Dear Vera, dear Colleagues
when our group was established Vera and I discussed whether
I should join because my point of view, as expressed already
at the initial meeting at Beijing in August 04, is certainly
rather conservative, traditional given the changes which are now
envisaged in the draft of the working group (3/22/2005). We agreed
at that time that I would stay and provide the information
from DESY, as was also agreed with the ZEUS spokesman, R.Yoshida.
Having read the draft I realise that it does not really express
how we feel about the current publication procedure. That procedure
has been adopted by all HEP collaborations for certainly good
reasons which are deeper than those listed at the end of page 6.
If one shares this view then it is difficult to propose basic changes
to the publication procedure, beyond possibly the Belle practice
which a collaboration may decide to adopt or not.
I thus felt I should explain my worries and try to raise certain
points below. I regret that after Beijing we did not meet again
in order to have a discussion which in my view is close to
impractical on critical items over the phone. A meeting in
person is desirable from my point of view.
As much as I would like to thank Vera for the draft I also
feel sorry about so many dissident points I will be raising
subsequently. It may well be that mine is just a minority
opinion but perhaps it helps for discussions which will publicly
come no doubt when the report is out. I yet discussed the main
points with the deputy spokesman and the physics coordinators of H1
and the spokesman of ZEUS. They all share the principal view
I am expressing below while the wording is mine. Please excuse
me for misunderstandings or things overlooked.
best regards Max
--------------------------------------------------------------------
Comments to the C11 Interim Report (Draft of 3/22/2005)
In the following some general comments are made first and then
specific comments are made to the draft. As at the end this became
too long I added a summary.
Summary
-------
The criticism of the current procedure and the proposals/suggestions
(the 'consortia' way) made in the draft go much too far. The reasons
for why all collaborations publish the way they do are not outlined
properly but considered to be questionable. We thus risk to initiate a
large debate (or to be ignored) about publication praxis which is
not based on really ongoing discussions, at least in H1 or ZEUS, and
which also risks to devaluate our field in a time in which HEP undergoes
a particularly difficult period with cuts and changes to master. I
thus suggest to: i) explain the current procedure, ii) present the
procedures of others, iii) put forward ideas as to how to reach
better knowledge and responsivity of those undersigned (Belle).
In particular I think there should be no proposal for consortia
and I would avoid sending a questionnaire which is too formal
and a too large effort given the rank of the publication problem
among other more pressing problems the community faces.
General Comments
----------------
1. The way we collaborate with the aim to reach new results for
publication is true, large scale collaboration between experts,
who come and go (the H1 Collaboration changed to a large
extent over the years). Each publication reflects a common effort
including detector building, operation, upgrades, maintenance,
trigger hard- and software, DAQ, luminosity measurement, data logging,
simulation of physics and detector including GRID developments,
reconstruction, data analysis, calibration, alignment, software
developments (H1 moves from FTN to oo),theoretical model comparisons,
paper writing, etc. It is accompanied by strong efforts in guiding
students, internal refereeing analyses and papers, teaching,
communication about preliminary results with the external community,
management etc. Publications thus appear as the outcome of joint
efforts, not (only) of some first authors. This in my view has been
the reason for ALL HEP collaborations to follow the current
publication procedure, with modifications as by Belle, up
to ATLAS and CMS who have decided the same way already. It is
the style of our work rather than the simple reasons listed at
the end of page 6 of the draft which is at the origin of our
publication practice. This is my main point and the reason
to vote against any big changes.
2. Contrary to what the draft states this procedure does not hamper
scientific discourse: conferences, seminars, direct exchange of
information between the experts of certain analysis take place
for many of our results. In particular it is easy to reach the
most knowledgeable persons. Note that a collaboration has also
all interest to let the specialists discuss specific questions.
3. This practice, in my view, neither complicates awards or hinders
careers: applications are done with weighted/commented lists of
publications, those one considers to be own and those shared.
Also talks, conference contributions, technical papers, theoretical
ideas can be and are published with less coauthors. Recommendations
and job hiring require personal interactions, this cannot
be replaced by any publication rule.
4. 'Consortia' for analyses exist, in H1 in the form of working
groups around some physics processes (e.g. diffraction, inclusive
ep cross sections, heavy flavour, final state, searches).
The suggestion to let these groups publish on their own their
papers is not helpful for a number of reasons: i) it is not appropriate
because the amount of common knowledge and effort usually exceeds
by far the genuine analysis work; ii) it is pointing a way towards
competition, in trigger budgets, data taking, simulation efforts,
subdetector use for example, instead of collaboration, i.e. to
a style of work which in my view is opposite to what a collaboration
needs; iii) it ignores the fact that a publication is much more than
its contents; iv) it potentially confines you, having build a specific
detector, to a subject for long while you contributed to the
collaboration as a whole by building some detector expecting to
share the full physics programme (the point Lorenzo made).
The split of a collaboration into subcollaborations would
in my view endanger the whole project, reduce the efficiency and
I thus cannot support any proposal which goes into that direction,
see also below.
5. I personally see one main problem of our long authorlists, this
regards the degree of knowledge of the contents of a paper by the
undersigned. Here the Belle practice, which is not a principal change
of the HEP publication procedure but an interesting modification,
may help. In this regard one may think of more formal requests
such that for example when a paper is circulating for comments
one requests each participating institute to send in comments which
ensures that each collaborating group, not necessarily each individual,
has taken note and raised its questions. I doubt that all Belle authors
who agreed to sign a paper know its contents fully. Yet to increase
the knowledge and this way to maintain the individual responsibility
I see as a genuine goal of our discussion, to me the overriding one.
6. I find the questionnaire questionable. For example, in 1)Q2 the question
is asked whether current practice is the best way to give credit to
'everybody'. Of course not, but this is not correctly handled by
giving it a -5. The addition about the publisher problems suggests
an even more negative answer to be given. I doubt that the whole problem
can be formalised (a concern I should have raised but wasnt present
when the group put the questionnaire idea forward) because it
is too involved.
7. There is a point which I want to mention even if that may go beyond
the publication practice questions. This regards the future of HEP.
While we were/are discussing the authorship questions it has been
decided to close HERA, Superbabar and BTeV, i.e. DESY, SLAC and
FNAL (with the exception of the neutrino programme) will soon have no
accelerator based HEP experiments on site anymore: the number of
HEP experiments will shrink further with H1, ZEUS, HERMES, Babar,
D0, CDF disappearing in sth like 5, perhaps a few more years.
To exaggerate, we will have only ATLAS, CMS. These, however, have
decided. A Collaboration like H1 which publishes since 1992 will
not change its rules for the last years unless deeply convinced
of advantages.
If our group, i.e. if C11 sends a questionnaire to the whole
community the attention it gives to this problem in my view
is much too high as compared to what really bothers us: the
shrinking support of particle physics and the future of our field
(which is not identical to the future of the ILC). I add here that I
was told by the ZEUS spokesman, and this holds also for H1, that the
authorship question is not raised inside the collaboration and
of small interest as compared to many other problems the
collaboration faces. If that is true for others as well, we
should be handling our task such that the impression is avoided
that C11 focussed on one of the less pressing problems of HEP.
Specific Comments
-----------------
p1 It seems to me unlikely that we all agree on the proposals
made in the report. There then needs to be a way to make
that clear which avoids people taking off their name.
I 1-3 as explained above the 3 worries are not of principal
character for me because: 1. is an unavoidable consequence of a
long author list (> 10 , not only > 100 or > 1000); 2. is no
problem as we know how to solve that and 3. is overcome by
applying with your own and the full list or marking those
papers one has mostly worked on. What is missing as a
problem is 4. that too few authors are really knowledgeable
about the contents of their publication.
The introduction implicitely criticises all HEP collaborations
for following the current practice and ATLAS and CMS to
have adopted that. I tried to explain in 1) why this in
my view can be justified and is done like that. I think one
should not miss to explain this.
p3 $5/6 it is first mentioned that large telescopes can be
'booked' and then said that space based astronomy is similar
to HEP. It should be made absolutely clear that experiments
in HEP, at least H1, cannot be booked in any way. The experiment
operation, data taking and checks etc. need to be done by
those who publish in order to keep the understanding of the
experiment high and to run the whole thing. We always
rejected the idea of technicians operating the experiment
for the great physicists analysts. We are thus back to the
relation between the style/character of our work and the
way we release publications. One cannot focus only on the
second question but neglect the former. I am afraid this is
what we have done to some extent.
p4 II.3 Regarding the 2nd Belle statement there is a danger that
it may be too easy to say 'I disagree': In H1 we usually dont
publish until consensus is reached which often changes the
analysis or interpretation. Would be nice to know how at Belle that
is handled in practice (of course an author may take his name off,
this happened in H1 without causing major debates).
p5 $1 The celebrated Belle price award: isnt it correct to say
that prices are awarded also to members outside Belle with the
traditional publication practice? The way that is presented here
sounds like the golden way for awards.
p6 $2 The sentence regarding the recognition of outstanding
young scentists suggests that there are no outstanding
young scientists in technical areas. One may change
'in general' to 'although' and remove the 7 lines from 'but there'
till 'others.'
$3 I have not had such experience.
$4 I dont think that the scientific discourse is hampered by first
finding the most knowledgeable people, as stated above already.
In the experiments I was since 1973 there were never
graduate students the first authors. That is sth one would still
consider. Yet as soon as one thinks about this there are doubts:
very rarely someone is really the most outstanding author, if
an ingenuitive analysis is done it usually requires an outstanding
detector, an innovative trigger, professional reconstruction or/and
alike. Behind those things there are often young or less young
people with energy and intelligence and career hopes. Thus even
that idea is nothing one would 'just' implement.
$5 see my criticism above 1). a-d are formulated such that one
concludes that those defending the current system are hesitant of
changes or dont think enough. Yet, it is the style and character
of our way of running the experiments and producing results,
much more than any of the points which are listed here which
is at the origin of the current practice.
p7 $1 I see no need to list these statements, and would not do so.
$2 The first statement is too hypothetical. I think it is upon us,
the authors and collaborations to define who is author. Technical
limitations need to be overcome with technical means and the
publisher's concerns may not affect us really. Note for H1 the author
list is rarely identical, people come and go. One is author only
as a member of the collaboration (not because one has built a
detector once) and can stay at most 1/2 year after leaving H1,
similarly one has to have contributed for 1/2 year before one signs
a paper.
IV The Belle model is interesting but since I am on a 'dissident
way', let me try to challenge the 5 points raised:
- the reduction of authors: this demonstrates a decreasing interest
of the Belle physicists in their results and it does not really
solve the problem of a long authorlist.
- this point refers to one example, as far as I know, of an award,
there have been many awards before.
- it would be good if that was true, is it really? do the 150-200
undersigned have really read the paper they sign?
- nothing one would have expected unless a complete institute
stopped signing papers.
- which readers and what happens for the alphabetically ordered
Belle papers, are they indeed discussed less?
p8 The 'consortium idea' is in contrast to how at least H1 operates and
obtains results as I tried to explain above. Moreover, it does not
solve the problem in principal. If I would divide H1 into 5 subgroups
then there are 60 people signing a paper and that is the level
of fixed target experiments to which the criticism about anonymous
authorship etc would apply similarly. ATLAS and CMS can in no way
be split now since they all work for discovering Higgs and SUSY, i.e.
if one was working in a QCD group and was told one can only sign
papers on pt spectra, for example, who would continue to work and
agree to not sign the discovery paper, and who is important, the guy
who saw the mass peak first or the one who has aligned a muon detector?
There is the danger to be less efficient: being a member of a specific
consortium I may overlook some work of others which may be of
use for my own analysis. A collaboration has certain standards which
are maintained by the current procedure. Being in group A and not
agreeing with a result of group B leaves someone, in the proposed scheme,
with no influence to oppose an analysis because one is not supposed to
sign the paper of B anyhow. In other words I see no real advantage but
too many counterarguments.
----------------------------------------------------------------
Dr.sc. Max Klein
H1 Spokesperson tel(fax) +49 (0)40 8998 3135 (3093)
secr Mrs. Alla Grabovsky -3144
email klein@ifh.de
DESY H1 Notkestrasse 85 22603 Hamburg GERMANY
----------------------------------------------------------------
|
Re: Draft of Write-up on WG on Authorship 
Search
|
Feedback
]
icon
and there is no indentation. 
