MIT video on Open Access
MITworld has posted a (76 minute) video of a seminar on Open Access. The speakers are John Wilbanks, the vice president of ScienceCommons; and Anna Gould, the head of the Libraries for Science & Engineering at MIT.
I think that Wilbanks is more or less dead on in his description of the issues in finding scientific data online today — the Google Search doesn’t do the job for finding data. You will find a pile of papers that may have the data you’re looking for buried within, but you really just want the extracted bit of information.
He goes on to point out that the copyrights enforced by the journal electronic subscriptions explicitly forbid actually doing any sort of digital indexing of the content. In other words you aren’t allowed to set up a search function based on journal text, let alone set up databases to extract data and allow for new combinations and cross-referencing.
He goes on to talk about some really exciting ideas. Of course a fair bit of time is spent on the Creative Commons-based copyrights for scholarly publications, which I am all for. He also describes attempting to set up a “transactional system” for repositories of scientific materials.
You don’t email Stallman to get the Gnu code
To me, this would fundamentally alter the practice of science. Most of science takes place in isolated labs, with perhaps some limited academic discussions between similar workers and maybe the occasional plasmid sent from one lab to another. The Science Commons project iBridge, however, streamlines the process and sets up something that looks almost like a store for scientific materials. Presumably these requests would go to the central repositories who would fulfill the request. I haven’t gotten a chance to investigate iBridge very much yet, but I’ll try to follow up on this in a later post.
He finishes up by talking about generalizing code to automate some scientific queries that are currently repetitive and manual tasks. It might sound dull, but he demonstrates some relatively simple code that can do data extraction from open access data. As the other tools of the Science Commons are developed, this will become even more powerful and will allow for much more complex and interesting data extraction, indexing, and recombination.
Gould goes on to focus more specifically things that people are doing today to develop new ways to look at scientific data. One in particular that I like is SciVee, where the paper authors can submit a video highlighting the results of their papers. She also mentions sort of a research library “union” being developed at CERN, called SCOAP3.
Overall it’s a very engrossing talk, and has sort of set my head spinning with ideas. If you get time, I’d recommend watching it (although at over an hour I understand that finding the time might be tough).
via Open Access News
March 26th, 2008 at 11:27 am
[...] few weeks ago, I saw (and blogged about) a video from a talk at MIT on Open Access. In it the president of Science Commons, John [...]