There's really a need though for more developers to get involved with building tools for more easily searching and working with the collection, ideally with a nice UI and integration with things like crossref. This is a massively valuable data set and it would be great to see what people can come up with. Lots of awesome potential for data mining too.
If it weren't for the legal issues (publishers using copyright law to restrict access to literature they got for free since they never pay authors for their work), there's no shortage of projects that could utilize this data and be enormously beneficial for the scientific community and humanity in general. Unfortunately such work can only be done in the shadows right now, which greatly limits the number of people/institutions likely to do so.