Great to see your comments. The blog is fairly new, so I’m still feeling my way about what I want to include here — think of it an experiment of sorts. I’m generally amazed by the amount of traffic it receives — my molecule pictures are quite popular, thanks to Google.

I’ll definitely investigate Taverna more closely. Personally, I have several workflows designed for the UNIX environment which make my life easier when dealing with 100s to 1000s of computational results. But it’s nice to see the beginnings of more general approaches.

And you’re flattering me about “precisely defined” function of Open Babel. We’re getting there, but any large codebase will have countless bugs. Prof. Donald Knuth, a famous CS professor, says “I can always take a program and cut out at least one line of code and fix at least one bug.” I think the exciting part about “Blue Obelisk” is that there are multiple open, reliable chemistry codebases for integrating all sorts of different chemical problems. Each has a niche in terms of project goals and use, but there’s a large amount of sharing of code, ideas, and support.

Have just mentioned the Blue Obelisk on the taverna list. Taverna is a workflow tool and althohough originally developed for bioinformatics is completely general. We see it as an important tool for uniting the sort of approach that the Blue Obelisk represents. it’s a unfying technology as it is paltform and OS dependent and allows people to integrate their applications without worrying about languages.

openbabel has a prominent place in our applications. It’s an excellent example of a tool - precisely defined function, inputs and outputs. This is how chemoinformatics should be built.

P.

Peter Murray-Rust and Henry Rzepa
University of Cambridge and Imperial College

ACS, 2005-03-14, San Diego

The Semantic Web is becoming a reality
…