With all the recent furore about academic publishing houses, I thought that PeterM’s blog post The scandal of publisher-forbidden textmining: The vision denied was just going to be another in a long line of rants about the publishers. How wrong could I be!

There in paragraph 6, 7 and 8 is something that (to me) is frankly amazing: someone has written a system that can parse natural text and turn it into the underlying chemical research. Now obviously, this is not strictly natural language as there’s likely a very stylised way of describing these things, but still very impressive.

Maybe the US governments wish to mine Facebook won’t fail after all.