This weekend, I got the chance to attend SciBarCamb in Cambridge, and thought I would write briefly about the experience. There were a number of very interesting talks and I’ve written about a few of them below.

–
In Ian Mulvany‘s talk we discussed the limits of knowledge and whether we can build a computer which could accurately represent anything in existence, something which could pass the Turing test, a computer which could build other computers etc etc. My immediate thoughts went to things that we believe to be impossible to replicate / predict in general, e.g. chaos theory – weather systems etc, and if these things hold true, then can we really build such a system ?

We went on to talk about whether or not we are losing knowledge through abstraction of information. For example, most people don’t understand how a smartphone works. Does this reduce what we know or what we will know as a society, or does it enable us to advance at a greater pace? This could have been a day conference in itself to be honest!

–
Another interesting chat was in relation to data producers and data consumers. A number of people spoke about the lack of information flow within the scientific community at the moment and how to improve upon that. One point which was interesting, was the lack of options for exposing your research data in conjunction with a particular publication. It was pointed out byCameron Neylon that Data Dryad was one such service.

Apart from a lack of services, it seems that scientists are quite conservative about exposing their data, and that these tools won’t take off until this attitude changes. The attitude was summed up by Ian Mulvany’s comment: “I hope in my lifetime to be at a data in science discussion where the topic is favourite sharing tools not whether they should”

–
This chat connected naturally to Matt Wood‘s demo of a tool which he has developed called Lark (not lark.com as it happens, which is a “silent waking system for couples” !). The tool was similar to a project management system for your data. The most interesting function was that the data could be replicated / forked by other users in a Github-like fashion, but more importantly, that in the event of a problem in the data, all users who replicated the data would automatically be notified. Pretty interesting.

All in all, it was a great event, and congrats to the organisers for a job well done. I’d definitely go again if only to try and steal some brain power from the incredibly smart people who attended :)