Level 1 Calo Software Meeting at QMW on Thursday 12 November 1998.
------------------------------------------------------------------

Present: Eric, Murrough, Norman, Tara.


We started with a general overview of our present status. We will shortly
have (part of) about eight people in the UK. (And a few in the rest of level1
calo community). Meanwhile we have a reasonable time window before we need
software for new prototype modules (excluding present Stockholm/Mainz and
Heidelberg demonstrators).

We do need to define a plan and timescale for our future software work.
We also need to consider:
- the organisation of the work
- what software process and tools we want to use
- how we coordinate within the level1 calo community
- whether and how we might coordinate more widely in level1

On the software process and tools issue, the Atlas offline group have defined
their "Atlas Software Process" and use the Software through Pictures (StP)
commercial tool to manage the requirements, analysis and design phases of
their project. The "Back End DAQ" part of the DAQ prototype -1 project used
it as well, though the DAQ Dataflow group used less formal design methods
and the level2 "lightweight DAQ" project used the Rational Rose commercial
tool instead of StP.

For managing the source code and binary releases, the use of CVS and SRT seem
to be becoming standard in Atlas.

Tara reported positive experience of using CVS. We agreed that we all need
to learn more about all these tools. Some of (eg StP) may seem a bit "heavy"
for our needs, but we also have to bear in mind the need for maintainable
and current documentation describing our software for the sake of the
inevitable flux of new people over the lifetime of the project.


We also briefly discussed our interaction with the DAQ group. Murrough is
planning to visit Livio & Bob Jones after the level 1 meeting in Heidelberg.
It looks as though we wont have much of a workplan by then, but we have at
least made a start.

The DAQ prototype software is quite extensive with many components and uses
many underlying packages. We will need to become familiar with a lot of new
things here as well. Perhaps at the next meeting we can divide up the DAQ
components and each try to become familiar with some of them. This seemed
too much to do _before_ our next meeting.


The next major step will be trying to define our software requirements. The
software backup document for the review panel in March outlines some of our
requirements in a few major categories. We agreed that we would divide up
these categories and each try to define more detailed requirements before
our next meeting. Murrough will act as the secretary collecting text for a new
requirements document. The major headings are:

(a) Test & Diagnostics:      Murrough
(b) DAQ software:            (leave till later)
(c) Calibration:             Tara
(d) Control software:        Norman
(e) Monitoring:              Eric
    - technical
    - "physics"
(f) Offline software:        (leave till later)


We also realised that we need to have a serious discussion about the (hardware)
architecture of the system. Specifically the division into readout crates,
locations of CPUs. This affects where we are able to carry out the necessary
software tasks such as monitoring, calibration etc.


We agreed that a small software group, at first probably a purely UK group,
should meeting about monthly, in between the main level1 UK meetings. The
AT1SOFT@LISTSERV@RL.AC.UK mailing list should be updated to included (only)
the interested people.

Our next meeting will be on Monday 14 december (just before Norman returns
to Heidelberg to discuss RODs with Cornelius). A draft agenda for that
meeting will look something like:

(1) Overview of DAQ -1 prototype............Murrough
(2) Reports on software tools
    (a) CVS.................................Tara
    (b) StP.................................Steve Fisher (we need to ask him)
    (c) SRT.................................Murrough/Tara?
    (d) Discussion
(3) Requirements doc........................Murrough
(4) Architecture............................Norman


Before that meeting we will have a short spot (30 mins?) at each of the
level1 UK and main Heidelberg meeting to discuss software and present our
initial discussions to our colleagues.