Friday, March 30, 2012

The Value of an Academic Publication

Russell O'Connor's paper was accepted into last years ACM SIGPLAN Workshop on Generic Programming. Russell put the final version of his paper on the ArXiv under a public domain dedication. Russell couldn't transfer author's rights to the ACM since he gave them up. After much discussion the ACM decided not to publish the paper in the proceedings. The abstract in the proceedings states
We note that one of the papers presented in the workshop is not included in the proceedings. This paper, "Functor is to Lens as Applicative is to Biplate: Introducing Multiplate" by Russell O'Connor, is accessible as arXiv:1103.2841v2 [cs.PL].
Russell gives his account but he focuses on ArXiv instead of the public domain aspect. In the STOC CFP we encourage putting your submissions on ArXiv and similar sites. The issue that worried ACM was the loss of rights. ACM could have published the paper, it was in the public domain, but it wouldn't have control of that publication and didn't want to set precedent. Scott Delman of the ACM responded here.

An academic paper you write has little direct value to you (the paper itself not the Intellectual Property within). No one is likely to give you any money for that paper. But your paper does have financial value as part of a collection in a journal or conference proceedings. Commercial and non-profit publishers know how to collect on this value. You might complain that this puts your paper behind a firewall but all the major publishers in CS allow you to post earlier drafts on your homepage and archive sites. As long as you make that effort, people will have access to your papers.

It does take some money (or considerable person hours) to maintain even an electronic journal or conference proceedings. But publishers get more value than that. For the ACM, the DL revenue is a major source of funding for ACM activities and those of the SIGs.

Of course you should never trust the opinion of someone who has a financial interest in a position and as SIGACT chair, we certainly make use of the DL revenue. We are hoarding some in case the DL revenue shrinks in the future.

It seems a shame to leave the monetary value of our papers on the table but also that value shouldn't be exploited. Big discussions will continue on the publications issue, at ACM and other publishers and at all levels of the CS community. There will be a Dagstuhl workshop focused on this topic in the fall. Figuring out the right model for publications will not be an easy one.

My biggest fear is that lack of a plan will lead to a degradation in the quality of our publications and then everyone loses.


  1. I see no reason to worry about a degradation in the quality of publications. If STOC were to disassociate tomorrow from ACM, registration fees might rise, but publication quality would not be reduced one iota. In fact, it will probably increase, since we won't have to stick to that ugly two column format.

    What I would worry about is that ACM, by continually pursuing anti-open-access policies, will brand itself as a "mini Elsevier", and cause scientists to distance themselves from it. The ACM should realize that the business model of charging access fees for publications produced by volunteer scientists paid by government or other funding is doomed. Clinging to this model at all costs will only harm the other good activities that the ACM does. That is something that I worry about.

    Boaz Barak

    p.s. I find Scott Delman's response highly disingenuous. Lance was at least honest enough to admit that the issue is not recuperating the negligible costs to actually publish the papers, but to finance other ACM activites.

    By now it's clear that the costs to actually publish electronic papers are negligible and should not be part of the argument. The annual arxiv budget (handling 30M downloads per year), is $400K. To put it in perspective, this is .00005 fraction of the NSF budget, roughly equivalent to three typical individual NSF grants. Sure, the arxiv and other open access entities rely also on unpaid volunteer labor, but this holds also for the ACM.

  2. I think we should look at alternative models for funding ACM.

    For example, how much revenue would you get if you put some tasteful Google Ads on the ACM web site? I am sure the Digital Library collectively gets a lot of traffic.

    This is just one idea (I know some people vehemently object to advertisement on principle).

    In general I think the value to the scientific community of free dissemination of information is far greater than protecting one particular revenue stream.

  3. My biggest fear is that lack of a plan will lead to a degradation in the quality of our publications and then everyone loses.

    Can you elaborate on how this could happen?

  4. I don't see how putting earlier drafts of the paper on one's homepage (which as you said, is allowed by ACM) does not hurt the revenue of ACM DL, but putting it on arXive under a CC license does. So, it seems to me that ACM's position is based on a principled opposition to open access, rather than a loss in revenue.

  5. Posting your papers on your website only works if a) you actually do it and maintain the links appropriately, which not everyone does (unlike you, Lance!), and b) you are still alive. Math papers from 100 years ago get referenced on a not-infrequent basis. There are already some CS papers from 40 years ago that are similarly referenced. Having authors post on their website is a good service, but is not a good solution for free access in the long run.

  6. There is no reason that individual authors couldn't give organizations like ACM "license to publish and republish" their works rather than transfer their copyright. The combination as a journal or proceedings could still be behind a paywall and fund these organizations but such an item could only be reproduced if all the authors also gave such a license to some other organization.

    Alternatively, all publications could be published under delayed open access. (I suspect that this would really cut SIGACT revenue from the DL since we get a lot of money from views of classic papers.) "Giving up" the value is also a matter of redistributing it to the research community.

    There is value in the aggregation and indexing that the ACM DL provides. I don't see a problem with them charging for that to defray costs.

    If organizations like ACM don't choose something like one of these two models, I think they will be in trouble.

  7. If ACM cannot support the other activities using other sources of revenue (membership fees, government grants, charitable contributions, advertisements, etc) then, perhaps, they do not add sufficient value to the community (SIGACT news?).

    And just to add: STACS proceedings are open access (

  8. Agree with Mohammad.

    If the paper has already been posted to arXiv, the author no longer has the ability to transfer copyright. Publishers, including ACM, need to accept this simple fact.

  9. I think we should kill the ACM. This organization is no better than Elsevier. It is trying to lock up our research, hiding it from other researchers and the public. The Elsevier boycott is because they are actively fighting against open access. Now so is the ACM! They are refusing to publish papers that are on the arXiv, because it might set a precedent? Evil.

    "The DL revenue is a major source of funding for ACM activities."

    These activities do not offset the downside of hiding our research away. Allowing authors to post papers on their homepages is not a solution, because it is not archival. It is nearly impossible to find online many old and important theory papers, because they are under lock and key, and will be forever.

  10. This act by ACM is inexcusable. The paper is public domain, and therefore there is no danger in publishing it in the proceedings. What exactly is ACM afraid of? That the government will suddenly retroactively make it illegal for journals to contain public domain papers? Even in such an extremely unlikely scenario, ACM could simply remove the paper when that happens. I've lost a lot of respect for ACM because of its position on open access. ACM is worried about setting a "precedent", but it IS setting a precedent: "Put your paper in the public domain and ACM won't publish it!" What kind of precedent is that? The ACM needs to join the 21st century or else be renamed to A20CCM (Association for 20th Century Computing Machinery).

  11. "It seems a shame to leave the monetary value of our papers on the table."

    Not to me.

  12. Conference proceedings limit access in other ways too.

    In order to publish in conferences you must be able to attend the conference.

    This results in the rich getting richer effect: you can only publish in conferences if you live in a country where there is sufficient grant funding. In the US you must be well funded.

    Moreover conference publications are important in evaluation of computer scientists - so even in the US it is the rich getting richer effect (and rich here means in money - not in intellect).

    Similarly attending conferences is easier for healthier people with good family support.

    Why should the main venues for publishing results in computer science should be so biased in the 21th century?

    Elchanan Mossel

  13. Lance,

    I find the argument "It seems a shame to leave the monetary value of our papers on the table" unconvincing for the following reason.

    ACM's mission is to "advance the field of computing". Therefore if we talk about value, we need to talk about value to the whole computing community, not just ACM's bank account. By charging for access to papers, ACM obtains funding but that funding ultimately comes mostly from the computing community itself (indirectly, via overhead on grants paying subscriptions). Even at this stage, it's not clear there is a net positive value for the community. If we then factor in loss of access to research, it seems likely closed-access becomes significantly negative for the community.

    At least, I have not seen any argument to rebut that. Scott Delman wrote, "it has been overwhelmingly clear to these members of the community [the ACM Publications Board] that ACM's approach is in the best longterm interests of the scholarly community." Perhaps we need to close the gap here because it is not overwhelmingly clear to a lot of other people. Why has USENIX, which has a much more open policy, not "shuttered its doors" as Scott fears would happen to ACM?

  14. DL revenue has long been seen as harmless or beneficial, but I think it's becoming clear that it is a corrupting influence on the ACM, causing it to act against the progress of science and against the interests of its own members. In my view, the only solution is to take the money out of scientific publishing, e.g. by using the arxiv for hosting papers, and overlay systems to handle refereeing and related activities.

  15. The ACM's copyright policies are offensive, actually worse than Elsevier's before the boycott. Currently, just about every reputable math journal allows authors to post manuscripts of accepted papers on the arXiv, but the ACM's policies do not. The American Mathematical Society makes a substantial profit off of journal publishing, which it uses to subsidize other activities, but it has very liberal copyright policies (including allowing authors to retain copyright) and all papers are available for free after five years. Even Elsevier has decided to make papers in 14 of their math journals free after four years. Most publishers do not yet do this, but I believe the idea of time-delayed open access is going to take off in mathematics, and before too long this will be part of the minimal expectations for a responsible academic publisher. The ACM should be ahead of this trend, not behind Elsevier.

    The ACM seems determined to exploit all the content in the Digital Library for as long as copyright law will allow, which is a very long time. When anyone suggests they should do otherwise, they react as if this were an unheard of and crazy possibility, while completely ignoring the experiences of organizations like the AMS.

    Any "professional society" worthy of the name should have policies that are as open as reasonably possible. This doesn't preclude using the subscription model, or making a fair profit in exchange for providing a valuable service. However, the fundamental question should be "how soon can we afford to make this paper open access," not "how can we collect as large a fraction as possible of the money authors are leaving on the table."

  16. "Leaving money on the table" reminds me of what a systems prof recently told me, which is that when he calculated the DL revenue for SIGOPS (the SIG for operating systems), it was about 1% of the research funding that went into the papers published in that year's systems conference.

    One way to view this is that if the paywalls reduce the scientific impact of the papers by 1%, then putting papers behind them means "leaving money on the table."

  17. Ha ha ha, this is a good April Fools joke, if a bit early. "Leaving money on the table" = increasing access to research as a bad thing? You're killing me, Lance!

  18. If you are claiming that ACM has a net positive effect on CS then you should write a post about it. From where I see things, ACM has become a big bureaucratic parasitic entity.

    When did ACM have a poll of its members about the issues like this? The decisions are being made in a completely undemocratic way right now. Please publish ACM's budget, particularly how much of it goes to its employees and things will become much more clear.

    If majority of members believe that ACM should turn into open-access then it must do it, otherwise it doesn't deserve to survive anymore.

  19. I am going to stop paying my membership dues if ACM continues to act the way it does, and I will encourage others to do the same, maybe a boycott will cause people running ACM to change their mind.