Present: Graeme Stewart, Teng Jian Khoo, Andrea Valassi, Paul Laycock, Serhan Mete, Eduardo Rodrigues, Meirin Oan Evans, Kyle Knoepfel, Michel Jouvin, Benedikt Hegner, Ben Morgan, Attila Krasznahorkay, Allison Hall, Krzysztof Genser, Sudhir Malik, Liz Sexton-Kennedy, Chris Jones, David Lange, Dorothea Vom Bruch, Host Severini, Josh McFayden, Kevin Pedro, Michael Villanueva, Mike Kirby, Philippe Canal, Daniel Elvira
Apologies: Andi Salzburger, Pere Mato, Efe Yazgan
The Generators WG meeting last week discussed the review - see below for a report from the WG convenors.
Key people in Simulation met yesterday (expts., projects, HSF) and are planning a mini-workshop where experiments will present some plans (from CTDRs), boundary conditions and thoughts about R&D impacts. Should be w/b 26 April, afternoon meeting CERN time (to be doodled).
Key people in Analysis met yesterday (expts., projects, HSF) and are planning a mini-workshop where experiments will present some plans (from CTDRs), boundary conditions and thoughts about R&D impacts. Should be w/b 26 April or 3 May, afternoon meeting CERN time (to be doodled).
DOMA project has appointed two chief editors for that document, Oliver Keeble (CERN IT) and Mario Lassnig (CERN EP, Rucio).
We have asked for first document drafts by 30 June.
Students who did good assessments are now working on their proposals, have to be done by 13 April.
Season of Docs? Waiting for news.
The basic idea, following on from the discussion at the last coordination meeting, is to bring together a panel of experts with broad experience (e.g. including experts from outside HEP) to review the annotated[1] DUNE requirements and make recommendations. The panel will be chaired by the HSF Framework WG convenors who would start contacting potential panel members ASAP. Review of the requirements by the panel could start as early as possible, with a “panel meets DUNE” workshop consisting of two 1/2 days after vCHEP (late May / early June) that could then focus on the recommendations. The workshop outcome would be a public document (it is the HSF!), DUNE could then take those recommendations and make a development plan.
[1] Annotated, as the DUNE requirements are written for a DUNE audience. These annotated requirements would need to be public, intend to use googledoc for this and freeze at the time of the workshop.
Started contacting individuals from various experiments for our upcoming meeting series, attempting to survey the wider HEP & NP communities for analysis computing needs. So far the response has been encouraging, and we are trying to get specific names. Contacted so far:
We hope to start the week of April 19.
Have also been working on metadata summary document. To flesh out tech specs then invite wider community input.
We had a nice meeting yesterday:
We recommend that meetings like this be recorded, when there is lasting value in the presentation and demo.
Meeting on Spack is being planned, but date not yet settled.
Had LHCC Review 2 kickoff meeting
Phoenix people want to start holding meetings as their user base grows - created a category for that in Indico.
ATLAS had reason to use an EvtGen header in Athena, for which it turns out the licence is GPLv3, which would have its usual viral impact on the rest of the code, which is baseline Apache2.
We contacted the EvtGen authors who were not in principle averse to changing to a more liberal licence, however they are dependent on Pythia8, Photos++ and Tauola++, all of which are GPL, tying their hands.
In the course of this discussion we realised that HepMC3 is also GPL. As the main interface between the generators and the experiment codes we would like to ask that the authors change this to (at least) LGPL. This is to be followed up.
Archeology showed that HepMC2 should have been under BSD license.
FastJet authors were not very willing to change their license when we discussed with them. There has been some recent development here to add thread-safety.
Propose dual licensing to generator authors? If they want to keep GPL for everyone and allow HEP to use it non-virally nonetheless, maybe this can be formalised?
Follow-up in a Generator WG meeting.
AIDAinnova kick-off meeting, 13-15 April: https://indico.cern.ch/event/1003419/overview.