CERN: J Apostolakis (Chairman), G Cosmo, G Folger, Jürgen Knobloch, M Maire, M G Pia, F Ranjard, S Sadilov, M Schroder, P Truscott (Secretary), L Urban, J-P Wellisch

KEK: K Amako, T Sasaki (before 11:00 CET)

University of Manchester: J Allison

University of Naruto (telephone): H Yoshida

Santender: P Arce

SLAC: W Lockman (part-time), D Wright

TRIUMF: P Gumplinger

M Verderi (Ecole Polytechnique)

Comments had been submitted by M G Pia as corrections to the minutes of the meeting of 25^th January 2001, however these remained to be incorporated. Subject to these corrections, the minutes of the meeting (currently at version 1.1) were accepted.

A request was made (F Ranjard, P Truscott) for future minutes to be shorter and reflect only the conclusions (rather than the detail) of the discussions.

Actions from TSB Meetings 9 and 10 were reviewed (see Annex A for Action List).

A collaboration board meeting was held in Paris on 12^th February 2001. Key points of discussion were the Geant4 licence, disclaimer of warranty and milestones for the past year.

The CB had proposesd to circulate a letter to copyright holders and ask whether they would accept Geant4 release under public release, GNU Public Licence (GPL) or have a prefered some other form of licence. This letter remains to be drafted.

The CB has drafted and circulated (for comment) a disclaimer for warranty. However, a response is being awaited from some institutes. The current time-scales prohibit the planned inclusion of a warranty disclaimer in the next (minor release), and it was considered essential to have one by the late June 29^th release of Geant4.

Action 11.1

It is anticipated that the final papers on the Comparisons Project will be completed in Summer (July to September 2001).

The draft report covering the software process milestone will be sent to the CB at the end of this week (any comments should be forwarded to G Cosmo).

The CB has accepted the training kit milestone as fulfilled. It was reported that the CB had identified(The TSB chair noted the need to make the training kits more accessible on the web-site, and state that they’re primarily intended for use by in lectures by people with Geant4 expertise.)

The CB had received two proposals for new members from universities at Helsinki and Barcelona. A final decision has not been reached by the CB. The TSB strongly encourage a decision to be taken soon, in order for the collaboration to meet programme deadlines. TSB eagerly looks forward to a decision.

Notes from the category coordinators meeting have been published at the Geant4 web-site and were circulated and discussed at the TSB.

J Allison reported on the status of the General Paper. The timetable is currently:

• 4^th April - Second draft contributions submitted
• 11^th April – Editors required to provide feedback
• 24^th April – Third consolidated draft
• 3^rd May – Final draft

The general comments about the first draft are:

• It is somewhat repetitious showing sections to have been written in isolation;
• There is a need to consider the target audience, which will influence the content and style;
• Wherever possible, concrete examples to showcase Geant4 should be introduced;
• The first three chapters should provide an overview and background, whilst the subsequent chapters should provide the detail.

J Knobloch asked whether copies of the sections could be provided in PostScript and/or PDF format. This was agreed

Action 11.2

J Apostolakis emphasised that the paper should be strong on the design and structure of Geant4 (which is a topic that was unlikely to be covered in similar detail in future publications). He also requested contributions covering Chapter 8, and in particular, if members of the collaboration were writing other papers on Geant4 applications, to provide1-2 paragraph summaries should be sent to him. Similarly, contributions were invited from those who had or would be conducting extensive runs of Geant4, which would reveal the robustness of Geant4.

Action 11.3

J Apostolakis stated that there was the potential for duplication between the "comparisons" subsection of Chapter 8 and comparisons with experiment in Chapter 5 (Processes). It was agreed that it was still valuable to have the subsection in Chapter 8, since it would cover the methodology for comparisons and present results where there are several combined processes involved.

The following alterations to the organisation of the paper were agreed:

• There would be a geometry subsection, and a material and particle subsection in Chapter 2.
• Handling of stack track and pile-up should be discussed.
• The last subsection in Chapter 2 should be removed and Chapter 8 should discuss how to build a Geant4 simulation.

Nuclear Instruments and Methods in Physical Science (NIM) was identified as the preferred target journal for the General Paper, whilst Computer Physics Communications was a possible alternative if the paper was not accepted by NIM.

MG Pia requested noted that the results from experiments or other groups on comparisons between data and simulation should be referenced from published papers, since the Geant4 Collaboration could not publish somebody else's data (unless properly referenced). J Apostolakis commented that approval to publish data from experiments has been acquired or was currently being would be sought, wherever required.

Although the majority of the contributions are present, the Activity Report is not in as advanced state as was hoped.

Action 11.4

Action 11.5

It is intended that the editors will generate feedback on the draft Activity Report in the first two weeks in April.

At the Category Coordinators meeting the following time-scales for Geant4 release were agreed:

• Geant4 3.1 – 6^th April 2001. This release is intended to be Minor release, between scheduled releases.
• Geant4 3.2 or 4.0 – 29^th June 2001 (release period to start 18^th May, with subsequent release deadlines 25^th May and 1^st June). This release shall include a disclaimer of warranty. The status (major or minor) will be confirmed closer to the date.
• Geant4 4.0 or 5.0 – 14^th December 2001. It is anticipated that this will be a major release, and there will be experiments dependent upon changes to the toolkit in this release.

A discussion was held as to whether it should be possible to change the behaviour of classes of Geant4 for minor releases, where traditionally only bug-fixes are permitted. Associated with this issue is the need to (and complication of) distinguishing between bug-fix tags and development tags, potentially using a flag for cases where a bug-fix had been made. There is also the need for supporting branches in the development process and the associated manpower overhead of having to effect changes to two (or more) versions. It was agreed that a concrete procedure should be developed on how to effect such modifications in minor releases, and this should be presented and discussed at the next workshop.

Action 11.6

It was agreed that it should be possible to update the documentation for any scheduled release, whether major or minor.

Action 11.7

G Folger presented the results of the survey of user platforms of Geant4 parties. The most popular configurations are PC/Linux and Sun/Solaris. There is however a modest-size DEC user community and a requirement for PC Windows systems. INFN require support for the next 1-2 years for DEC, but ultimately may dispose of DEC machines thereafter. The GLAST collaboratortion requires/ operates its applications software on Windows 2000 and Visual C++, as a result of collaborations with NASA. In addition, LHCb required Geant4 operation on Windows/NT. IN2P3 is also a DEC/Geant4 user, as is SLAC, which uses Geant4 with ObjectSpace STL..

With the anticipated disposal shutdown of DEC and HP machines at CERN by the end of 2002, it was agreed that partners with particular platform requirements should share the responsibility for porting and full system testing.

Action 11.8

The process of how testing and modification should be undertaken remains to be decided:

• Category groups could be allowed access to the machines to debug software;
• Feedback could be provided by the testing team of bugs and bug-fixes, and the category coordinators assess and approve the fix.

Concern was raised about the potential additional workload on the Category Coordinators, especially considering the limited human resources presently available.

Following discussion it was proposed to discontinue testing and support of Geant4 on HP platforms after at the next scheduled release after the June 2001 release. The release of December 2001 will not be supported on HP.

, and possibly discontinue teTesting on DEC machines would continue at CERN until at the end of the year at least. The site to which testing of DEC will migrate during 2002 must be identified.

It was agreed that coding standards will continue to be followed to ensure that, if at all possible, no action would be taken to deliberately or otherwise preclude any particular platform that is not specifically supported.

Action 11.9

Action 11.10

J Apostolakis stated that he was yet to receive a complete set of objectives from all working groups.

Action 11.11

It was agreed that the paper covering the architecture of Geant4 will be written by a task of the Architecture working group when it if formed.

It is anticipated that activities for development of the training kits this year could would include:

• Speakers’ notes to explain figures and schemes where general knowledge was insufficient.
• A video or recorded audio lesson synchronised with slide projector (this was an informal objective since no particular partner had requested this).
• Improved tools for maintaining training kits and exercises.
• Improved infrastructure for developing documentation, and increased linkages between different parts of the documentation.

A video or recorded audio lesson synchronised with slide projector was agreed as an optional objective.

It was agreed to forward to the CB a job description for a person to work on the structure of the training kits, the Web and the user documentation and potentially on tools to integrate some of their aspects - in order to ease in their maintenance.

A target of October 2001 was set for a review of the Geant4 user documentation.

Action 11.12

One of the objectives of 2001 will be to explore new comparison projects, especially those which that will stress the areas of physics which have no't been tested significantly up to now in comparison projects. Proposals were invited from the member groups on these, as well as new applications, but it was stressed that the timescales for conducting this exercise should be realistic to ensure that reasonable milestones are identified.

Action 11.13

The intention of the regression suite is to minimise the manual work required for base-lining outputs from comparisons used to assess the impact of changes on performance. The Processes Groups were invited to identify concrete examples which would be used for the suite, and to assess the magnitude of the task to implement these into the suite. To kick this activity off, J Apostolakis agreed to define better the requirements from the processes groups in order to support the development of the suite.

The need to create an infrastructure for the creation of the regression suite was noted. Resources to achieve theseis sub-objectives will be discussed at a later TSB.

Action 11.14

The requirement for specifying the lifetime for each instance of the B-meson was understood, and code modifications have been implemented and were in the process of being tested.

There was aA new requirement for Geant4 to handle "optical" processes of X-rays was expressed on behalf of ESA.

Action 11.15

It was requested the HyperNews system cover the following fora: General matters; Experimental setup; Physics; Control of runs, events, tracks, particles; Interfaces. There was concern that over the amount of work required for dealing with HyperNews queries could become problematic.

The requirement for traceability from user requirements to the design, implementation and tests was discussed at the category coordinators meeting. It was anticipated that one month of effort could be required for some cases to complete this task. Category coordinators were invited to submit estimates of the effort required for their specific categories.

Action 11.16

(This action is not relevant to the lowE EM Working Group, which has completed the software traceability task.)

J Apostolakis reviewed the terms of reference of the Architecture subgroup:

• The subgroup will be responsible for maintaining the global category diagram and the Geant4 URD;
• It would bring recommendations for updates, proposed by category coordinators, to the TSB;
• It will be responsible for producing the Geant4 Architecture paper (deadline Autumn 2001).

J Apostolakis stated that he wouldproposed to announce the membership of the subgroup by the end of the month once it had been finalised.

The date of the Geant4 Workshop in Genova was confirmed as 2^nd – 6^th July. The current members of the programme committee were MG Pia and M Schroder. It was requested that a relatively good draft of the programme be produced and circulated by the end of April.

Action 11.17

The following dates were identified for subsequent TSB meetings:

• End of May to mid June
• End of September to mid October
• Mid to late October

Action 11.18

An application of three students (Elena Guradincerri, Simeone Dussoni and Alfonso Mantero) made by M G Pia was discussed by the TSB. No objections were raised to their joining the Geant4 collaboration as part of the Low-Energy Electromagnetic Physics WG.

Similarly the applications of Dr Kouiti Murakami (Kyoto University), and Rachid Marzini (a PhD student at the University of Montreal) were accepted (subject to CB approval). It is intended that K Murakami join the collaboration as part of the UI working group.

A proposal for monitoring Geant4 downloads from the web-site was discussed. It was decided that some explanation would be required on the user registration page as to why details were being collected, and how they would be used. E-mail details of the user should be optional (e.g. if they wanted to submit their name to a mailing list).

The establishment of this facility was generally supported, subject to the availability of resources.

Action 11.19

It was reported that ESA plan to issue an ITT covering the requirements and development of radiation tools (expected to be based on Geant4) in support of ESA science missions. Proposals will be invited and subject to open competition for a prime contractor (with 60% of the 400kEuro budget) to run the programme of work. Part of that programme will involve smaller Announcements of Opportunity (envisaged to cost 10-50 kEuro each) for research groups to undertake specific studies.