Steve Dytman
This meeting was highlighted by the advances in data analysis of various run groups. Although only elastic and inclusive ep scattering cross sections are able to be compared with calculation, many specific final states can be cleanly identified and important physics distributions are being constructed. This is great news for the collaboration, but causes various changes in the way the collaboration works. For the next collaboration meeting, there will be many more results to be given. I will ask various physics working groups to fill short sessions (1-2 hours) with talks and they will have to choose how to use the time. The more significant changes are discussed below in the business meeting summary.
The initial hardware complement of CLAS is now installed and working. With the exception of the DAQ rate (about 550 Hz), performance is approaching specifications. This isn't totally surprising because the technology used for DAQ is less standard than that used for the detectors and the manpower for DAQ has always been too small. Dieter Cords outlined recent improvements that are expected to bring about a significant improvement this week.
A number of short talks on CALCOM, e1 and g1 results were presented. Many of the e1 and all of the g1 talks were given by grad students. A central problem presented was the remaining difficulty in getting good resolution. Empirical shifts in angle were made in various ways that had significant repercussions elsewhere. One of the highlights of the meeting came Sat morn when John McNabb presented the solution--the array for minitoroid magn field had been displaced by one word, confusing x, y, and z components and causing virtually all of the problems. A first run by Keongsung Joo found much improved (50% as I remember) missing mass resolution.
As usual, many talks presented software progress because that continues to be an area of significant advance. GSIM progress has been rapid and impressive, but much remains. Long runs are being made (and sometimes restarted when new bug fixes are made) for a first acceptance calc. Results are qualitatively fine in most cases, but problems are sometimes found when data distributions are compared with predictions using a physics generator based on previous data and present GSIM representation of CLAS. In some cases (e.g pi+ electro- production), the problem is that the new data is much better than old, so that the generator is wrong. I believe that means the runs will have to be redone with a better (to me, a less complicated) generator. e1 data has been 'cooked' to a large extent. The code is stable and existing linux boxes are capable of good turnaround for data files we have now, but will be insufficient when data rates increase. More linux boxes are on order.
A significant effort toward defining the physics and instrumentation for a CLAS upgrade for 8 GeV beams has been led by Bernhard Mecking. He gave a snapshot of what might be built. At the risk of introducing mistakes, I will try to summarize it here because it is important for all of us. Next weeks' Upgrade Workshop will give the opportunity for more discussion. Improvements not associated with the energy increase include daq rate improvement and better calorimeter coverage. For higher energies, the phi and low theta holes in CLAS should be filled. He suggests replacement of region 1 and the minitoroid with a solenoidal field (or polarized target field) and new detectors. High mass calorimeter segments will cover the 6 cryostats for neutrals. Some low mass tracking will be in front of the cal segments and stretch across the active area of the outer detectors. In front of that could come a simplified RICH detector. There is room for anyone wanting to contribute to this design effort.
Gail Dodge reported results of the Membership Committee meeting. At the January meeting, 4 candidates for full membership were approved after review of their proposed Statement of Service (which outline the service jobs they will do for the collaboration). The SOS documents are on the web. On Sat, Marco Battaglieri (INFN-Genova), Laird Kramer (Fla. Intern.), Allena Opper (Ohio U.), and Dave Tedeschi (So. Car.) were all approved by the collaboration and are now full members. This time, the Membership Committee discussed the new SOS of Derek Branford (Edinburgh). His membership application was forwarded to the collaboration for final approval, but they requested that the SOS be augmented with more detail on the proposed activities.
The PAC meeting dates for this summer were chosen to make review of new CLAS proposals in the normal way impossible, but I felt required to have some review since we have had none for about a year. The Coordinating Committee reviewed 3 proto-proposals and sent all 3 to the relevant working group. Each working group did a review of their proposal (there happened to be 1 for each group), though usually less public than our normal procedure. Two of the proposals were approved and the 3rd was deferred until the next collaboration meeting. Proposals approved were for rho electroproduction from the proton at high Q^2 (spokesmen= M. Guidal, C. Marchand, and E. Smith) and for R_{LT}' for (e,e'p) and (e,e'pp) for ^3He target (spokesmen= M. Holtrop and W. Hersman). This was an unusual situation and will not be done again. Nevertheless, I strongly feel that the proposals approved were of sufficient quality.
The Coordinating Committee discussed procedures for management of data analyses that were not approved by a PAC. One of the advantages of CLAS is that many experiments are done simultaneously and there is no need to have all of them passed by a PAC. To have better understanding of this situation, we request those of you doing data analyses to send a brief (1 paragraph) description to me for recording purposes only. This does not supercede any aspect of the charter and only extends previous agreements. If an analysis has a PAC approved proposal or if a thesis student is identified with an analysis, these have precedence. Recording the analysis does not establish precedence, but does let the rest of the collaboration know of progress. This way, those doing parallel analyses will have better chance to work together to head off conflicts at a later time. Ultimately, analyses must be approved by the appropriate working group and/or the collaboration via procedures that are not yet defined.
Mac Mestayer presented a proposal for a Service Committee that will make the service of CLAS groups more organized than it is now. This idea was overwhelmingly approved by those at the last collab mtg, by the Membership Comm. and by the Coordinating Comm. After some discussion, it was approved by the collaboration as a by-law. The exact wording of this by-law will be forwarded to the collaboration soon. The main justification is to be sure the service work for CLAS is done efficiently as we evolve from a collaboration building a complicated device to one that uses and maintains a complicated device. The Service Committee will be be composed of 3 from the Membership Comm and 3 from the Technical Comm. Each CLAS group representing a school or lab will be asked to submit a list of tasks they are willing to do, in many cases they will already be doing the task. The committee will help groups choose tasks. Anyone who wants to serve on this committee should contact me or Mac.
The collab has very few formal procedures to govern the way CLAS results are presented at conferences. Unfortunately, most of us have been very occupied making CLAS a working device. In the charter, requests to the collab are to be apportioned by the Coord Comm. At present, conference organizers are making requests only to individuals and these talks are largely outside the collaboration review. (The exception is that any conference report must be approved by the Coord Comm.) In addition, there is presently no mechanism for the collaboration to review the results to be shown at conferences. We discussed the mechanisms to choose speakers in more equitable ways in private and at the meeting Sat morn. Arne Freyberger (CLEO) and Jim Mueller (CDF) described ways high energy collabs deal with this issue. They require ALL requests for talks to come through the collab and try to distribute them in ways to shift some of the credit for results from the older experienced folks to the young people. Various people expressed doubt whether high energy collabs are as democratic as it appears and whether we can or should reproduce such a system. However, there is always a Speakers Comm and some of us (much of what was presented was first suggested by Volker Burkert) made a proposal for such a comm. The basic structure of this comm was overwhelmingly approved by the collab and I was delegated to convene such a comm with the charge to develop a formal proposal. This comm will be the 5 members of the Coord Comm plus 3 others chosen by the Physics Working Groups and 2 others that I will choose to be sure there is proper representation. One of the approved goals of this comm is to replace the Charter article (XII.A) requiring only notification of an invited talk to the collab. Another is to define a Speakers Comm (nowhere mentioned in the Charter) that will solicit and suggest talks to conference organizers and try to choose speakers in a reasonable way. (Anyone interested in the Charter language should go to http://www.physics.odu.edu/~dodge/charter_jan98.txt The proposal was that invited individuals are requested to ask the Speakers Comm for approval to speak through a ~1 page document. In many cases, this approval will be granted. Contributed talks (content and speakers) would be developed through the physics working groups and approved by the Speakers Comm. There will be a formal proposal that will likely be similar to the above ideas at the next collab mtg. In the meantime, those of us who get invitations are REQUIRED to notify the Coord Comm; in addition, they are REQUESTED to clear their results with the appropriate working group and to give a practise talk at the lab. Some people suggested transparencies should be put onto a secure web site; such a site exists.