It's the first day of the ACS meeting in a bakingly hot Washington DC, so I decided to dip my toes into the cool waters of chemoinformatics. One of my favourite talks was by Jean-Claude Bradley, who provided an update on his work using open notebooks. Jean-Claude points out that one of the advantages of Open Data is the way that it allows quick validation of results - because there's no filtering of the data, then it's easy for anyone to scrutinize any unusual observations. What's more, if your data disagree with someone else's, then it's possible to work out what might have been done differently, potentially leading to useful new discoveries along the way.

I also quite liked his use of crowdsourcing as a way of amassing lots of data. He's currently interested in accumulating solubility data, so he's found some nice ways of getting people around the world to help provide it. For example, one undergraduate lab asks its students to generate solubility data as a practical assignment; the data is then passed on to Jean-Claude's group for his project. What better way to motivate students in their practical work than by giving them an assignment that generates useful new information?

So I came away from the session thinking anew about the whole Open Data concept. What do you think about it?

Andy

Andrew Mitchinson (Senior Editor, Nature)