Hello everyone & welcome to the March newsletter. Get ready to be amazed by the crazy notion that there’s anything going on outside your four walls!
As of writing there are 22 £50 tickets left for the Software Sustainability Institute’s Collaborations Workshop 2021 taking place from Tuesday 30th March - Thursday 1st April.
You may just have time to enter the second annual AI Song Contest! Register a team by March 31st, with a deadline for song entries on 18th May 2021. The first contest last year invited participants to create a Eurovision style hit using artificial means, and the winning entry, Beautiful the World, did full justice to the genre.
If you’d like to help the BBSRC “better understand the computing needs of the bioscience community”, you can sign up for their workshop here, running all day on the 29th April.
The PyFR Seminar series continues with regular talks over the coming months on high-order Flux Reconstruction schemes and the PyFR flow solver. Up next is a talk by Imperial’s Dr Giorgio Giangaspero on Synthetic Turbulence Generation in PyFR, taking place on Thursday 8th April.
The Software Sustainability Institute is promoting an ELIXIR-UK survey about production, usage and handling of life science data in a research institute. If you’d like to participate, complete this 25-minute survey by 5th April.
Are RSEs naturally disinclined to blow their own trumpets, we wonder? As mentioned last month, the hidden REF has extended submissions until 14th May. Take a look at the wide range of submission categories; these include Training materials and courses, Community building, Citizen science and many more.
On Tuesday 9th March, a group of RSEs, researchers and academics came together for the Imperial/Sheffield ReproHack (reproducibility hackathon). Jointly organised by individuals from Imperial’s Research Software Community and the University of Sheffield RSE team, the 1-day workshop provided an opportunity for participants to select a paper from a directory of publications submitted by authors and to try and reproduce the results of the paper. Authors provided details of open data and software associated with their papers. The workshop also included a guest talk on “Sharing Reproducible Computational Environments with Binder” by Dr Sarah Gibson (The Alan Turing Institute).
We had a directory of 50 papers available for participants to choose from, including several submissions from authors based at Imperial. This was my first chance to participate in a ReproHack event and I found it a really interesting and valuable experience. In particular, it highlighted to me the huge challenge of ensuring long-term reproducibility of research results, even when authors have worked hard to try and ensure this, as tools, infrastructure and software stacks that research results are based on change faster than ever before. The participants were also able to provide feedback to several authors as an output of the event. It is hoped that this feedback proved useful in highlighting where authors have been successful in helping to ensure reproducibility of their research, but also in offering constructive feedback on where improvements could be made to help improve reproducibility of future research.
Thanks to all the authors who submitted papers to this workshop and to everyone who joined as a participant.
JC
Software Carpentry instructors, both trained & aspirational, the Software Sustainability Institute has set up a community hub for you.
Imperial is facilitating the redistribution of unwanted laptops to those in need, including an amnesty for that ancient college one you’ve had under your desk for years. I got a bit tangled up trying to thusly divest myself, because there seem to be two registration mechanisms in place. One is for drop-off at White City, but if that doesn’t suit ICT also have an ASK category via which you can drop off in Sherfield. There’s also (allegedly) a courier service if getting to South Kensington is not an option. ICT didn’t respond in person to my ASK ticket, but I dropped off the laptop anyway on Level 4 at the datetime I’d chosen, and it was received at the door.
Imperial has positions open for an [RSE Team Lead and Senior RSE (in ICT) and an RSE (R Developer)] (in the School of Public Health). Closing date for the first two jobs is 09/04/21, while the RSE (R Developer) closes on 14/04/21.
Should you be thinking of renaming your Github “master” branches, to “main” in line with Github’s own policy, or indeed any other name, here’s a helpful page on how to do it.
A blog post by Dave Horsfall from Newcastle University’s RSE team on Mental wellbeing in academia: Staying healthy as a research software engineer.
A short blog post and link to a related paper, recently published in PLoS Computational Biology, Using prototyping to select software for a research software project
Visualise your GitHub story in 3D. Mine looked more like a Toblerone than a Skyline, alas. But then my extensive body of work is on github.ic.ac.uk. At least that’s what I tell myself.
This podcast in the Code for Thought series is on the effect of “télétravail” on team structure.
Server-room enthusiasts, line up for some slo-mo drama with this time-lapse of ARCHER2 being installed! You can also see the moment its predecessor was switched off and packed away, although I do wonder if letting the world know it’s just a matter of typing /home/crayadm/bin/stop_archer was a good idea.
This month’s pick is OpenSlide. A recent piece of work I had to do involved processing some slide images to make them available in manageable chunks to a neural network. Virtual slides are large, high resolution (often multi-resolution) images used in digital pathology, and reading them using standard image tools or libraries is a challenge because of their size, and because there’s no universal data format so each vendor implements its own, typically undocumented, formats, libraries, and viewers.
My particular vendor (Hamamatsu) produces files in a format called NDPI, and after struggling for a bit to tile them up with a bespoke command line utility I discovered OpenSlide and its Python binding (its core is C; there’s a Java binding too). The API is very simple; there’s broadly speaking one method that reads a region of the image at the desired position and resolution and makes it available as a PIL image (in the Python case) for further processing.
RS Community coffee continues weekly on Teams - normally on Friday afternoons at 3pm but check our Slack workspace for exact times and connection details.
The Slack workspace also features an #OpenFOAM channel where regular code review sessions are announced (amongst other CFD-related discussions…)
Drop us a line with anything you’d like included in the newsletter, ideas about how it could be improved… or even offer to guest-edit a future edition! rse-committee@imperial.ac.uk.
If you’re reading this on the web and would like to receive the next newsletter directly to your inbox then please subscribe to our RSE Community Mailing List here.
This issue of the Research Software Community Newsletter was edited by Jazz Mack Smith. All previous newsletters are available in our online archive.