Develop mechanism for recognizing duplicate layers

[blewis]

There can be a variety of OGC and Esri URL forms that point to the same data on the same server.
There is often an OGC and an Esri version of the same data on the same server.
Cascading can also result in this situation.

In addition there are many copies of a given dataset (census for example) which are served from different servers but provide the exact same service to the end user.

The challenge is to use a combination of machine learning on endpoints and metadata combined with analysis of tiles returned from services, and perhaps combined with feature service requests, to develop a "likelihood of being the same" measure.

This information, combined with uptime and usage statistics could make it possible to provide recommendations to users when a the server behind a layer they are using disappears. Proper handling of duplicates will also improve search.

[capooti]

We do not need this for 0.3.5 milestone, correct?

[blewis]

yes, not needed for initial worldmap rollout

[capooti]

Moving to 0.4