Tuesday, August 02, 2005

Two New Blogs

Chris Leonard who edits the Elsevier journals in theoretical computer science has started a weblog Computing Chris. He plans to address some of the concerns of the community to commercial publishers and Elsevier in particular. Feel free to suggest topics to Chris.

Jeff and Suresh point to David Eppstein's new weblog 0xDE. Eppstein always had a number of fun and useful stuff on his website including a catalog of the complexity various games and puzzles.


  1. I wonder if this guy's blog is just going to be a constant source of infuriating propaganda.

    "So as things stand, there is no good reason to stop printing journals."

    NO good reason? Couldn't you at least humor us and say that there's at least one?

    "It is interesting to speculate what kind of business model this may usher-in too. Since there is no printed journal, annual subscription prices to a journal are not as applicable as they once were and maybe we will see new kind of licensing models?"

    Maybe we will see your company crumble into the ruins of no-longer-serves-a-purposeness?
    I'm just speculating.

  2. Elsevier has been a grossly exploitive steward for our community. They combine the worst of opaque and monopolistic pricing strategies with aggressive copyright policy. STOC finally abandoned JCSS (an Elsevier publication) for precisely these reasons.

    Of course, they are not to blame---we are! They have absolutely no incentive to change their practices so long as our libraries continue to pay their high prices and we continue to send them our best work. When you next consider where to publish the hard-won flowers of your research, select a society journal (SIAM, ACM, IEEE) or Theory of Computing, which actually has no cash flow and leaves the copyright with the author.

  3. While the blame does fall largely on our shoulders, we should still hold Elsevier accountable for their treatment of the community.

    It seems that a good first step would be to start a public web page on which we (the members of the CS/TCS community) could pledge to no longer submit our papers to Elsevier and to no longer referee for their journals. Hopefully this would create a sense of solidarity that encourages others to follow. What do people think about this (given, for instance, that many of our friends are editors for these journals)?

  4. Blaming Elsevier for problems in scholarly communication is a bit like blaming Detroit for American gasoline consumption. Yes, they share some culpability. But (a) Elsevier is entitled to maximize profit. (b) The real problem is structural; it is not as simple as evil companies vs the public good. And (c) there are practical solutions that have made an enormous difference in certain areas.

    I am one of many supporters of the arXiv. In most of physics and in much of mathematics, it is the foundation of the new structure of serious scholarly communication. It is not by any means a perfect or complete solution, but it is pretty good. It works well enough that many people in its most invested areas, for example in quantum computation, are too bored with Elsevier to criticize it.

    The real question in these areas is how to build the next level on top of the arXiv. I discuss this in an arXiv article, math.HO/0210144.

    Computer science, or at the very least CS theory, could have been where physics and math are today. There are couple of legacy reasons that explain why it isn't. One of them is ECCC, which is an honorable but Lilliputian alternative to the arXiv. It would be a big step forward if ECCC and the arXiv shared papers. This may seem like a heavy-handed proposal when there are networked emulations of paper sharing like CiteSeer and Google Scholar. However, the arXiv is as much a work of social engineering as a technical success (as is ECCC, on a smaller scale). CiteSeer is a much weaker solution in this regard.

  5. First of all, people should--without question--submit to _both_ ECCC and the arXiv. (In fact, I would prefer that ECCC act as a "complexity theory" front end.)

    I can't really speak to physics, but I don't think you will find any serious mathematician submitting to the arXiv and not to a refereed journal as well, thus this doesn't really solve the problem with respect to Elsevier (and, to a lesser extent, Springer). For instance, Elsevier still hosts the special issues at other theory conferences (e.g. ICALP/TCS).

    The argument that "Elsevier is entitled to maximize profit," when left unqualified, doesn't make too much sense. At what cost, and to what ends? This kind of justification makes it easier to not see them as essentially a parasite on the public good, which makes it easier to justify cooperation with them in the future. The reasons that we have not seen mass-exodus from these publications escapes me. Is it only because we are so entrenched?

    I leave you with this quote from the "Computing Chris" (Elsevier editor) blog:

    "It saves some money, but not a lot. The majority of the costs associated with producing STM articles comes from supporting the editorial process, typesetting (in XML in our case) and maintaining servers where the information resides. Printing and mailing are minimal extra costs compared to this."

  6. I recently had a paper typeset by
    Elsevier journal Discrete Applied
    Math. They did a terrible job despite having my tex file: bad
    typesetting, introduction of errors
    and leaving out common sensical things. The typesetting was done in India which is fine but they should maintain standards. I complained and got a rather callous response which did not acknowledge the bad quality of the process. After finishing the
    process they asked me to look at
    the electronically uploaded file
    by logging into Science Direct,
    their paid service. I wrote back
    nicely saying that I don't have
    access and they essentially said
    "tough luck". I responsed with a
    strong email threatening to not
    publish anymore with Elsevier and
    then they sent me the pdf file.

    Clearly they have the pattern of
    maximizing profit with very little
    consideration for the authors.
    The balance they strike is up to
    them but we can influence this by
    going elsewhere with our papers.

  7. To respond to the second to last anonymous message (I'm not sure why this discussion has to be so anonymized):

    A priori I see Elsevier neutrally as an actor in the public sphere. I am an unromantic capitalist. The first obligation of a publicly traded company is to maximize profit. Whether the company's relationship to the public is symbiotic or parasitic is a matter of judgment, market rules, and alternatives. The arXiv does make Elsevier's existing business model look parasitic. But I don't mean that as a moral condemnation, I mean it as a weighing of alternatives.

    I agree that the commercial publishers are entrenched, and you could argue that that is what mainly sustains them. But then, every important player in scholarly communication is entrenched. The arXiv is entrenched, Google is entrenched, the ACM is entrenched.

    It is true that most serious math papers still go to journals. But in areas where the arXiv is completely entrenched, that is done only for peer review. So you have publishers that print and sell not because authors need it, but because they can. Then the libraries buy the journals out of inertia, not because readers need it. Here Andrew Odlyzko has said that libraries spend more money babysitting journals than they do buying them. It is easy to heap odium on commercial publisher, but what about the expensive librarians?

    Lastly, and most importantly at the practical level, I agree that if there are no structural changes, people should submit to both ECCC and the arXiv. But no one enjoys double submission and it is the main impediment to the social engineering effect of both the arXiv and ECCC in complexity theory. So yes, it would be great to see automatic paper sharing. It is better to leave the details of this to direct negotiations, but suffice it to say that it would be good to negotiate it.

  8. "It saves some money, but not a lot. The majority of the costs associated with producing STM articles comes from supporting the editorial process, typesetting (in XML in our case) and maintaining servers where the information resides. Printing and mailing are minimal extra costs compared to this."

    This isn't really worth a reply but if it is true then I'm happy cars don't cost 1000000$ because the local vendor chooses to take them apart and rebuild them buy a bunch of preschool kids.

  9. Two years ago I decided no longer to referee or submit articles to Elsevier and Kluwer and I'm curious to see (on a web site, or otherwise) who has made a similar decision.

    I agree with Greg that there's nothing intrinsically wrong with commercially publishing academic material, as long as all parties involved get a good deal (see for example Springer Verlag's GTMs). But by the same token -at the moment- it just doesn't make sense for me to spend my time refereeing (non-stellar) articles for journals that clearly make money out of my work. That half-line on my resume "refereed for XYZ" just is not enough; if Elsevier/Kluwer would be willing to reward me in some way for my work, I might reconsider my decision. Of course, I can't submit articles to journals for which I don't referee, but there are enough non Elsevier/Kluwer journals for that not to be a problem.

    Recently, the American Physical Society has discussed some suggestions how to make refereeing more rewarding.

    And, yes, ECCC should give up its resistance and join the arXiv.

    - Wim van Dam

  10. Hi all - and thanks for the welcome to the blog world! Sorry it's taken so long to reply but I haven't been able to post comments on here recently. I did post a brief reply to some issues on here on my own site.

    I'm aware that I'm putting myself up to be shot down here, but I think Elsevier has been absent from these types of conversations in the past, and has suffered for it.

    It will probably take a while to find my voice on the blog (and not antagonize everyone), so please have a little patience with me.

    Also, some topics will need some research on my behalf, so I may not be posting on them immediately.

    But I do want to address these points - and others too. Despite the occasional roasting I may receive.