Dear Bramin,
I have been thinking about paper committees and various
things. Here are my confused ramblings that might be useful as a starting point
for discussion. If I disparage any individual or collaboration my apologies. I
have made lots of mistakes in the past. This file is also attached
as a word document.
1) We have an obligation and need to publish
a. The obligation is because we owe it to the taxpayers, our students, our
community and ourselves.
b. The need is because we will disappear if we don't
2) The "speed" of publication and disemmination is faster than at the AGS &
SPS. Both conferences and the preprint server mean that hot new results become
"old news" more rapidly than in the past.
a. This causes a tension between getting something "preliminary" ready for
a conference and slogging through the systematic errors, (and the correlation
matrix) to get results fit for publication.
b. Thus we have to strike a balance between no discussion before
publication (PHOBOS) and perpetually preliminary (NA49).
3) Our "brand" name is vital, it is important to keep our reputation for
producing reliable results.
a. The main topic of our text and figures is always the data. We should
not overload our papers with models that may not last. NA38-50 springs to mind
as a group that was always discovering the QGP.
b. Related to this PRLs are nice but on occasion we will need to publish
longer papers with the full exposition of the data and analysis.
4) We have to make other people want to use our data and make it easily
accessible. This means routinely using EPAPs to give a direct hyperlink from
our papers to our data as ascii tables, root macros and plots. We should also
have all of this on our own web site.
I was a member of the ratios paper committee and felt that it did not work very
well. One reason for this was that we only met in person once. Trying to polish
a text via email is very difficult, unless done by a small group very
intensively. Phone conferencing might have helped. Two way calls are themselves
very useful. We could make committees that are either exclusively European or
American. This would make it easier to meet but has its own dangers.
Traditionally large collaborations have had paper committees to manage
conflict, or as the Marxists might say producing synthesis out of thesis and
antithesis. We need the conflict to strip away all that shaky in our analysis or
redundant in our text. We need management to prevent the conflict ripping the
collaboration apart. For the ratios paper this process was not held within the
paper committee but was broad-cast to the whole collaboration. This began out of
convenience because Romero asked to be included in the physics discussion and we
didn't have a list server for the paper committee. However an unfortunate aspect
of this was that the paper committee itself never "approved" a draft to be sent
to the collaboration as a whole. Eventually it seemed to me that the paper
committee, as a committee, was irrelevant and that we interacted with the main
authors as individuals.
This brings up the question of weather BRAHMS is large enough to warrant paper
committees at all. Certainly we do not use them at the cyclotron and getting rid
of them has the advantage that no one is excluded from any stage of the
discussion. However I feel they offer some advantages
1) They offer a counterweight to the prejudice of the main author. Perhaps
we should think of them as "juries" whom the main author needs to convince.
2) It is their responsibility to critique the paper. Everyone is busy so if
its not explicity your job to review a draft there is always the temptation to
let someone else do the work.
3) It helps to have a devils advocate. Jack Sandweiss, while an editor on
PRL, said that the only group that consistently gets more than 1/3rd of their
submissions to PRL accepted are the big high energy collaborations which have
very strict internal quality controls on drafts.
4) Language counts. An APS poster at Nantes said that the most useful
thing you can do to get your paper accepted is to have it proof-read three times
by a native English speaker.
My experience with the ZDCs gave me an inside look at the preparation of the
STAR ultraperipheral paper. A collegue at the cyclotron was put on the
"god-father" committee of this paper precisely because he was not an expert.
However he asked 3 questions and the paper was not presented to the
collaboration until his questions were answered. The effort to answer his
questions led to many improvements and this paper was very quickly accepted by
PRL.
Since BRAHMS is not very large all of the above issues can probably be handled
without committees. However I would recommend strengthening them with a clear
mandate to verfity and vouch for both the final analysis and the text.
Yours Sincerely,
Michael
Michael Murray, Cyclotron TAMU, 979 845 1411 x 273, Fax 1899
This archive was generated by hypermail 2b30 : Fri Aug 09 2002 - 16:59:44 EDT