summary of video meeting 9/18-9/19

From: Dave Casper (dcasper@uci.edu)
Date: Sat Sep 22 2001 - 03:36:43 JST

  • Next message: Giampaolo Co: "(no subject)"

    From: "Dave Casper" <dcasper@uci.edu>
    To: <opc.nuint01@neutrino.kek.jp>
    Subject: summary of video meeting 9/18-9/19
    Date: Fri, 21 Sep 2001 11:36:43 -0700
    Message-ID: <001101c142cc$5d4c2f00$6401a8c0@LAGUNA>
    
    
    
    Participants: Y.Hayato, M.Sakuda, R.Seki, D.Casper
    
    A video conference was held (after some technical difficulties) to
    discuss the Monte Carlo session of the workshop.
    
    Dave sent an invitation to all known neutrino MC experts the day before
    the meeting, soliciting their interest in producing samples of events in
    a common format to allow detailed comparisons of the existing programs.
    
    Hayato-san was not able to review the mail prior to the meeting due to
    computer problems, but expressed his opinion that special emphasis
    should be placed on quasi-elastic reactions and comparison to
    theoretical calculations.
    
    Dave tried to make clear that the comparison of event samples was not a
    replacement for the sorts of benchmark calculations proposed by
    Seki-san, but rather an addition to it.  Whereas one can compare
    relatively well-understood processes like QE scattering with theory,
    there are many aspects of the physics for which there is no theory and
    hence the only possible comparison is with other Monte Carlos.
    
    The question of the organization of speakers was raised by Seki-san.
    Dave felt that the presently proposed agenda seemed fine and that no
    major reorganization of the agenda was being proposed.  Instead, the
    more detailed comparisons will perhaps allow the experts to meet in a
    smaller session during the workshop to discuss the results.  Dave also
    said that QE scattering is in some sense a fairly limited topic and it
    might be difficult to have an hour-long panel discussion about this
    topic alone.  Since the panel discussion is the first (ever) opportunity
    for specialists to discuss the challenges of simulating neutrino
    physics, it may be more appropriate to discuss a full range of topics -
    QE, final-state interactions, transition from resonant to
    deep-inelastic, future work on a universal generator, etc.
    
    At the end of the meeting, Dave asked Hayato-san what he thought the
    SuperK collaboration's attitude would be toward:
    
    1) Making Neut event samples public, for comparison with other
    generators, and
    
    2) Making the Neut source code public, for comparison with other
    generators.
    
    Hayato-san said he would make inquiries about these questions.
    
    The conclusions of the meeting were (I think):
    
    1) The present draft agenda will not be changed significantly; perhaps
    one talk summarizing what has been learned by the event-sample
    comparison and future plans of the event-generator working group which
    we hope to establish.
    
    2) We will await some feedback and response from the experts and tailor
    the event-sample comparison to the wishes of the participants.
    
    3) The program of benchmark plots will be refined in the next few weeks.
    
    I have so far received two extremely positive responses to my
    invitation, one from Flavio Cavanna (Icarus), who will attend, and one
    from Hugh Gallagher (Soudan/Minos) who is unable to attend for family
    reasons but willing to participate in the comparison.  I attach these
    responses.
    
    
    Dave
    dcasper@uci.edu 
    
    
    

    attached mail follows:


    From: "Flavio Cavanna" <flavio.cavanna@aquila.infn.it>
    To: <dcasper@uci.edu>
    Subject: Re: NUINT'01 Neutrino Monte Carlo Session
    Date: Tue, 18 Sep 2001 12:57:20 -0700
    Message-ID: <3BA7A707.D2DC4722@aquila.infn.it>
    
    Hi Dave,
    I'm very happy about hearing from you and about your proposal for 
    having at some point a common, well established MC generator 
    a-la-LUND. Since my early days in ICARUS (already ten years ago..)
    I was "frustrated by the necessity of writing my own generators 
    from scratch" and I was dreaming about a nu-LUND MC!!
    Therfore I'm fully sharing with you the the leading ideas about 
    the organization of the NUINT01 ws.
    This initial e-mail is just to show up with my interest in it 
    and to say that I'll quickly try to collect and send you
    information about what 
    I and some other people have done about our own MC generator.
    The exercise of producing n-tuples with generated events
    is very good and we can easily produce them according to any format/
    structure...  
    I'll send you the information in a few days.
    
               best,  flavio
    
    ps. I'd like to ask you to include the name of my colleague (icarus)
    Ornella.Palamara@lngs.infn.it in the list of interested people.
    She wrote a large fraction of our MC code 
    (we call it: GENEVE - GEnerator of Neutrino EVEnts) and
    both of us are going to come to NUINT01.  
    
    
    Dave Casper wrote:
    > 
    >         Hi,
    > 
    > This mail is an initial contact concerning the Monte Carlo session of
    > the NUINT'01 workshop to be held December 13-16 at KEK (see
    > http://neutrino.kek.jp/nuint01/ for additional information on the
    > workshop).  You are receiving this because you have been identified as
    > the principal author or maintainer of a neutrino event generator used
    by
    > one or more past, present or future experiments.
    > 
    > If you have grappled with the uncertainties and frustrations of
    > simulating neutrino physics, I hope you are as excited as I am about
    > this opportunity to come together, discuss and compare notes with
    others
    > facing the same challenges.  Moreover, we expect that NUINT'01 will be
    > the first of a series of annual meetings, and we hope to create
    working
    > groups which will remain in contact over the years.  Planned or
    proposed
    > precision measurements with neutrinos will require a much-improved
    > understanding and description of neutrino interactions, so our work
    may
    > benefit the field for years to come.  My own hope is that some day a
    > LUND-like program for neutrino physics can be created, sparing future
    > experiments the necessity of writing their own generators from scratch
    > (or at least giving them a widely-available program to compare with)
    and
    > eventually allowing improvements to be immediately available to
    > everyone.  It is difficult to imagine the situation where every e+e-
    and
    > hadron collider experiment for the past 10 years would have had to
    > create their own version of LUND - but this is precisely the present
    > situation in neutrino physics.
    > 
    > The preliminary plan for Monte Carlo session is two-fold.  One part,
    > being organized by Seki-san (seki@neutrino.kek.jp), will focus on
    > comparing neutrino generators with theoretical calculations.  Seki-san
    > is preparing a list of "benchmark" plots which participants will be
    > asked to show at the conference.  The second part, which I have
    > volunteered to organize, is perhaps more ambitious but also perhaps
    more
    > interesting to specialists.  The idea is to generate samples of
    > simulated events at several energies and produce Ntuples which can be
    > compared in a more open-ended way.  The motivation is that any
    > pre-determined set of plots for comparison with theoretical
    calculations
    > is by definition limited to fairly simple quantities, and may miss
    > important differences and features of the various generators.  In a
    > sense, the event sample comparison is more for our benefit as
    > specialists, in that we will be able to investigate whatever aspects
    of
    > the physics are of most interest.  A 0'th-order example of what this
    > might entail would be producing samples of, say, 10K events, neutrino
    > and anti-neutrino, at energies of 0.5, 1, 2 and 3 GeV.  But of course
    we
    > have several months to discuss what is both interesting and possible
    to
    > do and arrive at a final program.  Final state interactions in nucleii
    > is another area where comparisons may be helpful.  Many, hopefully
    all,
    > of us plan to attend the Workshop, but those unable to attend can
    still
    > contribute by participating in this study.
    > 
    > A good first step will be simply to confirm your interest in this
    work,
    > and perhaps to send a brief summary of your program (name, region of
    > validity, approximations and models used, output format, etc) and any
    > other comments you might have.  I will create a web page with the
    > information received which can be updated as we make progress.  Please
    > send your response to dcasper@uci.edu.
    > 
    > Some issues which should be addressed between now and the workshop
    > include:
    > 
    > 1) Choice of samples to generate and compare:
    > As above.  In addition, the question of the target needs to be
    > considered.  The known programs were originally designed
    > 
    > 2) Creation of a uniform Ntuple format for comparisons:
    > It is unlikely that the output of half a dozen different programs will
    > be compatible without additional work.  Nevertheless, the output of
    any
    > true generator must, in the end, be a list of particles and momentum
    > vectors, hence there is no reason in principle why translation to a
    > common format should be difficult.
    > 
    > 3) "Proprietary" status of the event samples:
    > We wish to compare as openly as possible, but some experiments may
    have
    > reservations about making the results of their generators public in a
    > wider sense.  In particular, publications using the samples provided
    for
    > this comparison should, I think, be agreed in advance with the person
    > who provided the sample.
    > 
    > 4) Are we missing anybody?
    > The present list of known MC experts (who are receiving this mail) is:
    > 
    > Andre Rubbia (Nomad/Icarus)
    > Dave Casper (SuperK/K2K)
    > Flavio Cavanna (Icarus)
    > Geoff Mills (MiniBoone)
    > Yoshinari Hayato (SuperK/K2K)
    > Hugh Gallagher (Soudan-II,Minos)
    > Paolo Lipari (Macro)
    > Piero Zucchelli (Chorus)
    > 
    > Looking forward to your responses, and to the upcoming meeting.
    > 
    > Dave
    > dcasper@uci.edu
    
    
    
    

    attached mail follows:


    From: "Hugh Gallagher" <gallag@hep.umn.edu>
    To: <dcasper@uci.edu>
    Subject: Re: NUINT'01 Neutrino Monte Carlo Session
    Date: Wed, 19 Sep 2001 15:08:08 -0700
    Message-ID: <0f1c01c14157$90b31460$80de6580@hep.umn.edu>
    
    Hi Dave,
    
    I am quite excited by your e-mail, I know that in the past our generator
    (NEUGEN) has undergone real improvements as the result of these kinds of
    discussions, and I expect the same will result from this meeting.  As
    Jorge
    may have informed you, I will not be able to attend the meeting as my
    wife
    is due to give birth Dec. 21, however I would like to participate in any
    other way I can.
    
    I think your e-mail touches on a lot of important topics and is asking
    all
    the right questions.  I was particularly heartened to hear that this is
    envisioned as being the first in a series of annual meetings.
    
    One point in your e-mail in particular struck a chord with me and that
    is
    the desire to develop a standard generator for neutrino physics.  I
    think
    that this is a very realizable goal at this point in time, particular as
    I
    suspect that other generators are, like NEUGEN, undergoing re-writes in
    C++
    to support future OO-based simulations.   I think the time is right to
    discuss organizing this kind of activity.
    
    I will give some thought as to the best way for me to contribute to the
    meeting from a distance.   I am sure there are more things to discuss
    but I
    wanted to start off by letting you know that I am quite enthused about
    this
    meeting and the activities you have outlined for it.
    
    Cheers,
    Hugh
    
    ----- Original Message -----
    From: Dave Casper <dcasper@uci.edu>
    To: Piero ZUCCHELLI <Piero.Zucchelli@cern.ch>; Andre Rubbia
    <rubbia@particle.phys.ethz.ch>; Dave Casper <dcasper@uci.edu>; Flavio
    Cavanna <flavio.cavanna@aquila.infn.it>; Geoff Mills <mills@lanl.gov>;
    Hayato@Neutrino. Kek. Jp <hayato@neutrino.kek.jp>; Hugh Gallagher
    <gallag@hep.umn.edu>; Paolo Lipari <lipari@suketto.icrr.u-tokyo.ac.jp>
    Cc: Makoto Sakuda <sakuda@neutrino.kek.jp>; Jorge_G_Morfin
    <morfin@fnal.gov>; <seki@neutrino.kek.jp>
    Sent: Monday, September 17, 2001 1:29 PM
    Subject: NUINT'01 Neutrino Monte Carlo Session
    
    
    >
    > Hi,
    >
    > This mail is an initial contact concerning the Monte Carlo session of
    > the NUINT'01 workshop to be held December 13-16 at KEK (see
    > http://neutrino.kek.jp/nuint01/ for additional information on the
    > workshop).  You are receiving this because you have been identified as
    > the principal author or maintainer of a neutrino event generator used
    by
    > one or more past, present or future experiments.
    >
    > If you have grappled with the uncertainties and frustrations of
    > simulating neutrino physics, I hope you are as excited as I am about
    > this opportunity to come together, discuss and compare notes with
    others
    > facing the same challenges.  Moreover, we expect that NUINT'01 will be
    > the first of a series of annual meetings, and we hope to create
    working
    > groups which will remain in contact over the years.  Planned or
    proposed
    > precision measurements with neutrinos will require a much-improved
    > understanding and description of neutrino interactions, so our work
    may
    > benefit the field for years to come.  My own hope is that some day a
    > LUND-like program for neutrino physics can be created, sparing future
    > experiments the necessity of writing their own generators from scratch
    > (or at least giving them a widely-available program to compare with)
    and
    > eventually allowing improvements to be immediately available to
    > everyone.  It is difficult to imagine the situation where every e+e-
    and
    > hadron collider experiment for the past 10 years would have had to
    > create their own version of LUND - but this is precisely the present
    > situation in neutrino physics.
    >
    > The preliminary plan for Monte Carlo session is two-fold.  One part,
    > being organized by Seki-san (seki@neutrino.kek.jp), will focus on
    > comparing neutrino generators with theoretical calculations.  Seki-san
    > is preparing a list of "benchmark" plots which participants will be
    > asked to show at the conference.  The second part, which I have
    > volunteered to organize, is perhaps more ambitious but also perhaps
    more
    > interesting to specialists.  The idea is to generate samples of
    > simulated events at several energies and produce Ntuples which can be
    > compared in a more open-ended way.  The motivation is that any
    > pre-determined set of plots for comparison with theoretical
    calculations
    > is by definition limited to fairly simple quantities, and may miss
    > important differences and features of the various generators.  In a
    > sense, the event sample comparison is more for our benefit as
    > specialists, in that we will be able to investigate whatever aspects
    of
    > the physics are of most interest.  A 0'th-order example of what this
    > might entail would be producing samples of, say, 10K events, neutrino
    > and anti-neutrino, at energies of 0.5, 1, 2 and 3 GeV.  But of course
    we
    > have several months to discuss what is both interesting and possible
    to
    > do and arrive at a final program.  Final state interactions in nucleii
    > is another area where comparisons may be helpful.  Many, hopefully
    all,
    > of us plan to attend the Workshop, but those unable to attend can
    still
    > contribute by participating in this study.
    >
    > A good first step will be simply to confirm your interest in this
    work,
    > and perhaps to send a brief summary of your program (name, region of
    > validity, approximations and models used, output format, etc) and any
    > other comments you might have.  I will create a web page with the
    > information received which can be updated as we make progress.  Please
    > send your response to dcasper@uci.edu.
    >
    > Some issues which should be addressed between now and the workshop
    > include:
    >
    > 1) Choice of samples to generate and compare:
    > As above.  In addition, the question of the target needs to be
    > considered.  The known programs were originally designed
    >
    > 2) Creation of a uniform Ntuple format for comparisons:
    > It is unlikely that the output of half a dozen different programs will
    > be compatible without additional work.  Nevertheless, the output of
    any
    > true generator must, in the end, be a list of particles and momentum
    > vectors, hence there is no reason in principle why translation to a
    > common format should be difficult.
    >
    > 3) "Proprietary" status of the event samples:
    > We wish to compare as openly as possible, but some experiments may
    have
    > reservations about making the results of their generators public in a
    > wider sense.  In particular, publications using the samples provided
    for
    > this comparison should, I think, be agreed in advance with the person
    > who provided the sample.
    >
    > 4) Are we missing anybody?
    > The present list of known MC experts (who are receiving this mail) is:
    >
    > Andre Rubbia (Nomad/Icarus)
    > Dave Casper (SuperK/K2K)
    > Flavio Cavanna (Icarus)
    > Geoff Mills (MiniBoone)
    > Yoshinari Hayato (SuperK/K2K)
    > Hugh Gallagher (Soudan-II,Minos)
    > Paolo Lipari (Macro)
    > Piero Zucchelli (Chorus)
    >
    > Looking forward to your responses, and to the upcoming meeting.
    >
    > Dave
    > dcasper@uci.edu
    >
    >
    



    This archive was generated by hypermail 2.1.0 : Sat Sep 22 2001 - 03:36:57 JST