[Rivet] Validation meeting

[Andy Buckley]

Andy Buckley wrote:

Good meeting, y'all. Action time:

 * All (?): I forgot to mention, but there's a Genser steering meeting
tomorrow at 0830 (!) GMT / 0930 CET. Since I think there may have been
an unaccountable executive decision made to push HepMCAnalysisTool, some
Rivet/MCnet presence would be good. Given the badness of the EVO
connection last time, it would be particularly good if someone happens
to be at CERN and in the room!
 * Andy and Frank: get validation system in a state where it can produce
basic plots and Web pages. To be improved while at CERN.
 * James: run Rivet (1.1.1?) in Athena with Pythia 6 S0-Pro tuning on
CDF_2001 and CDF_2008_NOTE (Drell-Yan UE). I've sent the S0-Pro tune
params around by email. Phone me if you're having any trouble. To be
presented as part of Andy's Atlas MC talk on Wednesday 3rd Dec.
 * Andy: make standard tune/gen comparison plots for MC group talk.
 * Andy: remangle the binreloc symbols in Rivet and AGILe to be
different from each other and HepMC.
 * Andy and James: meet at UCL with project students on Friday
(tomorrow) at ~3pm to do Rivet things / identify start-up difficulties.

"Minutes" below:

>>>  * How soon can we release a new Rivet/AGILe version? TODO: big
>>> documentation push (manual, wiki, Doxygen, more analysis class
>>> metadata), validation of W+jets & other analyses to be used for YETI.
>>> Move to using new command-line interfaces as standard, with rivetgun as
>>> deprecated but not removed?

We would like to release very soon, probably shortly after the Atlas
week. The prerequisites for release are

 * Disable Python interface by default for now: needs to build out of
the box.
 * Get validation system running so we can manually okay the W+jets
analysis (+ another?) for Sherpa, FHerwig and Pythia 6. Put plots online.
 * Improve documentation (undiscussed, but very true and important!)

>>>  * Status of validation system: how soon can we think about making a
>>> (static, initially) Web gallery of standard gen/tune comparisons?

Yes. The system is now at the stage where we can generate runs which
correspond to Rivet analyses in a pretty simple way. Frank and I will
work today on getting the output analysis script working and generating
the plot web pages that we want.

>>>  * Aims for next week's working meeting.

To miss the MC and SM meetings, I suggest that we meet at 10am on
Wednesday morning. Location: UCL Atlas office (01-01-035)? Jon/James can
comment on whether this okay or if we should find somewhere else, e.g.
the other UCL Atlas office whose code I don't know.

Aim to get our sets of validation plots up and running and to have a
clear path to a release in the following few days.

>> Three more from me, while I remember:
>>
>>  * Next gens for Prof tuning

As mentioned in the meeting: the 5 core MCnet gens are our next set on
the hitlist. We will finish the Py6 pT/interleaved tune to Tevatron
data, write it up, and then progress to Herwigs, Py8 and Sherpa. In that
order?

>>  * Suggested next data for Rivet: HERA, UA5, ISR, STAR...

Leave this for now: get visible plots/tunes from what we've got before
extending further.

>>  * More gens needing interfacing: PHOJET... others?

I will need PHOJET for personal UE studies if I want to be able to
compare to all the current Atlas MB/UE tunes. But it's not one to be put
through the Professor machinery.

Best wishes - see you tomorrow/next week,
Andy

-- 
Dr Andy Buckley
Institute for Particle Physics Phenomenology
Durham University
0191 3343798 | 0191 3732613 | www.insectnation.org