Guide to TEI usage

Jerome project: guide to TEI markup conventions

Overall organization

• Citation structure follows the logic of the MS
• Use a distinct TEI div unit for each era, where era is defined by a unique set of regnal fila
• Assign @n values to each div following guide in Helm edition. We will begin with Olympiad 1 = era 14.

Markup of tables

Within each era, the first table should be identified with an @n attribute t1; if there is more than one table in a single era, subsequent tables should be identified as tX.

Header rows

Tables generally begin with a special row giving header labels for the table. This row should have an @n value of header.

When labeling multiple header rows within the same division and table, label each as header_1, header_2, etc.

Every cell in the header role should a role attribute. Values should be one of olympiad, abraham, filum, annimundi, or spatium. Cells that have role filum should also have an @n attribute identifying which filum appears in that column. Use only values from the list on this page.

Data rows

Data rows fall in three categories:

1. Change of Olympiad: use @role = olympiad and @n = the Olympiad number in normal Arabic numerals (e.g., the row for Olympiad 1 should have @n attribute = 1).

2. Events such as change of rule without any numeric data: use @role = regnal with the @n attribute. For the @n attribute, use the format OLYMPIAD_r1 if the regnal row appears immediately after an Olympiad row, and the format OLYMPIAD_PRECEDING YEAR_r1 if the regnal row appears immediately after a data row, e.g. appearing after the second year of Olympiad 2 would have the @n attribute 2_1_r1. If regnal rows appear in an uninterrupted succession, the r1 would become r2, r3, etc. for every successive row

3. Normal data rows: no @role attribute needed; for @n attribute, use the format OLYMPIAD_YEAR, e.g., year 1 of the second Olympiad would have @n attribute 2_1.

4. If a row contains a mix of normal data, regnal information, or olympiads, the row is given an attribute in accordance with the highest priority data present, with Olympiads being highest priority and a data row being the lowest priority.

Identifying cells

1. If there is no text or number in a cell, identify each individual cell with the attribute @role=spatium.

2. If a change in rule occurs in the same row as a change in Olympiad, identify the row using the procedure outlined for any normal Olympiad row. However, in the individual cell where the change in rule occurs, identify that cell individually with the attribute @role = regnal.

3. If there is a change in name of the group in the header in the middle of the page, denoted most often by a circular data entry identical to those in the header row, the individual cell is labeled as @role = filium with an appropriate @n attribute consistent with the list on this page.

4. If there is only a number in the cell or plain text, it needs no additional identification.

Identifying Text

• When the author of the manuscript uses an abbreviation, use, first of all, the element choice. Within the choice brackets, add the elements abbr and expan. Within the abbr element, type the observed abbreviated form. Within the expan element type the inferred unabbreviated form. e.g. <choice>``<abbr>hr</abbr>``<expan>habetur </expan> </choice>

• If there is a name of a person, use the element persName to edit the text. E.g. <persName>Alexander`

• If there is the name of a geographic location - cities, countries, rivers, mountains, hills, etc. - use the element placeName to edit the text. E.g. <placeName>athenis</placeName>

• If an ethnic group is referred to, such as the Medes, use the element <rs> with the added attribute type="ethnic" to edit the text. e.g. <rs type="ethinc"> Medorum </rs>

• If the name of a person, place, or ethnic group is also abbreviated, then first use those types of tag and then put the abbreviation tag inside of the more specific ones. E.g. <placeName>``<choice>``<abbr>Iudea</abbr> <expan>Iudeam</expan> </choice>``<placeName>

  • Often, personal name will have adjectival descriptors, e.g. Nabuchdonosor secundus, Arctinus milesius, etc. In this case, the adjectival descriptor is included within the persName element. If the adjectival descriptor is also a ethnic group, continue to use the element rs within the persName element.

• Whenever there is a Roman numeral, either in the table proper or within the text, use num around the Roman numerals. If the number is written out as word, e.g. unus, duos, tres, etc., no additional element is required.

• If a certain word is unambiguously clear in its appearance, yet does not make syntactical sense, approach the word in one of two ways:

  • If the intended correct form can be easily inferred, i.e. a simple typographical error, use the element choice. Within the brackets of the choice, add the element sic first, followed by the element corr. Within the sic element, type the observed, incorrect form. Within the corr element, type the corrected form. E.g. <choice>``<sic>Ateniensium</sic>``<corr>Atheniensium</corr>``</choice>

  • If the intended correct form cannot be easily inferred, simply edit the text using the sic element.

• Sometimes, the scribe himself will correct the morphology of a word right on the page. Often this means writing the corrected form above the incorrect form. In this case, edit the text in the same way as the method for a typographical error noted above, using the syntax: <choice>``<sic>``</sic>``<corr>``</corr>``</choice>

• For numbers in the thousands, the scribe will often put a bar over the digits intended to be in the thousands. For example, if the number presented is IIIICCCC for 4,400, the following syntax must be used to represent this construction: <choice>``<abbr>``IIIICCCC``</abbr>``<expan>``<num>``IIII``</num>``milia``<num>``CCCC``</num>``</expan>``</choice>

• When the scribe refers to a written work, use the element title to edit the entire title of the work. E.g. <title>librum machabeorum<title>

• If there is a line spacing between two parts of the text in the same cell, use <p>``<p> commands to wrap each of the separated sections into individual paragraphs

• If the scribe uses a symbol in the text that looks similar to an '=' symbol (it often appears at the beginning of sentences), regardless if there is spacing between the sentences or not, interpret the symbol as meaning that the author recognizes that the succeeding text is a separate thought but part of the same cell. In such a case, once again use the <p>``</p> commands to wrap each of the individual texts within that one cell into separate paragraphs.

• If there is text in a cell with multiple regnal symbols to indicate the succession of multiple leaders in a short period of time, each regnal symbol should be interpreted as the beginning of a new paragraph. Therefore, the <p>``</p> commands should be used under this circumstance. A new paragraph can also be created when there is not a new regnal symbol if there are two consecutive regnal texts in different colors in the same cell.

• If the scribe seeks to indicate that text that is separated at a distance from a particular area of the page does belong in such a cell, a symbol may appear to the right side of the number that the text belongs with as well as in front of the text that matches with that number. In many cases three dots may be used, but other symbols have been used as well. In such a case, correctly place the text next to the corresponding number in XML, and use <note``</note> to wrap around the text that is being moved.

  • Note that sometimes the symbol may appear next to a number that the text does not correspond to in situations in which there is no number in the column that the text would go to. For example, in the CONSVLVM column, there are no numbers used, only spatium and text cells. While the symbol may be next to a number in the column on the left side of the CONSVLVM, it does not go in the spatium immediately to the right of that number, but in the column for the Roman consuls. The text can be identified to belong in the proper column based on its color, as it should match the color of the numbers for that column.