The other week I was at a gathering of developers, librarians and researchers with an interest in institutional data repositories. Amongst other things, we spent some time brainstorming the requirements for such a repository, but there was one minor-sounding one that caught my imagination.
It boiled down to this question: given only the DOI for a published article (or other artefact), how do you find an open access copy archived in an institutional repository? Some (rather cursory) Googling didn’t come up with an obvious solution, so I thought “How hard can it be to implement?”.
All that’s required is a database mapping DOIs onto URLs, and a spot of glue to make it accessible over the web. The data that you need is freely available in machine-readable format from most repositories via OAI-PMH, so you can fill up the database using that as a data source.
So, without further ado here it is:
A few caveats:
- I don’t get much chance to write code at work at the moment, so this was an opportunity to exercise under-used brain muscles and learn some new stuff. It could probably be done better (and the source code is on github, so feel free to fork it and add improvements). It’s written in Ruby using the awesome Sinatra web framework.
- It’s currently hosted on Heroku’s free starter-level service, so there’s very little capacity. It therefore only includes DOI’s from the University of Bath’s Opus repository, and the database is full.
Go try it out and let me know what you think. If it’s useful, I’ll look into how I can make it more robust and resolve more DOIs.