Attending
Present/Contributing: Graeme Stewart, Benedikt Hegner, Ben Morgan, Efe Yazgan, Paul Laycock, Eduardo Rodrigues, Wouter Deconinck, Sudhir Malik, Valentin Volkl, Michael Hernendez
Apologies/Contributing: Markus Diefenthaler, Josh McFayden, Nicole Skidmore
News, general matters, announcements
IRIS-HEP/HSF Joint meeting on Software Citation and Recognition
Workshop happened 22-23 November, with a wide range of participants from software projects, experiments, INSPIRE, Zenodo, journals. We covered a lot of topics, including having breakout sessions that touched on a few key points:
- A history of work and development in this area (see Dan’s opening talk)
- A lot of good background to get to the current “state-of-the-art”
- What purpose do we foresee for software citations - it is for recognition? reproducibility? These are different things.
- What should we be citing? Academic articles? Software DOIs? Something in-between, like JOSS papers?
- Should follow the recommendations of the software’s authors, usually
- Distinction between conference proceedings and journal articles is made in HEP (unlike CS); even refereed proceedings seem under-weighted
- Some feedback from RSEs (not specifically HEP) is that they do not want to write journal articles!
- Experiments do have editorial guidance on what to cite, which varies from experiment to experiment
- There would be value in having a set of overarching guidelines (e.g., curated by the HSF); which also gives software package authors a ~single point of contact
- Software DOIs are not really indexed at the moment, but they will be in the next round of INSPIRE developments
- In an automated fashion only
- The mechanics of providing information on software citation has improved a lot (
CITATION.cff files; Zenodo release harvesting, with authorship)
- Can foresee some improvements, e.g., better handling of the Concept (all versions) DOI in Zenodo
There will be a report developed in the next few months (see final section of Live Notes for skeleton). The organisers also will submit a CHEP abstract.
Sudhir - important to also invite people who make the decisions in the experiments, e.g., Pub Com and Spokespersons.
Graeme - we did not bad for Pub Com Chairs. Generally got a good feeling on the openness of experiments to improve in this area.
HSF Talk at SMARTHEP network kick-off
Benedikt gave a talk at this event - thank you to Caterina for the invitation.
Working Group Updates
General
We now have a good number of nominations; however, we are awaiting some final nominations from the experiments this week.
Search committee will meet next week.
In January we will have a review of the year gone, as well as looking forward to 2023 - so please think about it.
Meetings
Reminder: Please try and book meetings in Indico at least 2 weeks in advance!
That way they go into the calendar early and they will be included in the weekly email announcement that goes to HSF Forum.
Data Analysis
- Abstract for a talk on experiment training and on-boarding submitted to CHEP - this talk will accompany the paper that was discussed last time. All paper contributors are added as talk authors
- Still need to establish a timeline for the paper given everyone’s busy work-loads
Detector Simulation
- Next meeting on 28th November on Differentiable Computing for Simulation
- Last meeting of the year on 12th December covering Geant4 Physics Model Parameter variation
- Anticipate Geant4 Technical Forum in early December covering latest 11.1 release
Reconstruction and Software Trigger
PyHEP
- Next topical meeting is taking place Wednesday December 7th. The tool RootInteractive, from ALICE colleagues, will be presented.
- We plan to organise in 2023 some of the meetings jointly with IRIS-HEP, given the regular overap in terms of topics presented and meeting attendees.
Benedikt - is the Python 3.11 speed-up “real” in HEP use cases? See talk from Henry at PyHEP. Caution - this is speed-up of CPython, would not necessarily be reflected in code using pandas, numpy heavily.
Event Generators
HSF co-organized with CTEQ, EICUserGroup, and MCnet MC4EIC, a workshop on event simulation for the EIC. The workshop covered:
- In-depth reports on the precision of foreseen measurements and the related MC event generator (MCEG) needs: Inclusive deep-inelastic scattering (DIS), semi-inclusive DIS, exclusive processes, jets and heavy flavor, BSM.
- Status of MCEG projects and the thrust of future R&D: Herwig, Phythia, Sherpa, BeAGLE, CASCADE, eHIJING, ePIC, eSTARlight, MadGraph5 aMC@NLO, MLEG, and SARTRE.
- Connection between formal QCD theory and its implementation in MCEGs:
- Is there a QCD-based computational framework that allows easy implementation of new theoretical developments, while also making MCEGs predictive and versatile enough to be used successfully by experiments?
- How can we make use of existing expertise, experience and technology to advance the construction of MCEGs for the EIC?
- What measurements will be essential for tuning the MCEGs?
A workshop report is in preparation.
We are still working on a meeting in December and aim for a tuning tutorial in February.
Software Training
- Upcoming data preservation training week: Jan 16 - 20, 2023, on topics of docker, singularity, CI/CD etc.
- modified format: self-training with videos followed by discussion by experiment experts
- Data preservation hackathon: Wed 7 Dec
- Attended EuSSI Training Bazaar where similar challenges were discussed (instructor pool, retention and attrition of participants)
- Submitting multiple CHEP abstracts on sustainability of training efforts.
C++ Course and Hands-on Training
- (Sudhir on behalf of C++ team)- 5th HEP C++ Course, post-training discussion (Advance the HEP C++ Course - fixed monthly meeting) https://indico.cern.ch/event/1203350/
- How many people have been trained to date?
- Registration has been 75-100 for each course, so 375-500, but there is attrition wrt. original registration numbers
Other Interest and Activity Areas
Compute Accelerator Forum
- 2022 Calendar
- 2023 Calendar
- Do get in touch with Graeme, Ben, Stefan if you have an idea or a topic to present
Software and Computing Roundtable
The Software and Computing Roundtable restarts on Dec. 13 with an update on EIC Software. There will be presentations on the EIC Software: Statement of Principles and the software stack. The “Year in Review” event with contributions from the HEP Software Foundation, Brookhaven National Laboratory, and Jefferson Lab will be on Jan. 17.
AOB
HSF Calendar
We started filling in the calendar for 2023.
Should now be done!
Next Meeting
Our next coordination meeting is 8 December - this will be the last one of 2022.
Meetings for 2023 are now booked in Indico. We follow the tradition of meetings on the odd weeks of the year.
The first meeting next year is January 19, where we will start by reviewing 2022 and planning for 2023.