Re: Is science on sale this week?

Dear all,

obviously, this is a heated discussion, and I usually wouldn't try to
engage here because it's full of all kinds of pitfalls, but there certainly
are some aspects that need clarification and maybe, I can contribute a
little.
As for myself, I am a scientist and a founding member of the Open
Linguistics Working Group of the OKFN, so I am in general very sympathetic
to open publication models. Also, I've been editor for half a dozen
proceeding publications in the last few years. At the same time, I happen
to have some insight in the work of publishers because I have personal ties
to people working at a publishing house.

(0) Selected personal experiences: Last year, we conducted a workshop on
Linked Data in Linguistics. We tried to find a suitable channel for an open
publication, but back then, technical issues with Ceur (that seem to be
solved by now) prevented us from doing so. In the end, we went for a hybrid
model: Publishing draft versions of the papers online (we made sure that we
were allowed to do so), but having a commercial print publication of the
final book. In this way, the core information is accessable to everyone,
but all references, and all quotations should be made to the printed
edition.

(1) I have no specific knowledge of contract details for ESWC and ISWC, but
usually, little to no money flows between conference organizers and
publishers. Some publishers will "pay" you with free samples, but some
publishers will actually charge you for publishing your proceedings to
cover their expenses (and this does not necessarily mean that you keep the
copyright!). In my experience, Springer will not. But they can afford this
only because they are selective. And this selection process is one aspect
why commercial publications, in particular in established journals, or book
series will be regarded as an indicator of quality for the foreseeable
future.

(2) Given the way academia works nowadays, scientists must have an interest
in being able to convince people from other disciplines from the quality of
their work. These people will dominate commissions, and *they* will be
reviewing your grant proposals, not your peers. (This is true at least for
smaller fields like the SW world or Natural Language Processing.) And there
are no objective means to convince them of your quality as a researcher and
on the basis of what you actually have done and written, other indicators
such as impact, dissemination, presentation of your publications are
getting increasingly important. Having a printed book to present to a
commission actually helps, in particular if it is from the "right"
publisher (say, John Benjamins for discourse studies, de Gruyter for
general linguistics, or Springer for mathematics), in the "right" series or
in the "right" journal. This system is broke in many ways, true, but if you
want to survive, that's what you have to do.

(3) Publication contracts clarify the rights of the publisher and the
author/editor, and these are negotiable. As I remember from our Linked Data
in Linguistics book, Springer contracts *explicitly* allow authors/editors
to provide draft versions over the internet, and as this is fixed in the
contract, they cannot withraw these rights without contract renegotiation.
(I mentioned Springer, becaue other publishers I had business with before
were less flexible in this regard.) If your publisher won't go with that,
choose another one.

(4) One should probably ask someone from publication business for
confirmation, but in my understanding, arxive.org serves as a *pre*print
server, and if your contract gives you (or your contributors) the right to
make private copies available online, there is no legal way from preventing
you from publishing your draft papers there.

(5) An even better way than publishing drafts online would be to develop
contracts explicitly stating that and how the copyright can be regained by
the original author after a few years or for a certain amount of money.
(Buying it back is always possible, but certain publishers will charge you
several thousand bucks -- depending on the publisher and the type of
publication, of course.) Many publishers already have open access models,
so you can buy the right to let them distribute your work under an open
license, basically you pay for their infrastructure.

(6) Commercial publishers provide some infrastructure. This includes
long-term hosting and lecturing (although it is certainly true that
lecturing quality has decreased in the last two decades, if a publisher
offers this at all). So, publishers still provide an infrastructure from
which you can benefit. On the one hand, they are still holding the
"established" brands for publication series. But also, as large publishers
do not depend on selling individual books, but rather packages to libraries
(both electronic and printed), having your book in one of these packages
means that they are automatically disseminated among participating
institutions.

(7) Commercial publishers are struggling to survive given the
marginalization of their traditional area of expertise, printed books.
Whether this is something to be concerned or happy about is debatable, but
it forces them to develop new business models, so open access models are
being developed on their side, too. On the one hand this offers the
advantage of using their infrastructure with the benefits mentioned above,
on the other hand, this may be costly enterprise for the author. (Some
costs would also arise if you maintain a publication server yourself, and
one need to look in detail whether costs and benefits outweigh each other.)
Although I would in principle support such solutions, at least if the
results are eventually available under a truly open license like CC-BY, the
trouble is that this shifts the financial burden of publishing and
acquiring publications from libraries to the researcher. I don't know about
other institutions, but in my university, I don't see any additional funds
for creating open access publications.

(8) A better solution would be a free, community-maintained portal where
researchers are allowed to publish for a minimal fee (or no fee at all).
But there is no such thing as a free lunch, and long-term sustainability of
this platform for the next, say, 100 years, needs to be secured *somehow*.
So, it represents a considerable financial load. If we can come up with
some funding models that shift the financial burden from the author to,
say, sponsoring institutions or libraries, this would be a perfect world.
Obviously, people are actively working towards this, and I would actively
support that whereever I can, but at the moment, this is little more than a
vision on the horizon. (But let me know if I'm wrong. I don't know too much
about the financial future of existing open access platforms.)

Just my two (well, eight) cents ;) To sum it up: At the moment, the
double-publication strategy of free drafts online plus commercial final
publication (resp., open-access proceedings and commercial postproceedings)
seems to offer the best of both worlds, and depending on your publisher and
your contract, it should be possible to do so in a legally proper way
already at the moment.

Best,
Christian
-- 
Christian Chiarcos
Applied Computational Linguistics
Johann Wolfgang Goethe Universität Frankfurt a. M.
60054 Frankfurt am Main, Germany

office: Robert-Mayer-Str. 10, #401b
mail: chiarcos@informatik.uni-frankfurt.de
web: http://acoli.cs.uni-frankfurt.de
tel: +49-(0)69-798-22463
fax: +49-(0)69-798-28931