Minutes of the Jefferson Lab CLAS Collaboration June (7-9), 2001 Next meeting: October 25-27, 2001 at Jefferson Lab, CEBAF Center For you to do: 1) Read these minutes; check out the new bylaws. 2) Save the dates for the next CLAS meeting, given above. 3) Apply even more time/effort to you CLAS commitments this summer! ________________________________________________________________________ Over 95 members of the CLAS Collaboration met for a three-day summer meeting at CEBAF Center. During this meeting the spectrometer was running to do physics with linearly polarized photons for the first time. The biggest news items from the meeting were: (1) the Deep Virtual Compton Scattering proposal which will be defended next month was endorsed by the Collaboration; (2) Hall Crannell was selected as the next Chairman of the Collaboration; (3) we started a new Off-line Technical Working Group to streamline our offline software. Here are some numbers that characterize where we are: Since the last meeting in January, two more physics publications have a appeared in print, one in PRL and one in PRC. One publication is submitted to PRC, and one began its Collaboration review during this meeting (see separate email to CLAS members). All of these physics publications are available on our web site. Two instrumentation (NIM) papers appeared in print since the last meeting. At least four physics analyses are approaching the publication stage, meaning that they will be submitted (or be close to submission) by the time we meet again in October. The Collaboration has 82 identified analysis projects, most of them with data in hand, 41 current graduate students, and 8 completed PhDs. There are 9 post-doctoral researchers on site supported by outside user groups. Bernhard Mecking gave his overview of the status of CLAS (see separate mailing). Among other things, he emphasized the need to publish our results as quickly as possible now. New Proposal ------------ A new proposal entitled "Deeply Virtual Compton Scattering with CLAS at 6 GeV" was presented by Volker Burkert. It is available electronically at http://www.jlab.org/Hall-B/secure/e1/e1-dvcs/ The measurement would use CLAS with a retracted target and with the addition of additional calorimeter modules (lead tungstate) between the target and Region One to directly capture the emitted real photons. The proposal was submitted on about May 31, after a review by only a subset of the SoN. The plenary discussion showed support for the proposal from the rest of the Collaboration, but not enough for immediate ratification. It was decided to postpone the formal Collaboration vote until after the SoN meeting on Friday. On Saturday the issue was revisited, and the proposal was granted almost unanimous support. The proposal asks for 60 days of new beam time. Note that this is more than the 45 days Hall B got allocated for PAC20 in addition to the jeopardy time. Jeopardy Review --------------- The Structure of the Nucleon group has prepared for several months to defend the remaining "e1" running time (30 days) at the "jeopardy" review scheduled for July at PAC20. The plan for this defense was presented by Dan Carman. At the PAC meeting, an overview of the 17 e1 experiments will be presented by Volker Burkert. Detailed arguments for the remaining time will be made for lower Q^2 running for the N-to-Delta physics (Cole Smith), and for the multi-pion physics at higher energies (Marco Ripani). Three "stand-alone" experiments will also present defenses of their beamtime at the jeopardy review. One is 91-015, the helicity structure of pion photoproduction. Dan Sober will motivate this partly to build support for constructing a frozen spin target for CLAS. Looking for medium modifications in rho photoproduction, experiment 94-002, was re-animated by Dennis Weygand. A target design now exists, and studies of existing g6 data suggest that the necessary e/pi rejection factors are achievable. The double-delta component of the deuteron wavefunction, experiment 93-043, will be defended by Brian Quinn. It requires running CLAS with reverse toroid field with a deuteron target. Several other experiments that were eligible for jeopardy review as of the January '01 meeting have escaped this scrutiny by being scheduled to run within the first 6 months of '02. The rest of the CLAS experiments will undergo a jeopardy review in January '02. They are mostly real photon experiments. A strategy for how these experiments will be presented and defended is under discussion. Off-Line Technical Working Group (OTWG) --------------------------------------- This new group is charged with the long-term maintenance and improvement of the CLAS off-line software. The members are long-term employees, of JLab or of university groups, who have experience both with software and with the various detector systems in the spectrometer. The OTWG group provides working software "tools" and training for persons charged with calibrating or cooking data. The Run Groups must continue to provide the manpower to carry out the work of calibration and cooking. We plan to hold a meeting among the OTWG, the CCC, and Run Group representatives at each Collaboration meeting from now on. Each Run Group can appoint a knowledgeable person to represent its needs. Each Run Group representative should be prepared to assign some of "his" manpower to coordinated calibration and cooking of the CLAS data in the analysis pipeline. We will try to hash out a plan to coordinate calibration and cooking efforts across run groups at these meetings. Examples of what the OTWG can do to provide for continuity and improvement of CLAS software are: (1) Grad student M. Guillo has written a code to calibrate the EC timing using real photon data. Who will maintain this code after he leaves or is busy writing this thesis? Cole Smith has agreed to adopt/adapt this code, as needed, in the future. (2) The LAC timing code was originally worked out by Vladimir Sapunenko. A more general code was written by grad student Jeff Lachniet. Both codes are still under some development, but both people have moved on to other work as well. Who will keep track of these codes in the future, possibly merge them, or advocate one over the other? Mike Vineyard has agreed to do this job. The people who have agreed to serve on the OTWG are: Principal Vice-Principal Software Environment Mark Ito Harut Avagian (library,system,cooking scripts) Drift Chambers Mac Mestayer David Lawrence Scintillation (TOF) Counters Elton Smith TBD Electron, Gamma, Neutron Counters Cole Smith Stepan Stepanyan Photon Tagging/Normalization Eugene Pasyuk TBD Simulation packages (GSIM) Will Brooks Maurik Holtrop Large Angle Calorimeters Mike Vineyard TBD Cerenkov Counters TBD TBD The "TBD" (to be determined) entries are opportunities for University Groups to fulfill Service Work needs by having an on-site person take a role for a period of a few years. The OTWG Bylaw which was discussed and adopted (with revisions) at this meeting, is included here: -------------------------------------------------------------------------- BYLAW TO CLAS CHARTER FOR THE DATA CALIBRATION AND COOKING Adopted 6-7-01 I. The responsibility for calibration and cooking of CLAS data rests with the Collaboration as a whole, which includes the Jefferson Lab Hall B staff and all member Universities. Primary responsibility for calibration and cooking of a given data set rests with the Run Group which took the data. Each Run Group shall designate an Analysis Coordinator to expedite the process of calibration, cooking, and analysis. Primary responsibility for providing working software tools for this purpose rests with the Off-line Technical Working Group defined below. II. Off-line Technical Working Group A) Each major software element of the off-line CLAS software (to be listed below) will have a long-term employee of Jefferson Lab or a member University who has the following responsibilities: 1) Continued improvement and maintenance of the software package. 2) Ensuring that the software is applied properly to each data set, although the final judgment as to the validity of the output rests with the Run Groups. 3) Training of new calibrators/cookers who wish to use the CLAS software. B) A suitable person is generally NOT a short-term post-doc or graduate student, but rather someone who has a long-term position and who will likely be available for the duration of the CLAS program. Membership of this group can change, but the intention is that each of the senior persons have a long-term commitment to this task. C) Each major software element should have both a person with the principal responsibility, but also a vice-principal person to help maintain and improve the codes, documentation, and procedures. These latter persons can be at the level of post-docs and graduate students. D) It is NOT the function of the Off-line Technical Working Group to provide the manpower to run the calibration and cooking jobs. The manpower must be provided by the Run Groups. III. Governance: The Coordinating Committee, the Off-line Technical Working Group, and all the Analysis Coordinators will meet at each Collaboration Meeting. They will discuss progress made in CLAS calibration and cooking. In the event of scarcity of resources such as manpower or computing facilities, it is in this forum that decisions about resource allocations should be made. Service Work commitments of the Universities can and should include providing manpower towards these needs. -----------------End of Bylaw------------------------------------------ Other Notes ----------- In the plenary sessions, very interesting physics talks on advanced analyses were given by Cole Smith, Marco Ripani, Valeri Koubarovsky, Kyungseon Joo, Jean-Marc Laget, Barry Ritchie, Kim Egiyan, Ioana Niculescu, and Peter Heimberg. [These talks will not be recounted or summarized here: there was so much information in these talks it is impossible to capture fairly, and these minutes are long enough already! Hardware, Software, Infrastructure ---------------------------------- Dieter Cords reported on continuing gains in performance of the Data Acquisition system. It now can acquire data up to about 4000/sec. The present limit in the system the rate of writing data to disk, at about 25 Mbyte/sec. There is now 600 Gb of on-line RAID disk space. The photon trigger setup is now no longer a separate GUI, but is merged into the CODA configuration setup. [Hooray.] New beamline instrumentation software has been developed for the g8 experiments. The polarized photon beam coherent bremsstrahlung facility is operational. The new upstream Pair Spectrometer (will be used for PRIMEX experiment) is being read out with a separate CODA trigger. There is a discussion underway to restructure some of the online banks to reduce bank overhead on disk. F. Klein claims that such a change can save up to 25% of disk space with no penalty in terms of increased analysis computation time. Mark Ito discussed the offline situation. Most CLAS data is stored on the "old" Redwood tapes. New data and analysis results are stored on "9840" tapes which are believed to be more reliable. "Data Movers" are server computers which move the data to/from tapes in the silo using new "Jasmine" software developed at JLab. It is hoped that this new software will improve reliability and transparency in the off-line data handling. A second tape silo has been purchased (the first one is full), and is planned to be available by the end of July. There are now 4 Terabytes of work space available, over 10 Terabytes of raw data cache space, and 1.2 Terabytes of DST cache. The transition to using the calibration database CalDB, started in January, has been slow. A lot of "private" old-style maps now exist. More utilities for moving data to and from the CalDB database are under development by Mark and Harut. The eg1b Run Group is planning to use CalDB from the start. [For older run groups it is not so clear.] Mark distributed copies of a splendid compendium of FAQ for the offline software environment. Saturday Discussion ------------------- On Saturday the Chairpersons of the Physics Working Groups gave summaries of the PWG meetings; we will not summarize the summaries here. David Tedeschi was elected as the new Chairman of the Real Photon Working Group. He will replace Bill Briscoe on the Coordinating Committee starting sometime this summer. [We thank Bill for his two years of service as RPWG Chair.] The report of the Coordinating Committee was made by Reinhard Schumacher. For the PAC 20 jeopardy talks we decided to call for a set of practice talks on Thursday July 12. The CCC noted that the real photon group has to prepare its strategy for the jeopardy review in January. This process has barely started. The CDR for the JLab upgrade will likely be due about April 2002. Community input into the document would be helpful; Bernhard Mecking is organizing this effort. Membership ---------- Kevin Giovanetti ran the election of the new CLAS Chairman on Friday morning. Hall Crannell and Sebastian Kuhn were the candidates selected by a nominating committee selected by the CCC. The candidates spoke briefly and answered questions. Balloting was electronic (before the meeting) and on paper (during the meeting). Hall Crannell was elected, and will start serving in September 2001. [Congratulations and Good Luck, Hall. This is a nice "retirement" project.] An authorship bylaw was discussed and adopted to regulate how new members get added to the CLAS author list, and when departing people get deleted from the CLAS author list. It covers situations not considered in the early days of the Collaboration, and it does not restrict the requirements. It liberalizes the authorship rules slightly such that any new member who is doing adequate service work will be added to CLAS papers. The text follows: ----------------------------------------------------------------------- BYLAW ADDITION TO BYLAW C2, SECTION 7 Adopted 6-7-01 Term members who join the CLAS Collaboration will be included on CLAS publications after a six month waiting period. During this time the term members must demonstrate commitment to CLAS by doing service work for the Collaboration which was agreed upon at their time of joining. They will then be included as authors on CLAS publications (if they so choose) even if their own work did not directly contribute to the published analyses. The Membership Committee will judge whether the new term members have done suitable service work during the waiting period. Term and full members who withdraw from the CLAS Collaboration will be included (if they so choose) as authors on CLAS papers submitted for six months following their departure. (Nothing in this bylaw changes the authorship provisions specified in the CLAS charter, specifically see Charter Section 11.B) ------(end of bylaw)-------------------------------------------------- Talks ----- Jean-Marc Laget, chairman of the CLAS Speaker Committee, reported on recent trends in invited and contributed talks. A total of 134 invited and contributed talks have been given, and since last time there are six new CLAS speakers. We saw that the CSC is actively soliciting CLAS talks at major conferences and at APS meetings. We were reminded that all talk invitations to conferences must be given to the CSC for approval, and that the procedure for doing this is well documented on the web page maintained by the CSC. Service Work ------------ Barry Berman will be the Chair of the Service Work Committee starting this summer. [We thank Ken Hicks for his efforts in this role for the last few years.] Barry will assemble his committee and start the process of collecting new Statements of Service from all member institutions this summer. Discussion led to the recommendation this committee finish its task in time for the next Collaboration meeting. The main meeting adjourned at about noon on Saturday with 50 survivors still in the room. ------------------------------------------------------------------------ Any changes, additions or other clarifications of these minutes by the Collaboration members will be cheerfully considered. Amended minutes will be sent out only in case of egregious errors or omissions. These minutes are in no way meant to be exhaustive [despite being exhausting]. [...]=chairman's editorials