iSpecies is a test of E O Wilson's idea of a web page for each species. The data displayed are generated "on the fly" by querying other data sources
This news via Lost Boy (a.k.a. Leigh Dodds, who works at science publisher Ingenta). The iSpecies search engine was written by Rod Page (who wrote the LSID plugin for Firefox, see here). What is interesting about this "mash-up" is how the web pages are produced. Distributed web services such as Yahoo Image search, NCBI's EUtils, and Google Scholar (a scraper in this case) are used in creating each page, for example lion.
What is also cool is that each species has RDF formated metadata associated with it via an LSID, see here for an example. It would be nice if each species had its own permanent URL, which would be arguably more useful than an LSID, but I won't go there :)
Expect to see more of distributed web services integration in bioinformatics, as Lincoln predicted, everyone ignored and will now think is great !
Comments
Mash-up in bioinformatics
You are right, I do think it is great :). I wonder why is it that ideas that are produced in the tech sector take so long to be used in bioinformatics. It must have something to do with critical mass. There is just not enough people working on these things although the life sciences is clearly a space where the technology is needed. What other services would we like to have APIs for ? Would it be fun to have a customizable portal with bioinformatics services ? Copy-cat all these AJAX desktop pages popping up in the tech sector with some useful bioinformatics tools. You want to give that a try Alf ? :)
make a list on the wiki
Want to make a list of bioinformatics services with web APIs?
Added some web services....
I've added some web services to that Wiki page. The EBI has quite a wide selection but they're not very well documented...
nodalpoint project contender?
Now there's something we could work on as a virtual collaboration...
modular base for experimenting with web APIs
I had a go at this today: http://alf.hubmed.org/cgi-bin/binf.cgi
At the moment, there's just one example - a keyword search of the PDB that returns a list of matching IDs.
The script behind it is modular though, so anyone else can write modules to add to the page. Here's a basic example - send any others by email and I'll add them in.
Plugins needed
More information here