shadow:none;} .navbar .dropdown-menu:before,.navbar .dropdown-menu:after{display:none;} .navbar .dropdown-menu .divider{display:none;} .navbar-form,.navbar-search{float:none;padding:9px 15px;margin:9px 0;border-top:1px solid #222222;border-bottom:1px solid #222222;-webkit-box-shadow:inset 0 1px 0 rgba(255, 255, 255, 0.1),0 1px 0 rgba(255, 255, 255, 0.1);-moz-box-shadow:inset 0 1px 0 rgba(255, 255, 255, 0.1),0 1px 0 rgba(255, 255, 255, 0.1);box-shadow:inset 0 1px 0 rgba(255, 255, 255, 0.1),0 1px 0 rgba(255, 255, 255, 0.1);} .navbar .nav.pull-right{float:none;margin-left:0;} .navbar-static .navbar-inner{padding-left:10px;padding-right:10px;} .btn-navbar{display:block;} .nav-collapse{overflow:hidden;height:0;}}@media (min-width:980px){.nav-collapse.collapse{height:auto !important;}}@media (min-width:1200px){.row{margin-left:-30px;*zoom:1;}.row:before,.row:after{display:table;content:"";} .row:after{clear:both;} [class*="span"]{float:left;margin-left:30px;} .span1{width:70px;} .span2{width:170px;} .span3{width:270px;} .span4{width:370px;} .span5{width:470px;} .span6{width:570px;} .span7{width:670px;} .span8{width:770px;} .span9{width:870px;} .span10{width:970px;} .span11{width:1070px;} .span12,.container{width:1170px;} .offset1{margin-left:130px;} .offset2{margin-left:230px;} .offset3{margin-left:330px;} .offset4{margin-left:430px;} .offset5{margin-left:530px;} .offset6{margin-left:630px;} .offset7{margin-left:730px;} .offset8{margin-left:830px;} .offset9{margin-left:930px;} .offset10{margin-left:1030px;} .offset11{margin-left:1130px;} .row-fluid{width:100%;*zoom:1;}.row-fluid:before,.row-fluid:after{display:table;content:"";} .row-fluid:after{clear:both;} .row-fluid>[class*="span"]{float:left;margin-left:2.564102564%;} .row-fluid>[class*="span"]:first-child{margin-left:0;} .row-fluid .span1{width:5.982905983%;} .row-fluid .span2{width:14.529914530000001%;} .row-fluid .span3{width:23.076923077%;} .row-fluid .span4{width:31.623931624%;} .row-fluid .span5{width:40.170940171000005%;} .row-fluid .span6{width:48.717948718%;} .row-fluid .span7{width:57.264957265%;} .row-fluid .span8{width:65.81196581200001%;} .row-fluid .span9{width:74.358974359%;} .row-fluid .span10{width:82.905982906%;} .row-fluid .span11{width:91.45299145300001%;} .row-fluid .span12{width:100%;} input.span1,textarea.span1,.uneditable-input.span1{width:60px;} input.span2,textarea.span2,.uneditable-input.span2{width:160px;} input.span3,textarea.span3,.uneditable-input.span3{width:260px;} input.span4,textarea.span4,.uneditable-input.span4{width:360px;} input.span5,textarea.span5,.uneditable-input.span5{width:460px;} input.span6,textarea.span6,.uneditable-input.span6{width:560px;} input.span7,textarea.span7,.uneditable-input.span7{width:660px;} input.span8,textarea.span8,.uneditable-input.span8{width:760px;} input.span9,textarea.span9,.uneditable-input.span9{width:860px;} input.span10,textarea.span10,.uneditable-input.span10{width:960px;} input.span11,textarea.span11,.uneditable-input.span11{width:1060px;} input.span12,textarea.span12,.uneditable-input.span12{width:1160px;} .thumbnails{margin-left:-30px;} .thumbnails>li{margin-left:30px;}}
 /* Styling for Nav bar. Bg color, link color etc */
 .navbar-inner{padding-left:20px;padding-right:20px;background-color:#2E435E;background-image:-moz-linear-gradient(top,#2e435e,#205081);background-image:-ms-linear-gradient(top,#2e435e,#205081);background-image:-webkit-gradient(linear,0 0,0 100%,from(#2e435e),to(#205081));background-image:-webkit-linear-gradient(top,#2e435e,#205081);background-image:-o-linear-gradient(top,#2e435e,#205081);background-image:linear-gradient(top,#2e435e,#205081);filter:progid:DXImageTransform.Microsoft.gradient(startColorstr='#333',endColorstr='#222222',GradientType=0);-webkit-border-radius:0;-moz-border-radius:0;border-radius:0;-webkit-box-shadow:0 1px 3px rgba(0,0,0,0.25),inset 0 -1px 0 rgba(0,0,0,0.1);-moz-box-shadow:0 1px 3px rgba(0,0,0,0.25),inset 0 -1px 0 rgba(0,0,0,0.1);box-shadow:0 1px 3px rgba(0,0,0,0.25),inset 0 -1px 0 rgba(0,0,0,0.1)}.navbar .brand:hover{color:#49AFCD}.navbar .brand{padding:12px 20px 12px}.navbar .nav > li > a{color:#FFF;padding:12px 10px 12px}.navbar .nav > li > a:hover{color:#49AFCD}.navbar .divider-vertical{border-right:1px solid #4D4D4D}.navbar{margin-bottom:4em}body{font-size:1em}p{font-size:1em;line-height:1.6em;margin-top:0.2em;margin-bottom:0.2em}.p-footer{font-size:0.7em;line-height:1.6em;padding-bottom:2em}li{line-height:1.6em}h1,h2,h3,h4,h5,h6{padding:2px 0 2px;text-align:justify}.table th,.table td{line-height:1.3em}a:hover{text-decoration:underline}.article{padding-bottom:20px}.content-title{margin-bottom:30px;text-align:justify}.content p{text-align:justify;font-size:1.1rem}.content dd{text-align:justify}.article a:hover{text-decoration:underline}.iconmenu{margin-right:0.4em}.well.small{padding:6px 6px 4px;text-align:justify}.btn-info.xsmall{padding:1px 3px 1px;font-size:11px;float:right;-webkit-border-radius:2px;-moz-border-radius:2px;border-radius:2px}.btn-info.xsmall.left{padding:1px 3px 1px;font-size:11px;float:left;-webkit-border-radius:2px;-moz-border-radius:2px;border-radius:2px}.img.right{float:right;padding:10px 15px 10px 15px}.summary{margin-bottom:3em}.summary pre{display:none}.well{margin-bottom:5px}
-.highlight pre .hll{background-color:#ffc}.highlight pre{background:#f0f3f3}.highlight pre .c{color:#09f;font-style:italic}.highlight pre .err{color:#a00;background-color:#faa}.highlight pre .k{color:#069;font-weight:700}.highlight pre .o{color:#555}.highlight pre .cm{color:#09f;font-style:italic}.highlight pre .cp{color:#099}.highlight pre .c1{color:#09f;font-style:italic}.highlight pre .cs{color:#09f;font-weight:700;font-style:italic}.highlight pre .gd{background-color:#fcc;border:1px solid #c00}.highlight pre .ge{font-style:italic}.highlight pre .gr{color:red}.highlight pre .gh{color:#030;font-weight:700}.highlight pre .gi{background-color:#cfc;border:1px solid #0c0}.highlight pre .go{color:#aaa}.highlight pre .gp{color:#009;font-weight:700}.highlight pre .gs{font-weight:700}.highlight pre .gu{color:#030;font-weight:700}.highlight pre .gt{color:#9c6}.highlight pre .kc{color:#069;font-weight:700}.highlight pre .kd{color:#069;font-weight:700}.highlight pre .kn{color:#069;font-weight:700}.highlight pre .kp{color:#069}.highlight pre .kr{color:#069;font-weight:700}.highlight pre .kt{color:#078;font-weight:700}.highlight pre .m{color:#f60}.highlight pre .s{color:#c30}.highlight pre .na{color:#309}.highlight pre .nb{color:#366}.highlight pre .nc{color:#0a8;font-weight:700}.highlight pre .no{color:#360}.highlight pre .nd{color:#99f}.highlight pre .ni{color:#999;font-weight:700}.highlight pre .ne{color:#c00;font-weight:700}.highlight pre .nf{color:#c0f}.highlight pre .nl{color:#99f}.highlight pre .nn{color:#0cf;font-weight:700}.highlight pre .nt{color:#309;font-weight:700}.highlight pre .nv{color:#033}.highlight pre .ow{color:#000;font-weight:700}.highlight pre .w{color:#bbb}.highlight pre .mb{color:#f60}.highlight pre .mf{color:#f60}.highlight pre .mh{color:#f60}.highlight pre .mi{color:#f60}.highlight pre .mo{color:#f60}.highlight pre .sb{color:#c30}.highlight pre .sc{color:#c30}.highlight pre .sd{color:#c30;font-style:italic}.highlight pre .s2{color:#c30}.highlight pre .se{color:#c30;font-weight:700}.highlight pre .sh{color:#c30}.highlight pre .si{color:#a00}.highlight pre .sx{color:#c30}.highlight pre .sr{color:#3aa}.highlight pre .s1{color:#c30}.highlight pre .ss{color:#fc3}.highlight pre .bp{color:#366}.highlight pre .vc{color:#033}.highlight pre .vg{color:#033}.highlight pre .vi{color:#033}.highlight pre .il{color:#f60}
+.highlight .hll { background-color: #ffffcc }
+.highlight  { background: #f8f8f8; }
+.highlight .c { color: #8f5902; font-style: italic } /* Comment */
+.highlight .err { color: #a40000; border: 1px solid #ef2929 } /* Error */
+.highlight .g { color: #000000 } /* Generic */
+.highlight .k { color: #204a87; font-weight: bold } /* Keyword */
+.highlight .l { color: #000000 } /* Literal */
+.highlight .n { color: #000000 } /* Name */
+.highlight .o { color: #ce5c00; font-weight: bold } /* Operator */
+.highlight .x { color: #000000 } /* Other */
+.highlight .p { color: #000000; font-weight: bold } /* Punctuation */
+.highlight .ch { color: #8f5902; font-style: italic } /* Comment.Hashbang */
+.highlight .cm { color: #8f5902; font-style: italic } /* Comment.Multiline */
+.highlight .cp { color: #8f5902; font-style: italic } /* Comment.Preproc */
+.highlight .cpf { color: #8f5902; font-style: italic } /* Comment.PreprocFile */
+.highlight .c1 { color: #8f5902; font-style: italic } /* Comment.Single */
+.highlight .cs { color: #8f5902; font-style: italic } /* Comment.Special */
+.highlight .gd { color: #a40000 } /* Generic.Deleted */
+.highlight .ge { color: #000000; font-style: italic } /* Generic.Emph */
+.highlight .gr { color: #ef2929 } /* Generic.Error */
+.highlight .gh { color: #000080; font-weight: bold } /* Generic.Heading */
+.highlight .gi { color: #00A000 } /* Generic.Inserted */
+.highlight .go { color: #000000; font-style: italic } /* Generic.Output */
+.highlight .gp { color: #8f5902 } /* Generic.Prompt */
+.highlight .gs { color: #000000; font-weight: bold } /* Generic.Strong */
+.highlight .gu { color: #800080; font-weight: bold } /* Generic.Subheading */
+.highlight .gt { color: #a40000; font-weight: bold } /* Generic.Traceback */
+.highlight .kc { color: #204a87; font-weight: bold } /* Keyword.Constant */
+.highlight .kd { color: #204a87; font-weight: bold } /* Keyword.Declaration */
+.highlight .kn { color: #204a87; font-weight: bold } /* Keyword.Namespace */
+.highlight .kp { color: #204a87; font-weight: bold } /* Keyword.Pseudo */
+.highlight .kr { color: #204a87; font-weight: bold } /* Keyword.Reserved */
+.highlight .kt { color: #204a87; font-weight: bold } /* Keyword.Type */
+.highlight .ld { color: #000000 } /* Literal.Date */
+.highlight .m { color: #0000cf; font-weight: bold } /* Literal.Number */
+.highlight .s { color: #4e9a06 } /* Literal.String */
+.highlight .na { color: #c4a000 } /* Name.Attribute */
+.highlight .nb { color: #204a87 } /* Name.Builtin */
+.highlight .nc { color: #000000 } /* Name.Class */
+.highlight .no { color: #000000 } /* Name.Constant */
+.highlight .nd { color: #5c35cc; font-weight: bold } /* Name.Decorator */
+.highlight .ni { color: #ce5c00 } /* Name.Entity */
+.highlight .ne { color: #cc0000; font-weight: bold } /* Name.Exception */
+.highlight .nf { color: #000000 } /* Name.Function */
+.highlight .nl { color: #f57900 } /* Name.Label */
+.highlight .nn { color: #000000 } /* Name.Namespace */
+.highlight .nx { color: #000000 } /* Name.Other */
+.highlight .py { color: #000000 } /* Name.Property */
+.highlight .nt { color: #204a87; font-weight: bold } /* Name.Tag */
+.highlight .nv { color: #000000 } /* Name.Variable */
+.highlight .ow { color: #204a87; font-weight: bold } /* Operator.Word */
+.highlight .w { color: #f8f8f8; text-decoration: underline } /* Text.Whitespace */
+.highlight .mb { color: #0000cf; font-weight: bold } /* Literal.Number.Bin */
+.highlight .mf { color: #0000cf; font-weight: bold } /* Literal.Number.Float */
+.highlight .mh { color: #0000cf; font-weight: bold } /* Literal.Number.Hex */
+.highlight .mi { color: #0000cf; font-weight: bold } /* Literal.Number.Integer */
+.highlight .mo { color: #0000cf; font-weight: bold } /* Literal.Number.Oct */
+.highlight .sa { color: #4e9a06 } /* Literal.String.Affix */
+.highlight .sb { color: #4e9a06 } /* Literal.String.Backtick */
+.highlight .sc { color: #4e9a06 } /* Literal.String.Char */
+.highlight .dl { color: #4e9a06 } /* Literal.String.Delimiter */
+.highlight .sd { color: #8f5902; font-style: italic } /* Literal.String.Doc */
+.highlight .s2 { color: #4e9a06 } /* Literal.String.Double */
+.highlight .se { color: #4e9a06 } /* Literal.String.Escape */
+.highlight .sh { color: #4e9a06 } /* Literal.String.Heredoc */
+.highlight .si { color: #4e9a06 } /* Literal.String.Interpol */
+.highlight .sx { color: #4e9a06 } /* Literal.String.Other */
+.highlight .sr { color: #4e9a06 } /* Literal.String.Regex */
+.highlight .s1 { color: #4e9a06 } /* Literal.String.Single */
+.highlight .ss { color: #4e9a06 } /* Literal.String.Symbol */
+.highlight .bp { color: #3465a4 } /* Name.Builtin.Pseudo */
+.highlight .fm { color: #000000 } /* Name.Function.Magic */
+.highlight .vc { color: #000000 } /* Name.Variable.Class */
+.highlight .vg { color: #000000 } /* Name.Variable.Global */
+.highlight .vi { color: #000000 } /* Name.Variable.Instance */
+.highlight .vm { color: #000000 } /* Name.Variable.Magic */
+.highlight .il { color: #0000cf; font-weight: bold } /* Literal.Number.Integer.Long */
diff --git a/theme/pygments-tango.css b/theme/pygments-tango.css
new file mode 100644
index 00000000..0fa0beaa
--- /dev/null
+++ b/theme/pygments-tango.css
@@ -0,0 +1,77 @@
+.highlight .hll { background-color: #ffffcc }
+.highlight  { background: #f8f8f8; }
+.highlight .c { color: #8f5902; font-style: italic } /* Comment */
+.highlight .err { color: #a40000; border: 1px solid #ef2929 } /* Error */
+.highlight .g { color: #000000 } /* Generic */
+.highlight .k { color: #204a87; font-weight: bold } /* Keyword */
+.highlight .l { color: #000000 } /* Literal */
+.highlight .n { color: #000000 } /* Name */
+.highlight .o { color: #ce5c00; font-weight: bold } /* Operator */
+.highlight .x { color: #000000 } /* Other */
+.highlight .p { color: #000000; font-weight: bold } /* Punctuation */
+.highlight .ch { color: #8f5902; font-style: italic } /* Comment.Hashbang */
+.highlight .cm { color: #8f5902; font-style: italic } /* Comment.Multiline */
+.highlight .cp { color: #8f5902; font-style: italic } /* Comment.Preproc */
+.highlight .cpf { color: #8f5902; font-style: italic } /* Comment.PreprocFile */
+.highlight .c1 { color: #8f5902; font-style: italic } /* Comment.Single */
+.highlight .cs { color: #8f5902; font-style: italic } /* Comment.Special */
+.highlight .gd { color: #a40000 } /* Generic.Deleted */
+.highlight .ge { color: #000000; font-style: italic } /* Generic.Emph */
+.highlight .gr { color: #ef2929 } /* Generic.Error */
+.highlight .gh { color: #000080; font-weight: bold } /* Generic.Heading */
+.highlight .gi { color: #00A000 } /* Generic.Inserted */
+.highlight .go { color: #000000; font-style: italic } /* Generic.Output */
+.highlight .gp { color: #8f5902 } /* Generic.Prompt */
+.highlight .gs { color: #000000; font-weight: bold } /* Generic.Strong */
+.highlight .gu { color: #800080; font-weight: bold } /* Generic.Subheading */
+.highlight .gt { color: #a40000; font-weight: bold } /* Generic.Traceback */
+.highlight .kc { color: #204a87; font-weight: bold } /* Keyword.Constant */
+.highlight .kd { color: #204a87; font-weight: bold } /* Keyword.Declaration */
+.highlight .kn { color: #204a87; font-weight: bold } /* Keyword.Namespace */
+.highlight .kp { color: #204a87; font-weight: bold } /* Keyword.Pseudo */
+.highlight .kr { color: #204a87; font-weight: bold } /* Keyword.Reserved */
+.highlight .kt { color: #204a87; font-weight: bold } /* Keyword.Type */
+.highlight .ld { color: #000000 } /* Literal.Date */
+.highlight .m { color: #0000cf; font-weight: bold } /* Literal.Number */
+.highlight .s { color: #4e9a06 } /* Literal.String */
+.highlight .na { color: #c4a000 } /* Name.Attribute */
+.highlight .nb { color: #204a87 } /* Name.Builtin */
+.highlight .nc { color: #000000 } /* Name.Class */
+.highlight .no { color: #000000 } /* Name.Constant */
+.highlight .nd { color: #5c35cc; font-weight: bold } /* Name.Decorator */
+.highlight .ni { color: #ce5c00 } /* Name.Entity */
+.highlight .ne { color: #cc0000; font-weight: bold } /* Name.Exception */
+.highlight .nf { color: #000000 } /* Name.Function */
+.highlight .nl { color: #f57900 } /* Name.Label */
+.highlight .nn { color: #000000 } /* Name.Namespace */
+.highlight .nx { color: #000000 } /* Name.Other */
+.highlight .py { color: #000000 } /* Name.Property */
+.highlight .nt { color: #204a87; font-weight: bold } /* Name.Tag */
+.highlight .nv { color: #000000 } /* Name.Variable */
+.highlight .ow { color: #204a87; font-weight: bold } /* Operator.Word */
+.highlight .w { color: #f8f8f8; text-decoration: underline } /* Text.Whitespace */
+.highlight .mb { color: #0000cf; font-weight: bold } /* Literal.Number.Bin */
+.highlight .mf { color: #0000cf; font-weight: bold } /* Literal.Number.Float */
+.highlight .mh { color: #0000cf; font-weight: bold } /* Literal.Number.Hex */
+.highlight .mi { color: #0000cf; font-weight: bold } /* Literal.Number.Integer */
+.highlight .mo { color: #0000cf; font-weight: bold } /* Literal.Number.Oct */
+.highlight .sa { color: #4e9a06 } /* Literal.String.Affix */
+.highlight .sb { color: #4e9a06 } /* Literal.String.Backtick */
+.highlight .sc { color: #4e9a06 } /* Literal.String.Char */
+.highlight .dl { color: #4e9a06 } /* Literal.String.Delimiter */
+.highlight .sd { color: #8f5902; font-style: italic } /* Literal.String.Doc */
+.highlight .s2 { color: #4e9a06 } /* Literal.String.Double */
+.highlight .se { color: #4e9a06 } /* Literal.String.Escape */
+.highlight .sh { color: #4e9a06 } /* Literal.String.Heredoc */
+.highlight .si { color: #4e9a06 } /* Literal.String.Interpol */
+.highlight .sx { color: #4e9a06 } /* Literal.String.Other */
+.highlight .sr { color: #4e9a06 } /* Literal.String.Regex */
+.highlight .s1 { color: #4e9a06 } /* Literal.String.Single */
+.highlight .ss { color: #4e9a06 } /* Literal.String.Symbol */
+.highlight .bp { color: #3465a4 } /* Name.Builtin.Pseudo */
+.highlight .fm { color: #000000 } /* Name.Function.Magic */
+.highlight .vc { color: #000000 } /* Name.Variable.Class */
+.highlight .vg { color: #000000 } /* Name.Variable.Global */
+.highlight .vi { color: #000000 } /* Name.Variable.Instance */
+.highlight .vm { color: #000000 } /* Name.Variable.Magic */
+.highlight .il { color: #0000cf; font-weight: bold } /* Literal.Number.Integer.Long */