EIC Software: Lessons Learned II

US/Eastern
Description

In the EIC Software meeting on January 26, software experts from the ATHENA and ECCE proto-collaborations shared their experience and presented "Lessons Learned" from the simulation efforts for the detector collaboration proposals. In this meeting, we will summarize the lessons learned and start with the wider community the discussion about a possible path forward for EIC Software: 

We will use Zoom for the remote meeting: 

EIC Software Meeting on March 23 

45 participants 

Summary of "Lessons Learned” 

  • Maxim - pay attention to data & analysis preservation from the beginning. Important that the new collaboration define policies and embed practices for DAP from the beginning. Exactly what they are matters much less than that they exist. Particular tool choices don’t matter so much, there is a lot of commonality between them. e.g. am noticing that REANA and github actions have a lot of functional overlap.
  • Wouter - excellent balanced overview and summary, thank you Markus & EICUG SWG.
  • Sylvester - ditto, and I agree with Maxim’s important comment.
  • Wouter - interested in what ECCE has done re: DAP via centralized workflows. How is it different from other approaches? Maxim - not very different. Important point is having formalized and reproducible description of workflows, less important at this point how they’re actually implemented. 
  • David - also thank you, very nice summary. We should make sure there are multiple leadership roles in the collaboration that have to do with S&C, engage a number of people and their expertise and perspectives, and help advance S&C careers. We in this meeting should be thinking about what would be a reasonable org chart in S&C. Come up with concrete recommendations for the management.
    • Agreement that this is an excellent point.
  • Joe - re: David’s org chart comment, it will have the added benefit of identifying the group of key people, core developers, on whom we really depend, understand where single points of expertise/failure are, and so enable us to protect against loss of people, understand what has to be covered.

Statement of Software Principles

  • Jin - procedurally how will we consolidate diverse views into a common position? Discuss specific points/topics in later meetings? 
    • Markus - Yes that’s the idea. And start the discussion now with comments in the googledoc. Will help to narrow down the discussion to the important points. Or write your own version of the draft! Aim is to converge on something that presents our collective view as a software community, which we can collectively sign, then go to the wider community for broader input. 
    • Jin - OK makes sense. Collecting diverse opinions is very useful.

AIWG kickoff meeting next week

There are minutes attached to this event. Show them.
    • 11:00 11:55
      Lessons Learned II
      • 11:00
        Reflections on “Lessons Learned” From Detector Collaboration Collaborations 25m
        Speakers: Dr Markus Diefenthaler (Jefferson Lab), Torre Wenaus (BNL)
      • 11:25
        Discussion 10m
      • 11:35
        Statement of Software Principles 10m
        Speakers: Dr Markus Diefenthaler (Jefferson Lab), Torre Wenaus (BNL)
      • 11:45
        Discussion 10m
    • 11:55 12:00
      Announcement: Kick-off Meeting of the AIWG