We are starting to think of what our Poderopedia API must have, do and provide. For starters
1. We are working on the basic premise that a good API is one that it`s creators use for rolling out improvements and services for their own projects.
2. The API should be easy to use and must be based on proven standards already approved by the community
3. Here is the information you will be able to extract from Poderopedia
-Basic info of a person, company or institution (basic bio, avatar, industry, country, website, social media)
-News oriented profile
-Hard data sheet of connections (father of, son of, brother of, married with, owner of X company, stock owner of Y company, General Manager of Z company, Board Member of A company, Member of B private club, related to D religious group, financer of E Political Party, Government possitions, relation to think tanks, relation to non profits, etc.)
-Visualization map of those connections
-Documents related to the person, company or institution (we are using Documentcloud`s doc viewer integrated to Poderopedia)
-Sources of information
All the information above is marked with PoderVocabulary, the controlled vocabulary used as metadata by Poderopedia
Questions and comments:
1. Should we develop a full-service API or different API `s for separate services?
2. HTTP method and RESPONSE format: So far we are leaning to HTTP / GET and RESPONSE / XML and JSON. If you have comments or advice on this topics please send it to the discussion group. We can all benefit from your prior experience and knowledge
3. API`s that we like a lot (for services provided, how they are delivered and the documentation):