LCD Weekly
"Reconstruction and Analysis"
04/18/2005

Attendees:

NG (Norman)
TJ (Tony)
JM (Jeremy)
SM (Steve)
WM (Wolfgang)
JS (Jan)
MC (Matthew Charles)
WL (Willy)
MD (Mark)
NTM (Niels)
T? (Tim)

--

NG: Series of meetings to formalize the group.  "Critical mass".  Formal meeting to exchange information.  Developers forum.

NG: [slides]

NG: Snowmass -- detector concepts into a design.  Each is settling into a canonical design.  Want to put into detector CDR.  Barry Barrish would like "some idea" of cost in CDR for machine.  Reasonably detailed studies of detectors by/started/done w.r.t. Snowmass.  Comparisons between concepts.

NG: Less than 4 months away.  "Don't expect much."  Would like to certain amount of work on SiD before Snowmass.  Want full rather than fast MC.

NG: New reco in org.lcsim.  Porting existin capabilities from hep.lcd.  Limited by manpower, so rewriting tuts using the new software.  Upgrade path for data from SIO -> LCIO.

MC: SLCIO same as LCIO?

NG: Yes.

NG: New analysis -- start with org.lcsim.  Keep up with new users so see new functionality.

NG: Arbitrarily complex geometries using SLIC.

NG: environment...

    simu
      slic
      lelaps
      lcdg4
    reco
      java 1.5
      JAS3 0.8+
      org.lcsim

WM: Problems with Java 1.5?  Postings on news groups of 'autoboxing'.

TJ: Perhaps can be slow if do without realizing.  

NG: Do have collections keyed on integers, for instance calorimeter cells on ID -- primitive integers.

NG: Request WM point to autoboxing posting.

NG: software

    CVS
      head should build
      strongly encourage users to checkin code on regular basis
        gauge amount of work by others
        global refactoring
        allows collaboration / help among group
    CD or DVD
      release @ beginning of June
      maybe another snapshot prior to Snowmass
          
NG: Free CD, then people will actually use it.

NG: Data samples -- ~10k eventst of ~10 canonical signals.  Small fraction available on CD/DVD

WL: Website of datasets?

NG: Yes, will get to it.

NG: Large standard model inclusive using WHIZARD.  Regenerate with ILC specs.  Don't expect to push all through detector, but can use for FastMC and then to FullMC.

NG: Detector designs -- SLAC will spend most effort on SiD.  Various subgroups are signing off on subsystems.

NG: Start with compact XML description, which will be basis for all geometries in org.lcsim.

NG: Approximations of LDC and GLD also available.

NG: CVS package LCDetectors.  

NG: Match LCIO files with associated detector in LCDetectors.

NG: Resources [see slides]

NG: [show lcsim.org dataset page and ftp site]

WL: How tell which simu generated?

NG: Presumably in header.

NG: Data crawler / indexer in future.

WL: Interface to JAS?

TJ: No interface currently.  Just download and analyze.  Can put in data catalog and make data browser.

NG: Not a high priority but could be.

JS: All datasets on NFS?

NG: Yes.

NG: [show linearcollider.org forum]

NG: More activity on simu side.

NG: Encourage people to use the forum, because threaded and publically available.  
 
WL: No results.

NG: [show Confluence Wiki]

NG: Hopefully will became a major resource.

JS: Notification?

JM/TJ: 'Watch' command.

WL: How to change?

NG: [demonstrate editing interface]

NG: Encourage use of Wiki for 'dynamic' problems.  Documents about solving problems.  Collaborative.

TJ: Attach comments to pages.

NG: Can use 'snapshots' of Confluence pages for docs.

WM: LCIO format 'ready in June' to be used for clustering?  

NG: Nothing prevents right now.

TJ: Take cluster finding algo and port to org.lcsim framework from hep.lcd.

WM: RC said not to move over right now because 'information not available'.

NG: For all current detectors, info is available.

TJ: SDJan03 and SDFeb05 are in org.lcsim.

NG: [show example cluster finding algo in JAS3]

WM: Question about detectors in CVS -- includes various versions for SDFeb05 with different had cal impl?

NG: Each will be a different tag, e.g. 'SDFeb05 scintillating HCal'.  

JM: Embedded in org.lcsim?

TJ: No -- detectors are downloaded.  SDJan03 is embedded.

WM: Event display 

NG: [show detectors in LCDetectors]

JM: Detectors in CVS?

NG: Just local, but will be checked-in.

NG: Concurrence, then make Tues 1:30 regular meeting.

NG: If get results, post to forum.

WL: Reminder before meeting?

NG: Yes.  Email just to 'local folks', but will send to larger group.

JM: Using Jira/Confluence for 'bugs'?

NG: Definitely post valid bugs to Jira.

TJ: Useful to post feature requests to Jira for tracking purposes.

TJ: Status updates?

NG: Tracking -- getting sim hits to digi hits.  Track finding software based on space points should then work.

NG: JS working on vertexing.

JS: Waiting for modifications to track.  Would like track accessed via symbolic values, e.g. enum.

JS: Mods to space point.

JS: How many people using space point vector?

NG: No one, so easy to change.

JS: Can checkin now.  Should compile.

JM: Who include in LCDWeekly?

NG: Lima, Vishnu, Dhiman, Eckhard, Graham, UofC, developers.

NG: Tracking, cal, etc. exist, so don't want to conflict with them.

NG: Time is short between now and Snowmass.