Attendees: David L., Elliott W. Sandy P., Simaon T., Richard J., Ryan M.
Curtis M., Franz K.

Agenda:

1. Make Systems: handling platform dependancies

2. Documenting code: Root-THML, Doxygen, ...
 (see http://www.jlab.org/~davidl/HallD/src_doc/)

3. Hall-D analysis framework proposal (see URL in 2.)

4. Code Repositories

5. Cherenkov update.

-------------------------------------------------------

1. A discussion of make systems occured. We are now running into places
where there are platform dependant features which must be handled in
the code somehow. The specific examples are: a) basename being defined
in string.h on Linux, but not on OSX  and b) xdr_longlong_t not being
defined on OSX. The discussion focused some on what policy we should
adopt to handle the inevitable platform dependancies so that we don't
end up with multiple approaches sprinkled throughout the code.

There seems to be at least some experience with automake and imake
by Richard and Franz. David descibed a few features in the BMS system
he's been writing for Hall-D based on gmake. David agreed to look into 
automake a little closer.

2. David showed a website generated by doxygen of the code he wrote
as the basis of the analysis framework (see 3.). Richard showed a page
he'd generated with the ROOT class THTML. Richard reported that he
had no problem with using THTML, but David reported running into some
problems with it. Elliott suggested there are a lot of "literate 
programming" tools out there that could be looked into as well. Elliott
"volunteered" to look into these and give a report at the next meeting.

3. David briefly discussed an anlysis framework for Hall-D that he's
developing. To begin with, the system will only implement HDDM from
files, but is designed to easily add additional media and formats for
input data. Everyone was encouraged to take a look at what's there and
think critically about what is needed in such a system.

4. Elliott reported that a near final draft of the document discussing
BitKeeper as an alternative to CVS is complete. He will incorporate 
forthcoming comments from persons reviewing the current draft and then
post it as a Hall-D note..

5. Ryan reported that work on the Cherenkov is progressing. He has 
gathered all of the most current information on it and it appears
to be in a state of flux. He will continue to work on the design
with monte carlo studies.


There was strong interest expressed to discuss what's needed for
simulation in the short term. Specifically, goals for the detector
review. It was clear we would need Ed to coordinate this discussion.