We’re In Ur Librariez, Controlling Ur Recordz

As I understand it, these conditions are specifically intended to assert effective ownership over catalog data that has originally been compiled by OCLC participating institutions and hence preventing others from using it.

“Preventing” should be “prevent”.

Anything worth saying as an evil copyright-owning publisher and non-academic who *still* thinks it’s outrageous borderline-crooked behaviour on OCLC’s part?

Almost as important as getting the catalogs fixed is getting the interlibrary loan system and the Library of Congress fixed. It doesn’t do much good to know that a book theoretically once existed if you can’t actually get your hands on a copy.

[OCLC is] now demanding that every library that uses WorldCat give control over all its catalog records to OCLC. It literally is asking libraries to put an OCLC policy notice on every book record in their catalog.

Yes, it literally is “asking”. But “asking” = “demanding”? There was a suspiciously feverish tone to Aaron Swartz’s post.

So I went and looked at actual OCLC policy. (I know, I know: primary sources. How old-fashioned.) Note section B, and in particular how OCLC’s definition of “WorldCat Record” concludes, in B.3:

An OCLC Member or Non-OCLC Member may Use or Transfer the following without complying with this Policy: (i) a WorldCat Record designated in WorldCat as the Original Cataloging of the OCLC Member or Non-OCLC Member; or (ii) a bibliographic record which is not Derived from WorldCat whether or not the OCLC Member or Non-OCLC Member adds the OCLC control number to the record.

In plainer English (and I hope without violating the sense): if your library contributed the record to OCLC, it’s still yours to use/transfer; if your library got the record from some non-OCLC source, you can use/transfer it without worrying about OCLC policy. And it doesn’t matter whether your library is an OCLC member or not. (Other policies might apply, depending on the record origin, but that’s none of OCLC’s affair, and they understandably would not want legal exposure in that case.)

Still, the definition of “Reasonable Use” is problematic; see B.13.b: “the term “Reasonable Use” does not include any Use of WorldCat Records that–”

substantially replicates the function, purpose, and/or size of WorldCat . . . .

You can see where OpenLibrary would have a problem with that. Especially with that “size” thing. Size does matter. OpenLibrary is up to about 20 million titles, but WorldCat is over a billion. One could argue that, for all practical purposes, at this point, you can’t really replicate the function and purpose of WorldCat without WorldCat’s size. And yet, for anything like OpenLibrary to become the default source, it would have to approach that size. Somehow. Without “deriving” from OCLC-owned records.

I’m no expert on intellectual property, copyright law, or library science. I haven’t even read the entire OCLC policy. Still, the way I read the basic intent of OCLC’s new policy is: “Everybody’s better off if we’re a virtual monopoly — but don’t worry, non-commercial use won’t be impeded at all. So please support us in not getting copied wholesale by upstarts, now that such a bulk transfer is so much harder to stop.”

With the appropriate caveats (most of which are answered by antitrust law, I think), I’m not sure I have a big problem with this. Maybe I’m too Schumpeterian by half, but “monopoly” is not, per se, a dirty word to me. I think we’re all better off with Google as a virtual monopoly, for example, even though a bunch of people are filthy rich off of Google. I think we’d all be better off if Firefox were a virtual monopoly — even though the CEO makes a stinking half-million dollars a year (no big deal in Silicon Valley according to her.)

At the same time, I’m open to a Schumpeterian argument against OCLC’s new policy: that OCLC ought to be creatively destroyed, that all its arguments about its members’ decades of sunk costs don’t mean squat, because we’re now all better off just letting bibliographic records become a total information commons, with the means of production made radically cheaper; and besides, it’s possible that OCLC, under its own policy, doesn’t even own a very large fraction of the records, maybe most of them are member-contributed and can be freely assigned to other open records projects.

But you know who I’d leave the final word with? Librarians. Not civil libertarians. Not open source enthusiasts. Not even intellectual property lawyers. Librarians. The problem with librarians is sometimes you have to get into a quiet environment and listen carefully as they speak softly, in civilized tones and measured words. But they tend to understand these sorts of issues better than anyone, in my experience.

“monopoly” is not, per se, a dirty word to me. I think we’re all better off with Google as a virtual monopoly…. I think we’d all be better off if Firefox were a virtual monopoly

Count me in opposition to all of these statements. Google is a (mostly) fine outfit and deserves their wealth, but that doesn’t mean we want them to determine the future of search, freezing out everybody else. Firefox is another fine product, but we did have a browser monopoly a while ago, and it was not Firefox, and it was a disaster whose effects are still felt.

You have to do business with the monopolists you have, not the ones you want. And OCLC doesn’t look like one that we want.

Your point about librarians = good is well taken, but even librarians need help now and then.

I went looking for some major non-profit organization that seemed to be reasonably democratically organized, with a history of responsible opinion, with libraries as voting members, devoted to the problems of making bibliographic records readily available on-line. Surely, they’d have a well-reasoned position on all this, right?

And I found a dot-org that matches almost perfectly! From what I can tell, it’s an outfit called the Online Computer Library Center.

Or, uh, “OCLC” for short.

Other fun facts to know and tell: Aaron breathlessly (how else?) informs us that the chief executive for the Whore of Babylon OCLC comes from Deloitte & Touche. Well, no, that’s their exec VP/CFO/Treasurer, Rick Schwieterman. OCLC’s actual CEO and President is Jay Jordan, since 1998; his previous experience in the private sector was with a commercial provider of databases. Hardly inappropriate. Also hardly Rupert Murdoch’s Mini-Me.

In other news (to spell out that last), Somalia, which all-too-quietly surpassed Sudan in the category of humanitarian relief crises last year at this time (with this harrowing disaster largely kicked off by the Bush administration’s irresponsible support of Ethiopa’s invasion) is in an increasingly deadly civil war, with suddenly larger flows of refugees to neighboring countries. So unless OCLC sunk all its members’ fees into bad mortgage backed securities, or played a role in deep-sixing evidence that Iraq had no significant WMD program, I think it’s really time to move along to a real issue. Not necessarily Somalia. Just a real issue. We’re not exactly short of them, are we?

RLG Eureka has been assimilated. See also OCLC’s executive suite: Vice President of RLG Programs Development, Jim Michalko. Formerly president/CEO with RLG, with them for 16 years.

The Borg will even resort to being reasonable and flexible. Clearly, they will stop at nothing.

For what it’s worth, this was just posted on the OCLC-Cataloging list:

From: OCLC-Cataloging [mailto:OCLC-CAT@OCLC.ORG] On Behalf Of Whitehair,David
Sent: Friday, November 14, 2008 3:14 PM
To: OCLC-CAT@OCLC.ORG
Subject: [OCLC-CAT] OCLC Record Use Policy

The following is being posted for Karen Calhoun, OCLC WorldCat and Metadata Services:
Dear OCLC-CAT readers,
You may have seen a posting from Aaron Schwartz of Open Library on his blog; Aaron is reacting to OCLC’s updated policy for the use of WorldCat records. We respect the perspective that Aaron brings to the important discussion of information access. We are likely in solid agreement with him on what is probably the central point of his most recent blog posts — increasing information access to users around the globe.
Increasing access to information and to the world’s libraries is OCLC’s mission. OCLC, like most communities and organizations, establishes guidelines and policies to support the goals and obligations of its membership. OCLC needs to update guidelines as the environment and the tools of the Web evolve.
The purpose of the Policy for Use and Transfer of WorldCat Records is to update the current Guidelines for the Use and Transfer of OCLC-Derived Records, for purposes of facilitating the freest possible use of WorldCat bibliographic data while maintaining the strength of the OCLC cooperative and the network effects generated for libraries by WorldCat.
What is of concern is Aaron’s misinterpretation of the facts surrounding the policy.
OCLC would like to clarify several points about the policy that have been incorrectly stated on Aaron’s blog. Specifically, it is important that the library community is aware of the following:
1. Access to WorldCat is freely available to all at http://www.WorldCat.org

2) There is no intention to restrict the downloading of bibliographic information from WorldCat-derived records into Zotero and other bibliographic software. As is true today, there will be no impact on noncommercial use that supports learning, teaching, academic research, scientific research, private study, verification of bibliographic information, development of bibliographies, and similar pursuits.
3) There is no intention of ceasing free access to WorldCat data and to OCLC member library collections through WorldCat.org. In addition, there is no intention to constrain the operation of union catalogs and resource sharing systems that have been in place for years, nor does the revised policy impact noncommercial record exchange between libraries using protocols like Z39.50. Our FAQ covers many use cases that may be of interest to this list’s readers.
4) The Policy carries forward the principles, if not the wording of the current Guidelines, which have been in place since 1987. The changes from the Guidelines to the Policy may be summarized as (a) increased access to WorldCat data for museums and archives; (b) clarification of conditions for data sharing; and (c) provision of an easy method for proposing new uses that aren’t covered by the policy.
5) The Policy provides a mechanism for commercial use, provided that the use is in the best interest of libraries, benefits members and is consistent with OCLC’s public purposes.
6) The existing Guidelines remain in effect until mid-February. During this period we will continue to listen to comments from OCLC member libraries and other key stakeholders about the policy. OCLC has already made substantive changes in response to concerns raised by members.
7) OCLC welcomes collaboration with Open Library. We respect the goals of Open Library and believe that our members and other libraries, museums, and archives could benefit a great deal from cooperation between our two organizations.
The FAQ for OCLC’s updated policy addresses these topics and more here: http://www.oclc.org/us/en/worldcat/catalog/policy/questions/default.htm And the updated policy can be found here: http://www.oclc.org/us/en/worldcat/catalog/policy/policy.htm; the Guidelines are here: http://www.oclc.org/us/en/support/documentation/worldcat/records/guidelines/default.htm
A podcast of a discussion of the policy was held yesterday between Richard Wallis of Talis and Karen Calhoun and Roy Tennant, of OCLC, and can be found here http://blogs.talis.com/panlibus/archives/2008/11/oclc-talk-with-talis-about-the-new-record-use-policy.php
If you have questions about the policy, please contact OCLC at recorduse@oclc.org
–Karen Calhoun, OCLC WorldCat and Metadata Services

David Whitehair
Global Product Manager, Cataloging & Metadata Services
OCLC
6565 Kilgour Place
Dublin, OH 43017-3395 USA
Voice — +1-614-764-6483 or 1-800-848-5878
Fax — +1-614-718-7292
Email — david_whitehair@oclc.org

Thanks for the correction, Aaron. I’m not an expert in these things. Oh, and by the way, if by “senior executive” you meant “top manager who’s been at OCLC longest”, yes, you’d be right: from Deloitte & Touche. As CFO/Treasurer, however, his biggest question about new records policy would probably be: “Will I still have beans to count?” And he’d probably be satisfied with “Yes, actually, that’s kind of the point.”

Now, as long as you’re here, Aaron, could we have some details on those threats of legal action against OpenLibrary by OCLC? How about some examples of where they’ve “insulted” OpenLibrary in the press? Neither are immediately visible in a Google News archive search, so far, but maybe I’m not searching right. BTW, pointing out objectively verifiable shortcomings of OpenLibrary isn’t necessarily “insulting.”

I’m sure there are cases to be made for and against OCLC, both for its specific behavior in this case and about the viability of the organization as a whole. I’m reading those cases. (I even offered some of my own pro and con, above.) One thing should be pretty clear, though, after reading some of the more level-headed exchanges and observations: OCLC isn’t out to “steal” our libraries from us. It’s not some “monster”. It’s a dues-paying membership organization that offers services to its members, and is apparently trying to work out long-term survival issues in an evolving technological landscape.

Maybe OCLC can’t survive in its present form. Maybe it shouldn’t. Maybe it’s too big. Maybe it charges too much. Maybe its executives are paid too much. Maybe it’s a bloated, overreaching bureaucracy that now owes its survival mainly to bureaucratic intertia in library systems, as suggested above. But “monster”? Library-thief? “Sorcerer’s apprentice”? Wow, so this is the most dramatic thing to happen in the world of libraries since that big one in Alexandria burned down? Thanks for the heads up. Mind if I go back to reading more sane-sounding stuff now? (Which is to say, almost everything that almost everybody else has written on the subject.)

OK, before I go, let me offer a couple of crazy conspiracy theories of my own.

Note that in the OCLC’s definition of “Reasonable Use”, there’s a reference to a certain internet protocol for bibliographic database queries. To the inexpert eye, it might seem a very minor technicality. It looked a lot to me like a Dewey decimal number (snooze) and isn’t even identified as a protocol in the policy text. When I looked into it, though, this protocol seemed to me to provide a way to quickly and easily siphon off huge bibliographic databases, possibly pre-laundered so that record origins were obscured. You just poke around the world of open ports for that protocol, looking for libraries whose servers strip out the record origin. It is, after all, a known problem with this protocol: differing implementations return different results on the same query. So that’s conspiracy theory #1: some people are out to do just that.

Here’s conspiracy theory #2: the group at OCLC tasked with updating the records policy noticed this very possibility themselves. They realized that if they opened up discussion of new records policy to any and all, they’d have to talk about this otherwise obscure protocol (or it would inevitably come up). Any such discussion would be widely advertising a kind of OCLC security hole, eventually if not sooner. Could they send out “quiet” requests for the relevant port to be more tightly policed, without a legal pretext, before the discussion went public? Nope, that’s just asking for leaks. So they decided: avoid public discussion, the new policy wouldn’t be “opt-in”, and it would be in force from day one of the policy’s unveiling. Perhaps they also decided that, on day one, they’d send out warnings to library computer system administrators to shut down the relevant ports as a precaution against litigation. But now the cat’s out of the bag, isn’t it? Now, they just have to rely entirely on the essential goodness of human beings, on the kindness of strangers.

It’s just science fiction, I haven’t really researched it that deeply, pay it no mind . . .

Keith, I said almost nothing about how the protocol works, so I don’t know how you get that this “doesn’t work quite the way you think it does” without making some assumptions about what I think.

As for “hardly anyone uses” Z39.50, I suppose The Library of Congress is “hardly anyone.”

Anyway, glad to hear we don’t have to really consider it within the realm of possibility that someone could find an unsecured Z39.50 port and siphon off records, since to do so would require “bitchin hakzor sk1ll5.” If hardly anybody has such skills, we don’t need to update our antivirus software (finally, hooray!)

Besides, people with such skills, upon arriving at Mike Taylor’s Net::Z3950 Perl Module page would probably stop reading at his “Do Not Use This Module!” warning about how that Perl module is obsolete. And even the ones who got past that and noticed that there’s a better Z39.50 component now would still balk at how ancient it is. Why, there hasn’t been a release of ZOOM-Perl since way back in . . . uh, June of this year?

Well, anyway, even if you could get ZOOM going, there are still some things we can be sure of. Not only is the Z39.50 protocol decrepitly senile now, hardly supported by anyone’s servers, it was also, as you say, only designed for “dedicated record sharing in a secure network”. So the last thing OCLC itself would do would be to make available a Z39.50 port for anyone to . . . uh . . . scratch that.

Well, I think it’s best if I leave the issue here. Otherwise, I might be tempted to waste the afternoon cobbling up something with ZOOM using my slightly rusty Perl skills. (Or my slightly less rusty C++ skills? Or maybe learn enough Python to get by? Looks like there are supported Z39.50 APIs for a number of languages. Admittedly, all old languages. It’s not like you’d find anything for one of those kewl new languages like Ruby, y’know, Ruby Zoom or something).

No, I should resist temptation. Why, if I got any traction at all, I might start thinking I have bitchin hakzor sk1ll5. I’m 52 years old, it wouldn’t be dignified.