MIT has an interesting project underway that has a promising potential to revolutionize the way we utilize information systems.
There has always existed a large gap between computer programmers and the people who actually need to work with the information the computers produce. Programmers often have no idea what the end user really needs – they are more interested in creating wonderful things. End users often are unable to articulate, or even know what it is they really want, other than in the most vague and often contradictory terms.
Hugo Liu and Henry Lieberman at the MIT Media Lab are working on a project called Metafor that might one day allow end users to get what they need without having to bother with the surly and mean programmers, and allow the programmers to focus on what they are passionate about rather than having to listen to the ramblings of an undisciplined mind.
Actually, Metafor is a long way from that. Currently, it just creates a program “scaffolding”. But it’s a very good start.
A predecessor project, MontyLingua, is a free little thang that goes a long way, taking normal sentences as input and outputting a semantic intpretation the text that can be used for such things as… well… data retrieval or processing.
Interesting pages he has, too – worth a visit. Mr. Liu seems to be a smart cookie. One of the hyper-rational/intelligent sorts that is in a constant state of “balancing” it with the un-sensible, or perhaps, aesthetic. He’ll find his balance. 😉
You can view a little animation that shows what Metafor is currently doing.
There are also some publications on it (in PDF format):
[view] (2005) Programmatic Semantics for Natural Language Interfaces. Proceedings of the ACM Conference on Human Factors in Computing Systems
[view] (2004) Toward a Programmatic Semantics of Natural Language. Proceedings of VL/HCC’04: the 20th IEEE Symposium on Visual Languages and Human-Centric Computing
And, in a MS Word format – a document that shows much of their original thinking that went into Metafor. It’s certainly the best at revealing the inherent challenges…