I do want to address one point, Stevan, as it’s been bothering me a bit. You write on BoingBoing:

– NH: “Why do they harass people who create link indexes to the author-hosted papers (the Ke-Sen Huang incident)”
– SH: ACM rightly withdrew its take-down request. So what’s the point here?

The point is that some of the ACM people involved in publications are, and continue to be, on the side of protecting publishers’ rights, not members’ rights. The ACM did finally come down (mostly) on the right side of this issue, in large part because Naty and others quickly got the word out and ACM members wrote Bernie and Holly. As it is, the compromise solution is that Ke-Sen should not directly link to the paper at the author’s web site, only the ACM Digital Library should do that. Some people with the ACM were pushed into doing something they would rather not have done.

The ACM still claims that deep-linking directly to an author’s paper is a violation of their policy and is somehow illegal (it isn’t). I don’t believe the rank and file ACM members feel the same way; I’ve yet to see anyone outside of ACM publications defend some of the more publisher-biased sections of ACM’s copyright policies. The value-added by the ACM is to provide peer reviewers (which is done by unpaid editors and reviewers, like me), provide copy editing (beyond reviewers’ comments, rarely done by staff; papers are supposed to be camera-ready for conferences, for example), and distribute the publications (the DL *is* expensive; if it disappeared that would be sad, but Ke-Sen Huang shows that a single person can outperform the DL in the aspect of accessibility, for example). Are these services from the ACM enough to justify forcing authors to hand over most of their republication rights?

As a book author, in the past decade I’ve seen the ACM become much more protectionist and money-oriented. A decade ago there really wasn’t much of a copyright policy, and that was fine by most people – no one in their right mind “steals” copyrighted research papers, and the authors were assumed to have control of their work. Reprinting images was simply a matter of asking the researcher for permission, for example. Under Mark Mandelbaum (now retired) the policy moved strongly to the ACM owning papers and images for redistribution (along with accompanying fees), with authors effectively losing these rights. I plan on posting more about the image ownership issue soon, since image production is a key element for computer graphics researchers.

I’ll add my own posting to the OSTP blog here, which is mostly a comment on Bernie Rous’ concept of balance between benefit to the public and a publisher’s desires. I think my most important point is the last one, that a public government repository will be incomplete (and also, that not all individuals put their work up on the web); serious researchers will need access to all papers, and so will subscribe.

Here’s my post, for what it’s worth:

On “balance”: I do understand the desire for profit by the ACM, to support its other activities. The ACM’s mission statement: “ACM, the world’s largest educational and scientific computing society, delivers resources that advance computing as a science and a profession.” The ACM is a non-profit entity ostensibly run for the benefit of its members. It should therefore not need to worry about balance, as researchers who contribute to its publications are not paid and so would incur no loss of income by wider distribution of their results.

A few quick points:
1. As has been said many times: the basic principle is that taxpayers paid, so they should have free access to the results.
2. ACM’s Digital Library is for pay: $10/article or $100/year for an individual. Cost clearly limits distribution of research.
3. Individuals can put their papers on their websites, it’s true. The ACM disapproves of direct, deep links to these papers (even though this practice is legal). Why is this a problem for the ACM, if their goal is dissemination of members’ research? I believe the answer is that they believe their Digital Library is then less competitive. See http://www.realtimerendering.com/blog/things-i-learned-from-the-acm/
4. Why should there be this limitation of the papers being on just an author’s/institution’s website? If the paper is not being sold for profit at a site, who cares what the IP address is of this site, as long as it has the author’s permission? The advantage of having government funded work in a central repository is clear.
5. A few papers are not available outside the ACM’s Digital Library (not due to willingness so much as due to problems for the individuals to host the works) and cannot be hosted elsewhere by the ACM’s current copyright rules.
6. If an author retires or dies (or simply changes jobs), it’s unclear whether his work can still be hosted by the institution. In this way older papers fall out of public access and become sole property of the ACM.
7. If you were to poll researchers (government funded or not) as to whether they would want wider and more reliable distribution of their work, I expect you would get a 98+% positive response.
8. Finally, the ACM should not consider a free central repository of only government-funded research results to be a true competitor to the Digital Library. By its nature, such a repository will be incomplete. Serious researchers will still subscribe to the Digital Library; other users that simply want to access a few articles a year will have a reasonable alternative.

You have posted the exact same comment on the Boing Boing post (and on your own blog, and perhaps several other places – you do seem to have a knack for repetition). I have addressed your points elsewhere, and do not care to repeat the exercise again here.

(1) ACM is among the 51% of publishers who are completely green on immediate self-archiving: http://romeo.eprints.org/stats.php

(2) For authors — as well as for institutions and funders who are attempting to mandate OA, it makes an enormous difference *where* deposit is mandated, because divergent institutional and central deposit mandates mean multiple deposit for authors, and put funder mandates in competition with institutional mandates, whereas convergent inititutional deposit mandates by both funders and institutions reinforce and facilitate one another: http://openaccess.eprints.org/index.php?/archives/369-guid.html

(3) For users, it does not matter in the least where an OA paper is deposited (as long as the repository is OAI-compliant), because all deposits can be, and are being, centrally harvested, by multiple central OAI harvesters (like citeseer, base, oaister, scirus, google scholar, and the even better central harvesters whose creation will be inspired by Green OA deposit mandates *if we will only help them happen by not over-reaching needlessly instead of grasping what it already fully within reach* (by supporting Green OA institutional deposit mandates, and those publishers, like ACM, that facilitate rather than obstruct them): http://openaccess.eprints.org/index.php?/archives/494-guid.html

Yes, the interests of learned-society publishers like ACM — and indeed those of any refereed journal publisher — are not more important than the interests of research, researchers, their institutions, their funders, and the tax-paying public that funds the funders, but research interests are not well-served if we demonize even the publishers, like ACM, that are already on the side of the angels, nor if we needlessly over-reach instead of grasping what’s already within reach:

Send OSTP and President Obama the simple, convergent message that is guaranteed to bring us universal OA in short order: Mandate immediate deposit into the fundee’s institutional repository the final refereed draft of all funded research, immediately upon acceptance for publication. That’s all.

ACM — unlike 49% of publishers — is not standing in your way.

(And there is absolutely nothing wrong with ACM continuing to produce the fee-based Digital Library to try to compete with the free central harvesters of OA content, just as there is nothing wrong with ACM continuing to produce the fee-based proprietary ACM print and online edition of the articles to try to compete with the OA drafts. The future will take care of itself, but please let us not keep holding it back by gratuitously insisting on more than necessary today.)

I’ll be adding my own comments to the OSTP blog soon, and urge others to do the same. I think the main point, to me, is that federally-funded research should be free to access to the taxpayers that funded it, not $10/article or $100/year via the ACM. A year delay is annoying, but it’s much better than the 95-year copyright that ACM currently has. I don’t consider this proposal a true competitor to the ACM’s Digital Library, as the DL will contain all papers, vs. only federally-funded ones in the OSTP’s proposal. If you asked a researcher if he would want his research to be easily available to the public, I suspect 98% would say “of course” (and the other 2% would be worried about the DL’s profitability).

ACM’s copyright policy is a separate issue, one that deserves attention from all interested people. If the ACM publishes your article, for example, their contract says that only the ACM has the right to permit others to reprint any images in that article – you cannot. This is a pretty key right for a computer graphics researcher, one that film and game production studios certainly never sign away to the ACM but most individuals do not fight, let alone know about.