Re: What is needed for an online journal

Re: What is needed for an online journal

[Marino Miculan]

On 31/dic/06, at 21:30, John Baez wrote:

>> What about having an editorial board, which would look at papers  
>> on the
>> arxiv, say, have them reviewed and revised, and then put them back  
>> on the
>> arxiv in final form, and listed elsewhere as having been through that
>> process and "blessed" so to speak by the editorial board?
>
> This is what many journals do, after someone submits the paper.
>
> For example, with Advances in Theoretical and Mathematical Physics,
> you submit a paper merely by sending them its arXiv number; when
> it's accepted you prepare a version in their preferred format and this
> gets put on the arXiv.

Another, and quite successful, example is the Journal of High Energy  
Physics (http://jhep.sissa.it), an on-line journal which I have  
worked for at its beginnings, many years ago (back in 1997).
Papers can be submitted by indicating an arXiv number, or by  
uploading a (PDF, LaTeX...) file.
The editorial procedure is fully automatized, in the sense that it is  
fully operated on the web site, with minimal human intervention. This  
allows to reduce maintenance costs, and to speed up the publishing  
process of one magnitude (the average time from submission to  
publishing is something less than 2 months, which is mostly due to  
the referees). Once, accepted papers were freely available online  
from JHEP site; nowadays, these are available online through IOP's  
Electronic Journals service; but I guess that papers are still  
available for free (at least for some time), or at a reasonable price.

Ten years ago, JHEP was started by an academic consortium as a  
spontaneous answer to the (already!) outrageously increasing prices  
of the main journals in the HEP field, especially Nuclear Physics B  
(which is run by Elsevier, and costs more than 15200 euros/year... so  
Vico was right, after all.) Nowadays, JHEP has become one of the  
major journals in the field: for instance, Ed Witten, Ashoke Sen and  
Cumrun Vafa regularly publish papers on JHEP.
For what it is worth, in 2005 the JHEP Impact Factor was 5.944, that  
of Nuclear Physics B was 5.522.  So on-line journals can compete with  
"standard" journals also on this point.

As far as the "real existence" of online papers, especially for grant  
applications, career advancements, etc: if I remember correctly, from  
a formal point of view the only important thing is that to have an  
ISSN number - which means that the journal is officially recognized  
as a periodical publication. The existence of a "printed version" to  
store and forget in dusty (and increasingly deserted) libraries is  
not needed.


-m

--
Marino Miculan - http://www.dimi.uniud.it/miculan/
Department of Mathematics and Computer Science, University of Udine
via delle Scienze 206, 33100 Udine - Italy  --  skype: marinomiculan
vox: +39-043255-8486 - fax: +39-043255-8499 - mob: +39-3292606452

Re: What is needed for an online journal

[Ronnie Brown]

Mike Barr has put his finger on the key points: papers are needed not just as a contribution to the advancement of knowledge but also for career prospects and the awarding of research grants, which are also related to career prospects. Further there are the notions of `impact factor' of various journals, and that of citation indices. 

On the last,  some Governments  take the line that a journal to be rated for research promotion has to be on the list of the Institute for Scientific Information (ISI: http://scientific.thomson.com/). It is possible that the UK Government, in moving to using `metrics' for research impact, will take this line.  A search of the ISI list shows that it has of course journals of major publishers and of national academic institutions, and that it claims to  have a procedure for adding journals to their list. But the operation of the listing is based on the idea that not all journals need to be listed, in order to assess significance. TAC is not on the list, and I think neither is NYJM, Cahier. It is also difficult to get on the list, as evidence from editors shows, and there is no evidence of an academic input to the ISI  procedures. The organisation is a commercial organisation, which has some control of scientific information. 

Eugene Garfield writes in 2004: 
Garfield, Eugene] ARE YOU SUGGESTING ISI COVER THE LOWEST IMPACT JOURNALS AND PASS LESS ATTENTION TO THE HIGHEST? 
[Garfield, Eugene]  WE CANNOT CONTROL HOW THE DATA IS USED. I HAVE DONE MY BEST TO PREVENT ITS ABUSE BUT I HAVE NO POWER TO CONTROL IT. 

Perhaps also mathematicians in research assessment forget the long time scale of the impact of new ideas. It is very easy to rate work which is related to famous problems. It is less easy to rate work which opens a new range of ideas, as has category theory, for example. I am grateful to David Corfield for pointing out a quotation from Rota's `Indiscrete thoughts' p.48: 

 ``What can you prove with exterior algebra that you cannot prove
without  it?" Whenever you hear this question raised about some new
piece of  mathematics, be assured that you are likely to be in the
presence of something important. In my time, I have heard it
repeated for random  variables, Laurent Schwartz' theory of
distributions, ideles and Grothendieck's schemes, to mention only a
few. A proper retort might be: ``You are right. There is nothing in
yesterday's mathematics that could not also be proved without it.
Exterior algebra is not meant to prove old facts, it is meant to
disclose a new world. Disclosing new worlds is as worthwhile a
mathematical enterprise as proving old conjectures.

There is a discussion of ISI and related issues of impact factors etc in 
http://en.wikipedia.org/wiki/Institute_for_Scientific_Information



Ronnie Brown
www.bangor.ac.uk/r.brown

Re: What is needed for an online journal

[Michael Barr]

I guess reality will have to intervene.  Young people need refereed
journal publications if they hope to get tenure and promotion.  Canadians,
and doubtless others, require them in order to get research grants.  If
you are going through the process of refereeing them for ArXiv, you may as
well start a journal; it costs no more.  TAC is a good model and I am
about to find out about the NY J. Math. since I am a coauthor on a paper
that has no category theory in it.

I have probably published as many papers in TAC as anyone else and have so
far managed to convince NSERC that these are genuinely refereed papers.
I recently had a paper turned down for TAC, incidentally, so the
refereeing is real.  I had my doubts whether the paper was publishable and
now I am sure.  On ArXiv, there would have been no judgment.

However, I do not believe that this model of totally cost-free publication
is viable in the long run.  Since the universities will be the main
beneficiaries of the eventual demise (I hope) of the commercial journals,
they ought to be giving us the modest help we need.

Michael

On Sun, 31 Dec 2006, Bill Rowan wrote:

> Hello,
>
> We have the arxiv preprint server, and other preprint servers.  Why is
> this not sufficient?  Because the papers are not refereed and subject to
> being made better through the work of editors (not that they always
> improve things).
>
> What about having an editorial board, which would look at papers on the
> arxiv, say, have them reviewed and revised, and then put them back on the
> arxiv in final form, and listed elsewhere as having been through that
> process and "blessed" so to speak by the editorial board?
>
> What we would all like to have is more prestige for having our papers
> blessed, and hopefully, read by somebody.  In this proposal, the quality
> of the editorial board's work and the usability of its published listings
> would be the most important thing.  They could even give a negative
> recommendation for someone's paper, although authors might hope for the
> opportunity to withdraw their paper if they didn't appreciate the final
> recommendation.
>
> The pulbished listings could include more information than just an up- or
> down- recommendation to read the paper, it could include comments by other
> mathematicians.
>
> Bill Rowan
>
>
>

Re: What is needed for an online journal

[John Baez]

On Sun, Dec 31, 2006 at 09:16:57AM -0800, Bill Rowan wrote:

> What about having an editorial board, which would look at papers on the
> arxiv, say, have them reviewed and revised, and then put them back on the
> arxiv in final form, and listed elsewhere as having been through that
> process and "blessed" so to speak by the editorial board?

This is what many journals do, after someone submits the paper.

For example, with Advances in Theoretical and Mathematical Physics,
you submit a paper merely by sending them its arXiv number; when
it's accepted you prepare a version in their preferred format and this
gets put on the arXiv.

Other electronic journals, like Geometry and Topology or Algebraic and
Geometric Topology, require that you send them LaTeX in their preferred
format when you submit a paper.  If and when it's accepted, the final
version gets put on the arXiv.

It's a small extra expense for a journal to keep its papers on its
own server as well as the arXiv, so that practice will probably continue.

It would be a huge amount of work for someone to edit ALL papers
on the journal, so that won't happen - unless some billionaire decides
it's worth setting up a foundation to do this.

Best,
jb

What is needed for an online journal

[Bill Rowan]

Hello,

We have the arxiv preprint server, and other preprint servers.  Why is
this not sufficient?  Because the papers are not refereed and subject to
being made better through the work of editors (not that they always
improve things).

What about having an editorial board, which would look at papers on the
arxiv, say, have them reviewed and revised, and then put them back on the
arxiv in final form, and listed elsewhere as having been through that
process and "blessed" so to speak by the editorial board?

What we would all like to have is more prestige for having our papers
blessed, and hopefully, read by somebody.  In this proposal, the quality
of the editorial board's work and the usability of its published listings
would be the most important thing.  They could even give a negative
recommendation for someone's paper, although authors might hope for the
opportunity to withdraw their paper if they didn't appreciate the final
recommendation.

The pulbished listings could include more information than just an up- or
down- recommendation to read the paper, it could include comments by other
mathematicians.

Bill Rowan