Attending
Present/Contributing: Graeme Stewart, Giulia Casarosa, Phil Ilten, Liz Sexton-Kennedy, Claire Antel, Torre Wenaus, Pere Mato, Valentin Volkl, Efe Yazgan, Kyle Knoepfel, Nick Smith, Kilian Lieret, Ale Di Girolamo, Alex Held, Juraj Smiesko, Nick Styles, Wouter Deconinck, Zach Marshall
Apologies/Contributing: Krzysztof Genser, Eduardo Rodrigues, Bendikt Hegner, Caterina Doglioni
News, general matters, announcements
LHC Experiment Feedback
We got some LHC experiment feedback - see presentation from the last Architects Forum, slide 13:
- A useful list of activities foreseen to be discussed during in 2023
- No committed effort to the various activities (by construction!)
- Nice to have events to spread new and technical insights (Computer Accelerator Forum, Frameworks WG, SW Tools and Packaging WG)
- Perhaps to be held more regularly?
- HSF: Foundation or useful Forum?
- E.g. it gathers together people from various communities to share ideas and discuss possible synergies, for example in the areas of trainings (e.g. C++).
- Nick Styles: HSF is useful, but ATLAS concerns that WG interests not well aligned with experiments
- Valid concern, though HSF is wider than LHC experiments (feedback on important topics always welcomed)
- Pere Mato: feedback was via SFT and we wanted to know if the SFT effort that goes to the HSF is thought to be well used
- Nick Styles - is the HSF really a foundation? No assignable resources and activities are best-effort
- True that we had more foundation-like ideas in the early days (modeled on the Apache Software Foundation)
- While effort is ‘best-effort’, people contribute to HSF to enhance things they would be hoping to do anyway
- So little risk that activities go away
- Albeit that there is limited capacity to ask the HSF for things
- At the moment not really keen to change the name, even if ‘forum’ is more what we do
- We have an identity well established
- We don’t want to preclude different feature activities
- We should explain better what we do (see website revamp ideas below)
Pre-CHEP Workshop
This is pretty much all ready to go now. See indico for final agenda. The Zoom link will be created soon, should be ok for presentations, but still it’s YMMV for discussions.
We plan to circulate the AFF report and a list of topical questions in advance to that we know what we are trying to achieve.
Google Summer of Code, 2023
- Preparing proposal evaluation overview
- 66 proposals for ~39 projects received, for ranking by April 27
Google Season of Docs, 2023
- CVMFS proposal submitted, but it was not accepted by Google (and no feedback given)
Working Group Updates
General
The PR for the planning summary of 2023 is here: https://github.com/HSF/hsf.github.io/pull/1410.
Please comment on the PR to add clarifications or additional information. (Thanks to those who already did - also PRs fine.)
A lot of the WGs have had slow starts this year (see also LHC comments above), so it becomes a bit urgent to plan pre-summer activities, so that people know we are still alive.
Software Training
- Hackathon next week, Monday April 17, 16h-18h, indico (todos following last analysis preservation workshop, taking inventory of ML material)
- Possibly two students with ~1 month FTE to work on pytest module to be integrated with analysis preservation workshop and brushing up the REANA training (still looking for mentor with more REANA experience)
- Lively discussion regarding the format of our training:
- The Carpentries are switching to a new framework “Workbench” with a new static page generator built in R. This would require some porting effort without clear benefits to us. Example in new format: shell-novice
- At the same time, Jupyterbook is a static page rendering with much more familiar technologies (for a HEP person) and has been used for training in our and the RSE field.
Just to note - Peter Elmer gave a very nice talk yesterday on workforce development at a US HEP planning (P5) meeting at BNL (with the HSF figuring prominently in his talk of course)
C++ Course and Hands-on Training
C++ course in the UK:
- Dates: August 28-September 1st, lunchtime to lunchtime
- Costs: TBC (would have been 500 GBP with accommodation, lunch and breakfast + 1 social dinner but see below)
- Structure
- 3 days of lectures + 2 days of hands-on / team work (will coordinate with training team)
- seminar by the Software Sustainability Institute https://www.software.ac.uk
- Teachers/demonstrators:
- Day 1, Day 2: Sebastian Ponce
- Day 3: Caterina Doglioni (or other volunteers welcome!)
- Looking for demonstrators (some available locally)
- Swift-HEP has some budget to cover costs
- Logistics:
- since the Monday is a holiday we can’t have lodging at the university
- hotel rooms will be pre-booked
- Next steps: finishing up and sending around “interest” questionnaire via indico
Detector Simulation
Reconstruction and Software Trigger
Kick-off meeting scheduled for 26/04/2023 @ 2pm CEST
- Created indico agenda: https://indico.cern.ch/event/1273894/
- Introduction of the WG
- Discussion on SW reco & trigger between different experiments (slides)
- will present slides with some aspects in common/different among “our” experiments (accelerator/beams, trigger, reconstruction)
- Plans and topics for 2023, Google doc to collect ideas here, reduced our proposal to 3 topics:
- Sustainability in SW development for HEP
- Lessons learned from recent trigger upgrades and new data-taking campaigns
- Biased triggering & missed physics
- Planning to circulate email in the following days - share with HSF coord for x-check beforehand?
- Weekly meet-ups until kick-off meeting.
PyHEP
- Organisation of PyHEP.dev 2023 in-person workshop progressing well.
- Set of participants essentially finalised.
- Now focusing on the preparation of the agenda with short kick-off presentations in the mornings.
- Grant proposal submitted to the PSF.
Event Generators
- Tuning tutorial - started contacting potential contributors and trying to fix the date(s).
- Graeme: Updates from Swift-HEP project would be well worth presenting (see Christian Gutschow’s talk)
Other Interest and Activity Areas
Compute Accelerator Forum
Next meeting is 26 April, https://indico.cern.ch/event/1264290/:
- CADNA, tool for studying numerical precision (Fabienne Jézéquel, Sorbonne)
- The case for AI-augmented facilities (Michael Bussmann, HZDR)
AOB
Website
Benedikt started an issue in GitHub to gather ideas about reorganising and revamping the website to better reflect our areas of actual activity.
Next Meeting
The next meeting will be on 27 April.
Please sign up for chairing this or one of the future coordination meetings.