\documentstyle[12pt,a4]{article} % pre-amble to re-set sytle options \headheight 0mm % No header text \headsep 0mm % No distance between header and body of text \parindent 0mm % No paragraph indentation \parskip 5mm % Size of gap between paragraphs % end of pre-amble \begin{document} \begin{center} {\normalsize\bf MINUTES OF THE WORKING GROUP 2 MEETING OF CCP4} \\ {\normalsize\bf HELD AT BIRKBECK COLLEGE ON 12th JUNE 1991} \end{center} \\ \begin{tabular}{ll} \it{Present:} & \\ & \\ David Rice (Chairman) & Sheffiled University \\ Phil Evans & MRC, Cambridge \\ Eleanor Dodson & York University \\ Sheila Gover & Oxford University \\ Kim Henrick & MRC, Cambridge \\ Don Akrigg & Leeds University \\ Peter Daly & Daresbury Laboratory \\ Sandra McLaughlin & EMBL, Hamburg \\ Peter Brick & Imperial College \\ Jim Raftery & Manchester University \\ Cameron Dunn & Bristol University \\ Ian Tickle & Birkbeck College \\ Francesca Young & University of Glasgow \\ Ian Clifton & Oxford University \\ Paul Taylor & Edinburgh University \\ Wojciech Wolf (Secretary) & Daresbury Laboratory \end{tabular} 1. {\it Minutes of the previous meeting} The minutes of the meeting of 10th April 1991 were agreed. 2. {\it Matters Arising} David Rice thanked Phil Evans for chairing the previous meeting. 3. {\it The 1992 Study Weekend} Eleanor Dodson distributed a list of speakers and a provisional timetable for the Study Weekend drawn up by herself and Sheila Gover. The names of the speakers on the list had been contacted and had responded positively to the invitation. It was agreed that afternoon sessions should consist of a number of short case studies. They should concentrate on practical aspects of the applied theory rather then provide a deep analyses of the investigated structure. There was some discussion on the length of the talks. David Rice thought that the time allocated to the theory was a too short and that consequently speakers may be tempted to cover the subject rather more rapidly than they would normally do with the risk of losing the audience quite quickly. One suggestion was to split some of the theory talks into two parts (possibly with two speakers); the first part being a treatment of the theory and the latter a `walk--through' a worked example. It was thought that, for example, the topic of rotation functions would benefit from this approach. There was general consensus that all the speakers should produce pre-printed notes to be circulated before their talk to enable the formulae used to be recalled more readily. The meeting was also strongly in favour of speakers being issued with a list of requirements concerning the preparation of slides and transparencies, detailing minimum letter sizes, use of colours, etc. Other suggestions for the study weekend included case studies of {\it Porin} with Richard Tillman, a multi-domain problem (Simon Phillips was suggested) and Phil Evans thought that Gerard Bricogne could give a lecture on performing better Patterson searches using maximum likelihood theory. The venue was again discussed and the Secretary confirmed that the {\it Lord Daresbury Hotel} had been booked. Concerns over the suitability of the lecture hall were again raised. The Secretary reported that the speakers platform can be up to 1m high and large screens, microphones and a TV relay were still being investigated. The close packing of chairs within the room was raised as a potential problem. Discussion then turned to costs. Eleanor Dodson pointed out that all speakers contacted so far have been clearly instructed to take advantage of Apex fares and to stay over on the Saturday night. The fee was provisionally set at last year's level of \pounds45 per participant. The requests for subsidy are to be on the same basis as last year, generally a `first come, frst served basis' but the deadline for applications was thought to be too late last year and a tentative date of {\bf 10th December 1991} was agreed. The planning of the rest of the timetable and the publicity has been left to Eleanor Dodson, Sheila Gover and Wojciech Wolf, and it has been thought that an announcement of the weekend could be made in August giving enough time before the deadline. 4. {\it Report on the EACBM/WG2.1 Meeting at Como} Sandra McLaughlin distributed the minutes of the meeting of WG2.1 that took place during the Como School\footnote{Copy is attached to the minutes.}. Those who attended felt that it was a very worthwhile meeting since many people who may not normally attend these meetings did gain the opportunity to join in by holding it at the School. The meeting was an update of the current state of MTZ format (see discussion below) and the Input and Output requirements of CCP4 programs. The current confusion over different packages was described, for example, should the programs use Motif, PHIGS, PEX or XView. Alwyn Jones thought that Motif is likely to be the industry standard and has already started to produce an interface with it for his ``O'' program. Using such interfaces to input into CCP4 programs was also considered. The two competeing options at present being David Wild's work and Phil Bourne's Crystallographic Workbench. There is still, however, general agreement that most CCP4 users are not concerned with pull-down menus and windows. Alwyn Jones described his template system, as it already exists within ``O''. The output from CCP4 programs was again discussed. There is already a wide agreement that the tabular data that is currently output is more readily understandable in graphical form and Jan Zelinka has produced a small program, Xgraph, to perform this function. It uses the Athena Widget Set. Kim Henrick remarked that there was no need to write a single line of code for X Windows since there was a wealth of public domain software already available on the Daresbury Info-Server. He further suggested GNUPLOT as an appropriate package to draw graphs of tabular data - the input is in ascii and can easily be filtered from the current output tables of CCP4. VOGLE was re-emphasised as the replacement to PLOT84 and it was requested that Peter Daly write the PLOT84 routines to use VOGLE and distribute this, Peter said it was on his (long) list already. David Rice complained that a major problem with the output is that it simply did not explain what the output meant and that students were not likely to look at the current output in any detail because a) important indicators (such as large values) were well hidden in a mass of irrelevant output and b) they don't know what thay mean anyway. Much time can be wasted using a `black box' approach to crystallography where output is not carefully sifted and students will continue in their work unaware that there is a serious error in the data. David suggested a revision of the documentation to include an explanation of the output (and its meaning) and program authors should detail a student to look through some output and send a note to Daresbury regarding how the documentation to that program can be improved. Kim Henrick considered this impractical since most authors do not have the time and suggested that Daresbury and Hamburg tackle the documentation problem in their way and the rest of the community adopt it wholesale. David Rice disagreed. 5. {\it The ESF Grenoble Meeting} The Secretary reported that David Blow had instructed him to subsidise up to six participants to this meeting from the UK using CCP4 funds. The cost is thought to be about \pounds2000. 6. {\it Newsletter} Contributions to the next Newsletter were called for. David Rice reported a opinion of David Blow that the appendix to the first article (Wilson and Drenth) of the last Newsletter was inaccurate and did not include enough British groups. This was partly the reason for the funding of the Grenoble meeting. It was also pointed out that future funding of CCP4 might be in jeopardy if it was thought that CCP4 was being absorbed by Europe. There was a suggestion that the minutes of the EACBM/WG2.1 meetings should be included in future Newsletters in sysnopsis form and selected highlights or an annual report on CCP4 should also be included. The provisional deadline for contributions to the next Newsletter was set up on 1st November, 1991. 7. {\it Report on MTZ Progress} MTZ is progressing forward and a good opportunity exists for a major ``version release'' of CCP4 to become a reality. This should enable to purge program suite and again classify particular programs among {\it supported} and {\it unsupported} software. The supported set should include the new MTZ programs and Kim Henrick's Unix startup routine CCPFYP. The libraries (in general) need to be sifted and common blocks made internally consistent and calls to arguments adjusted so that main programs should not have to use the commons in the library. Work was started on this with the parser, which now uses no commons in user code. Unsuported (and scrapped) programs would include all LCF programs (except the libraries), RFACTOR, LOCAL, COMBINE, PHASE, UNIQUE, REFINE ans ANSC. A list of programs that use MTZ was identified (many of which have been converted) including LCF2MTZ, MTZDUMP, MTZ2ASCII, ASCII2MTZ, SORTMTZ, FFT, GENSFC, RSTATS, SIGMAA, PHARE (PHARO), ALMN, ANISOSC, ECALC, TSFGEN, TFPART, POLARRFN, HEAVY, CADMTZ, RSEARCH, MTZUTILS, MTZEDIT, ABSCALE, ABSURD, ROTAVATA, AGROVATA, TRUNCATE, POSTREF. 8. {\it HDF} The discussion on the Heirarchical Data Format (HDF) was short. The HDF idea is to make binary files machine independent and was raised by David Wild via Daresbury and Hamburg. After a heated e-mail debate between interested parties the idea has been dropped out. Kim Henrick pointed out that since most CCP4 programs are I/O bound then accessing data files across a network is going to kill any performance of your local machine and is a waste of resources. The problem of missing columns in MTZ files was raised and the idea of zero standard deviation of a large negative number were considered as options, although no firm conclusion was reached. 9. {\it Any Other Business} David Rice raised the topic of Oxford Molecular, the Secretary reported that since the last meeting they had been in contact again and were prepared to consider any suggestions from CCP4 regarding commercialising the software including distributing the source code free to academic instituions. There are still many problems with any proposal to commercialise CCP4, a new factor is the ownership of the MTZ source since it has been jointly funded by CCP4 and EACBM. It was generally agreed that such proposals are not in the interests of CCP4. It is highly likely that commercialising the package can stop inflow of the new programs into the suite. It is also unclear what benefits would accrue to CCP4 if any agreement could be found. Don Akrigg mentioned that a huge effort is required to `teach' Oxford Molecular programmers what the programs actually do. The Working Group 2 felt not to be competent to discuss this subject further and passed discussion to the next Working Group 1 meeting. However the common feeling was that the bottom line for any discussion should be free distribution of the package including source code to all academics worldwide and that this condition was not to be negotiable. Concern was raised over the collaboration with Lynn Ten Eyck. David Wild does have contact with him but nobody at the meeting has heard anything from San Diego since last February, when Ten Eyck joined the Heidelberg meeting. It is likely that his funding is up for renewal within the next twelve months.It was agreed that David Rice would write to Dr. Ten Eyck and ask for a report on the work that he said he would undertake. 9. {\it Venue and date of next meeting} {\sc\small A date for the next meeting was set for Wednesday the 2nd October 1991, at York University for 11-30 am.} \bigskip \medskip \centerline{W. Wolf (Secretary)} \end{document} % End document