LifeWatch project reports back in Amsterdam

Yesterday I attended the LifeWatch meeting at the University of Amsterdam. As my own office is based just across the road, this is possibly the shortest work related trip I’ve ever had to make – I’m still working out the travel claim for the shoe leather used.

The LifeWatch gathering marked the close of the preparatory stage of the project, and reported back on some of the successes achieved. Alex Hardisty of Cardiff University summarised LifeWatch’s aim – to build a distributed research infrastructure for the biodiversity community. According to Alex, setting this up is a bit like running a railway in the UK. The infrastructure doesn’t buy its own rolling stock or run the franchise but it does keep the track running and take responsibility for safety.

LifeWatch will be gathering, managing, merging and manipulating data from diverse applications across multiple scales. The infrastructure itself will be based on an open access, single portal. It will also be a community driven e-Infrastructure, consisting of centres distributed across countries offering services to users. This is a well-known model, used for example by EGI, but more novel is perhaps its bottom-up approach to governance of community developments, similar to the Apache open source model.

The biodiversity community has a “jigsaw” of requirements to get to grips with but are focusing on five main challenges - heterogeneity of data sources, the gap between current practice and future vision, the scale of the user community, the fast pace of change in ICT and how LifeWatch will fit with mainstream practice. Solutions to these challenges will evolve from using a service oriented architecture, establishing workflow paradigms, thinking globally while acting locally, using reference models and cookbooks and having a technical construction plan.

More about the applications later…