1st conference of Research Software Engineers
Update: Slides from many of the talks are now available via the conference website.
Software engineers – twisted sense of humour #rse16 #forgotgitadd pic.twitter.com/XpqqqWLeKJ
— Tim Nicholls (@timcnicholls) September 15, 2016
The first conference of Research Software Engineers (RSEs) was held by the UK RSE Association at the Museum of Science and Industry in Manchester on 15-16 September. Billed as the first conference “to focus exclusively on the issues that affect people who write and use software in research”, it drew over 200 attendees from all over the world.
Incidentally, as noted by Simon Hettrick in his opening remarks, one should always refer, especially in writing, to ‘An RSE’ and not the alternative.
Caroline Jay’s talk “Not everyone can use git: Research Software Engineers’ recommendations for scientist-centred software support (and what researchers think of them)” was a highlight. She reported on a survey of researchers at Manchester, looking at why they do and don’t publish software, what might encourage them to do so, and related issues. Among other things, she reported that 76% responded positively to the question “Would having access to RSEs on campus be something that would help scientists publish their software?”. The fundamental role of software in much of modern research was emphasised.
Software can exist without a paper but a paper and the results can't. Wrong software = wrong science. #RSE16 pic.twitter.com/zOdpNgjHrU
— Karthik Ram (@_inundata) September 15, 2016
The results from the survey are available on GitHub.
There were a wide range of parallel sessions, covering specialist and more general technologies, “How to be a happy RSE”, and understanding the research software landscape and the role of RSEs, and more. Among the technical sessions were no less than three on Docker, which were very popular. A technology for packaging up software and the environment in which it runs, there’s been quite a bit of buzz about it over the past few years and it is particular value when it comes to making research reproducible. The material from one of the sessions, by Mark Fernandes, is available on GitHub.
The RSE Association AGM saw the introduction of a new committee; if you want to get involved in the discussions going on in the RSE community or even to influence the work of the association, check out their website.
Conversations at the conference highlighted that within institutions, RSEs are organised in a number of different ways. Many are embedded within research groups, others are part of a dedicated RSE team either within a central service (as with the Research Computing team here) or an academic department (e.g. the Digital Institute at Newcastle). The Research Computing network we’ve been building here is far from unique – see the Research Software Developers’ Network at Oxford, for example.
The next conference has already been pencilled in for the same venue on 6-7 September 2017.
There was much more going on at the conference than has been covered here. Look out for upcoming papers in the Journal of Open Research Software to expand on many of the talks and check out #RSE16 on Twitter for more. The RSE Association have gathered some of the many tweets from the event in Storify.