I'm on my way back from a great week in England. I spent most of the week in Norwich at The Genome Analysis Center (t-gaaaaaaaack), hosted by Vicky Schneider-Gricar. I gave a talk, taught two workshops together with Aleksandra Pawlik -- one for biologists and one for bioinformaticians -- and met quite a few people, including Sergey Lamzin, Bernardo Clavijo, Dan MacLean, and Mario Caccamo. Matt Drew and Tanya Dickie went above and beyond in helping organize things, too!

It was the usual poster-trip for science: good food, great company, wonderful discussions, and lots of fun. I also got to spend a day in London wandering about, which was really the first time I've explored London on my own; while there, I even serendipitously met up with The Kaitlin Thaney of Mozilla Science Labs.

One of the highlights of the visit was the chance to participate in an ELIXIR meeting, hosted by TGAC. ELIXIR is a European bioinformatics organization, with part of their mission focused on training. Several of the ELIXIR attendees also stuck around for a GOBLET meeting; GOBLET stands for the Global Organization for Learning, Education, and Training, and is in no way connected to Harry Potter and the Goblet of Fire. I had to miss the GOBLET meeting but I got a chance to meet many of the attendees.

The ELIXIR meeting (which I did attend) was all about how to coordinate on training and material development. ELIXIR is not, alas, a global organization, but there were people attending from Australia, Israel, and North America to help connect to those communities-- I got to represent all of North America, including Software Carpentry :).

The main thing that struck me is how little coordination already exists. We talked about some issues, like undergraduate and graduate programs, where I have little knowledge or primary interest, but we also talked about issues like common material development, sharing data sets that are good for training, reusable or syndicatable training materials, train-the-trainer initiatives, a map between standard needs in training and training materials, shared training methods, assessment, etc. Somewhat oddly for a group of hi-tech folks, there was very little social media "presence" and little awareness of resources like biostar.

One personal take-away was this: the bioinformatics post-grad education field (think: workshops & "remedial" training in data analysis) is ripe for some sharing, and I'm planning to participate in and/or run some materials and technology development workshops on this subject. (This was one of my conclusions from our SESYNC workshop on what to teach biologists about computing, so not too surprising!)

Another take-away was: I should sign up for GOBLET and participate in their development of a global network of trainers and trainers-who-train-trainers. Maybe we can get some good social media connections going there, too, and start to draw people in; we need active researchers to participate in training more than they do, and GOBLET strikes me as something that could help pull them in.

And, finally, there's something I hope to do more of: provide opinionated suggestions for "good enough" software and training approaches, to help guide people through the existing melange. For example, Software Carpentry chose Python as a good-enough language; we've bringing IPython Notebook to people's attention; and Greg Wilson has been pioneering an online trainer-training course that sounds really good. These kinds of "suggested practice" are becoming increasingly important as software and training expands and diversified, and bioinformatics training needs this.

--titus