Minutes of February 6, 2012 Physics Coordination Meeting

(Link to the Meeting Agenda)

News (CGL)

Higgs papers (p3)

The Higgs combination paper (HIG-11-032) will be submitted to PLB tomorrow (a similar paper is being submitted by ATLAS). Hopefully, we will submit the rest of the papers (9) this week. CERN will release a short public statement.

  • Q: What is going to be presented tomorrow? When can the supporting note be posted?
  • A: We can put the document on CDS immediately.

Object identification: reviews and central twiki (p4)

  • Object reviews : three reviews in the past week: Tau, Muon and BTV. Overall, a wide success, largely due to PAG object experts.
  • Approved object identification: central object ID twiki
    • The list of approved object ID is now linked from the main physics twiki page. Cross linked from PPD group(?)
    • The idea is to centralize information about object ID, efficiencies & fake rates, resolution and systematic uncertainties, with maximum clarity.
    • May be more adequate for some groups than for others and we will need feedback on the form, but the idea is not to have numbers on this ‘top level’ twiki, but rather to gather links to more detailed and relevant POG twikis.
    • Hope to be filled within 1 week.
    • Comments and questions:
      • This is a major improvement compared to what we used to have, i.e. disparate sources of information, sometimes not reliable, or simply scattered all over the place.
      • It is a first step, sketching what we have to converge to, also to motivate people from PAGs to help in POGs
      • Q: We would like to take this opportunity (the object review) to form a team of experts - we are not in the case where we can put finite numbers in these tables
      • A: It may not be possible for all groups to fill it (for the tau group it seems to be okay and ready, but for e/g it might not be quite as easy). So if a global recommendation cannot be given is a compact way, just use it as a list of bookmarks to the most up-to-date information, which already lays in POG’s twikis. In addition, we could make an expert call.
      • Q: What exactly should be achieved in the 1-week time scale?
      • A: Conveners should first prepare for the review and then will be told more precisely how to port most relevant information to the central object ID twiki.
    • Action item: prepare the draft, a minimum guidance, of what you want to put on the twiki

Central Monte Carlo production (p7)

  • We can now produce MC samples at a very high priority. All central samples should appear on DAS/DBS.
  • Some groups produce their own private samples, which cause many compatibility and cross-check issues
    • Some use CRAB, some others don’t. Samples are not all accessible from DBS. Sometimes end-users don’t have access to the metafiles used to produce a sample.
    • (Even from central production,) the probability to make mistakes is very high (wrong tags, wrong pileup conditions,...).
    • For analysts who need to produce their own sample, they must ask experts and request centrally produced samples for check. In addition, this will benefit from more robust codes and faster queuing.
    • Action item: If not already the case, we suggest that everyone goes into the habits within the group, of listing what is private, what is not, and put most useful ones on DBS so that they are available by all who need them.
  • Comments and questions:

    • Q: GEN-SIM step is stuck at 20K?
    • A: We don’t know how to change the priority

    • The TOPs can change their priority so everyone should be able to do it too. They rose the priority in case of problem. We have to ask experts for help and details.
    • The priority is set at submission, but it can be changed by accessing the interface (?) or contacting the computing experts (Gulielmo). It has to be done anyway at RECO level.
    • Some samples, such as muon samples, take too long and are not all produced centrally.
        • Trigger menus of the whole year should be taken into account (to be taken offline and figured out)

    • Q: We have our approval next week and we don’t have our final samples. What’s going on with 11-096?
    • A: Send numbers of PASs to the Physics Coordinators. You cannot miss the approval because of this.

Moriond 2012 Approval Marathon (p8)

  • A preliminary agenda has been prepared (link): about 40 analyses are shooting for approval. This is about the maximum capacity, but we can add in a few more talks.
  • Action item: Please check the proposed agenda, and let us know ASAP if an approval talk in your group hasn’t yet been scheduled, or if an analysis won’t make it.
  • The “central” or “plenary” session will be chaired by the Physics Coordinators. For parallel tracks, conveners are invited to chair the meetings, and attend as many sessions as possible.

p10

Luminosity will ramp up by ~%: new constants used for Moriond analyses and publications.

AOB

  • Rumors: If you’re asked (by the press or) approached by a theorist, you are not aware of any suspicious excess. We don’t have any idea of where the rumors come from; let’s keep denying.
  • Production with new beamspot has started. We don’t change the beam conditions unless we are forced to do it. The production is being restarted with 2011 beamspot. (for the pixel detector it is not dangerous, so we decided to stick to the current conditions.)

  • Q: What is going to be presented on February 14th at the LPCC meeting, results from 1-2 fb-1 of data?
  • A: No idea, they did not say. So far, nothing has been shown above 2 fb-1.

  • Spokesmen from the ATLAS and CMS experiments decided upon a gentlemen agreement: before releasing spectacular results, they will talk to each other.
  • The physics upgrade collaboration is launched: we promised to engage with them, make people available and now where are focused on the preparation of Moriond. But conveners should think beyond Moriond. The generation part is defined; once the simulation upgrade is done, once the code is available, we want to run the physics process to start analyzing the samples within the PAGs. It needs to be followed up. Chris and Albert are already watching the process. In order to follow our promises: if there are more PAGs interested in the upgrade, and have manpower avail, let us know.
  • 25 ns run: the accelerator division thinks we can higher up the luminosity, it seems to be no showstoppers.

Conference Proceeding Handling (Paris Sphicas)

We need to change the workflow in our conference reports. We are going to our final implementation.

Principle:

  • Current system:

Instructions to subeditors for conference reports (and abstracts?)

p4/ cannot correct the language except if what is said is wrong paris doing it himself, by hand...

p5-6/ CR ==author, rather CMS at large many mistakes in the CRs... plots not approved, intempestive copy-pastes. => A CR needs to be reviewed by the associate convener!

p7/ the conveners will appear as subeditors CR submitted, routed to conveners, who choose referees, conveners in any case responsible for the content of CRs.

p8/ improvements in the iCMS interface acquiring a CR, assignement of referee.

p12/ conveners should in addition watch the email exchange between the authors and referees

in short: acquire, edit, read, accept

p14/ joint CMS-ATLAS: special button in the interface to fwd to ATLAS and vice versa For now, paris will keep on receiving/treating requests or versions... Com: conveners can be by-passed by authors who were not waiting for the approval of the convener, but submit the report to the conf organization directly without refereeing. A: cannot be Q: what if the level of english is too bad and cannot be understood? A: direct people to tutoring? Greg: should watch for plagiats as well... (example of the miraculous improvement)

-- AnneFleurBarfuss - 06-Feb-2012

Edit | Attach | Watch | Print version | History: r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r1 - 2012-02-06 - unknown
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback