May 22, 2007 Software

From GlueXWiki
Jump to: navigation, search

Agenda

  1. Review minutes from May 1, 2007 Software meeting
  2. Portal/Wiki Security (Zisis P.)
  3. Update to FDC response model in HDGeant (Simon T.)[1]
  4. CD-2 document (David L.)
  5. Action Items

Location

Cebaf Center F324-325

To connect by telephone: 1.) dial:

800-377-8846 : US
888-276-7715 : Canada
302-709-8424 : International

2.) enter participant code: 39527048# (remember the "#")

We may have a VRVS connection depending if the equipment is in good working order. No guarantees.

Minutes

Attendees: David L.(chair), Mark I., Elke A., Elliott W., Simon T., Paul E.

On the phone/VRVS: Mihajlo K., Alex D., Zisis P.

Announcments

  • Elliott announced that monthly Hall-D DAQ/online meetings will be starting up soon. Scheduling is currently underway.
  • Mihajlo asked to be added to the agenda to talk about the new PID library he's started if there is time. (It turns out we ran long so this was derferred until next time)
  • Alex announced that he and Eugene C. are working on getting Pythia running for generating GlueX physics background. Elke has provided them some code she used at Hermes. Alex and Eugene plan to use Pythia to generate cross-sections they can compare to known ones. Results will be presented next meeting.
  • Alex announced that he is also working on understanding the BCAL energy sharing between sectors and layers using parameterizations found in the literature. He will report at next week's calorimetry meeting.
  • Richard announced 2 new wiki pages. One on the geometry options for the lower material designs for the FDC. The other for the low energy response for the BCAL.

Portal/Wiki Security

Zisis followed up on a discussion he'd started at the last collaboration meeting regarding security in our common access web-based tools. Namely, the portal, DocDB, and the Wiki. The "private" part of the DocDB is accessed almost all of the time through a single account with a well-known username/password. This same username/password is used for the accessing the Wiki. Since this username/password has been in use for a few years now, it may be time to change it. Elke noted that Hermes routinely changed group account passwords once a year. It was agreed that we should probably change the password, but it should be done at the next collaboration meeting where is can be widely announced without recording it on electronic media.

We briefly discussed the idea of certificates which may be needed in the future. The advantage is that you don't need to type your password all the time. The disadvantage is you can then only access the resource from machines which have had the certificate installed properly. Further discussion of this was deferred until another time.

Coupled with this issue is the question of exactly who should have access. It was noted that an updated collaboration list is being prepared by Paul and Richard for the upcoming spokesperson election. They will will convey this list to Zisis who will update the names in the DocDB with it. David noted that once we have this list, we should make a definitive home for it so that it may be more easily maintained. When asked, Mark briefly described how he'd dumped the PrimEx collaboration list into a relational database. This included fields for name, e-mail, and institutional affiliation. It was agreed that we should not spend too much time on this now, but should at least put the ASCII file with the list Paul and Richard make into the repository as a well-known, central location until a list maintainer can be identified.

Update to FDC response model in HDGeant

Simon gave an update on work he's been doing to implement the systematic smearing effects of the magnetic field on the avalanche position in the FDC. He has parametrized the effect using input generated by GARFIELD. His implementation uses a lookup table to smear the hits in hdgeant based on the location of the track. Initially, he sees no discernable effect in the momentum resolution a'la mctrk_ana. He does not believe this is correct (subsequently, he proved himself right). Work will continue on this as well as implementing algorithms in the reconstruction to account for this effect.

Action Items

  1. Put updated collaboration list in DocDB and commit list to repository