One of the great aspects of Web 2.0 is the availability of numerous APIs that are attracting both professional and hobbyist programmers to build cool new applications. The mashup has been borrowed from Hip-Hop culture and re-envisioned as a combination services and data from multiple locations online. Do you care about the modern views by location on 17th century poetry? You can cross-reference your collection of poems with the one of the news archive APIs and visualize the results on a Google Map.
The flip side to this is that each of these APIs started with a person who saw the benefit to letting data be available and re-usable. API creation is a daunting task, but it can be made easier. The Walden’s Paths project at Texas A&M, which I am currently the lead designer on, has found that by coupling modern database access techniques and the RESTlet library for API creation, we can easily produce APIs that can be successfully used for creation of interesting interfaces.
I propose a hack-a-thon where we would discuss as a group API design, issues to be concerned of when exposing your data, and then put together simple APIs that would allow easy data access. This might be even more useful when combined with others who have knowledge of creating mashups so we can quickly see what an open-API allows us.
#1 by neal on March 28, 2011 - 9:16 pm
This sounds like an interesting topic. API’s have a lot of potential for leveraging shared data sources, for defining a shared base of content and services that scholars can then repurpose and extend (possibly with some technical help, possibly by learning to do it on their own) to present their own analysis and interpretation without having to create the entire project from scratch.
Perhaps we could spend some time talking about the design principles for creating RESTful APIs as well as the some of the benefits/drawbacks to these approaches (e.g., what are the sustainability implications if I create a project that depends on someone else’s API).
For the hack-a-thon, it might be useful to have one or two project domains in mind. I’d be interested in walking through a general purpose API for representing digital facsimiles. I think that might be something that is broad enough that we can get both technical and non-technical people working on and yet (hopefully) have something cobbled together in an hour or so.
I’m sure that there are other interesting projects out there as well. Are there other types of resources that people would be interested in hacking up?