Friday, July 30, 2010

SkyRiver/III v. OCLC, Part II

In my previous post I covered what I saw as the stronger arguments made in the complaint. In this post I will cover points that either puzzled me or seemed to be off the mark.

The OCLC Number
The complaint states that
"This OCLC number has permitted OCLC to police its members to ensure that their records are not shared with unauthorized users." (p. 5)
Since anyone can add or delete an OCLC number from a MARC record in their own database, I don't see how this could be the case. I would like to see how this claim is supported.

The ILS Market
"OCLC is rapidly gaining market share in the ILS market by leveraging its monopoly power over its bibliographic database... " (p. 6)
Can they supply the figures to support this rapid gain in market share? They do state the number of WorldCat Local installations ("624", p. 22), but WCL is not an ILS (even though it may eventually become the basis for one).

Academic Libraries only
The complaint appears to only address academic libraries. (p.7) This could be because the evidence that they claim to have only relates to academic libraries, but both OCLC and III serve many public libraries. The complaint also states that:
"The relevant geographic market ... is the United States, because academic libraries cannot turn to suppliers of these products in other countries to meet their needs." (p. 10)
This may just be poorly worded, but if it intends to mean that there are no extra-US companies providing the service then it should have said so. The way it is worded it sounds like there are prohibitions on using non-US suppliers that pertain to academic libraries... could that be so?

New Products
In numerous places in the document, the complaint states that OCLC members are required to participate in product development as part of their membership obligation:
"Membership also obligates libraries to assist OCLC in developing new products and services to compete with for-profit firms." (p. 5)

"OCLC developed, and is still developing, WorldCat Local and WorldCat Local "quick start" through pilot programs in which many of its member university libraries have agreed to participate, without compensation, purportedly to meet the requirements of their membership in OCLC." (p. 20)
I have never heard of this requirement, and would be interested in hearing from institutions who did find themselves essentially forced to participate in pilots as part of their membership.

Acquisition of Other Companies
The complaint states that over time OCLC has expanded by acquiring 19 library industry companies, 14 of which were for-profit. (They fail to mention that at least some of those companies magically became non-profit when acquired by OCLC, cf. netLibrary.) The remainder of the sentence reads:
"... either to obtain software and other products that enable it to offer library services in competition with the remaining for-profit providers or simply to eliminate products from the marketplace." (p. 23)
These are strong words that the complainants should be prepared to prove. I'm not saying that it isn't true. However, in the few cases of which I am aware (WLN, netLibrary, RLG) the acquired company was in financial free-fall and OCLC's purchase was viewed at the time as a rescue that benefited the library community as a whole. In the case of netLibrary, OCLC had agreed to be the escrow agent for the ebooks purchased by libraries, to be called upon should netLibrary go out of business. In that case, OCLC was pretty much pre-obligated to rescue netLibrary or provide some service of its own. (I don't know what the monetary arrangements of the escrow were.) As for WLN and RLG, it's hard to know what would have happened if OCLC hadn't purchased those agencies. I suspect that the libraries using those services would have had to become OCLC members in any case in order to continue functioning as libraries. This only covers three of the 19, and may or may not be representative of OCLC's acquisitions.

[Partial list of acquisitions, gleaned from press releases and annual report:
Dewey Decimal System (1988), Information Dimensions (1993) [sold in 1997], Public Affairs Information Service/PAIS(1999), WLN (1998), netLibrary (2002, with MetaText eTextbook Division, a for-profit subsidiary), Openly Informatics (2006, OpenURL services), RLG (2006), EZproxy (2008), Amlib (2008, Australian web-based ILS), PICA (1997), Fretwell-Downing (2005), Sisis Information Systems (2005). Note: these may not be the same companies referred to in the complaint. This is my cobbled together list, and should only be seen as such.]

Another strange statement is about OCLC's use of head-hunters to hire staff away from other companies:
"In addition to acquiring for-profit companies, OCLC also uses headhunters to identify and recruit employees from for-profit firms. Plaintiffs are informed and believe and based thereon allege that OCLC is using its tax-free dollars to recruit employees of for-profit vendors of library services to eliminate competition and extend OCLC's monopoly to the ILS market." (p. 26)
There's obviously a story here, but I don't know what it is. Using headhunters is standard industry practice for a well-heeled high-tech organization. Has OCLC engaged in predatory hiring behavior? And can that allegation be proved?

Access to WorldCat
The strangest thing in this complaint is the repeated insistence that OCLC should give access to the WorldCat database to potential competitors.
"...As a result of OCLC's conduct... Innovative [and SkyRiver, in another paragraph] has suffered and will continue to suffer irreparable harm ... unless this Court orders defendant OCLC to provide access to the WorldCat database to Innovative and other competitors, on such terms as are just and reasonable." (p. 31; same but ref. to SkyRiver p. 29)
This argument comes as a surprise to me. I had always assumed that the goal was to allow libraries to provide their bibliographic records freely to anyone they wished, including for-profit companies. I see that as very different from giving competitors direct access to WorldCat. It seems to me that the former goal would be very easy to argue, but direct access to OCLC's own database seems much more difficult to justify. I'm quite puzzled by this, unless I am drawing the wrong conclusion about what it means.

There's a part of me that wants this to go to court so that we can get answers to these intriguing questions. There's another part of me that sees the possiblity that this could be a lose-lose proposition. Given the overall stress in the library community, both monetary and technological, in-fighting looks to be the worst thing we could do to ourselves.

There is no doubt that a large, union catalog of library holdings is key to providing the kind of web-scale (sorry, but I couldn't think of another word) services that libraries absolutely must provide today. That said, that database does not have to be WorldCat, although WorldCat performs that function at this moment in time. The main thing is that we must have a union/universal catalog that serves libraries and their users. It shouldn't be a limited access asset that is being fought over for market share. I don't have a solution to offer, but it's clear to me that the solution is: FREE THE DATA.


Jennifer Parsons said...

I'm not sure if the infighting will be harmful, actually, because I'm not sure if OCLC *or* SkyRiver/III could be counted as parts of the "library community"-- yes, they work with libraries and librarians, but their aims and goals are quite different from the people they serve. They're almost a class unto themselves.

But I do agree that the solution to this would be to simply free the data, since it's contributed and used by OCLC users to begin with, and not necessary the organization itself. Doing such a thing might actually be helpful to OCLC in the long run, as that way it could legitimately claim non-profit status. Certainly it would be the only way out that would benefit *libraries,* if SkyRiver's profit margins.

Anonymous said...

RE: Access to WorldCat
There are repeated arguments in the complaint saying that OCLC has been utilizing its existing advantage (aka WorldCat) to build add-on services (e.g. ILL, ILS). Since WorldCat is the de facto bibliographic database of the world, the plaintiffs claim that it is impossible for them to create/provide services that are competitive to those provided by OCLC because of the pre-existing disadvantage. Therefore, seeking access to WorldCat by the plaintiffs can be seen as a way to level the playing field. However, why not asking OCLC to allow member libraries to supply records to third party commercial vendors? From the efficiency standpoint, granting access to WorldCat could avoid the cost to reinvent the wheel by replicating all the bibliographic records and holdings in WorldCat. However, from a conspiracy theory viewpoint, the answer probably is the difficulty (in terms of both technological and financial burden) to build and maintain a quality bibliographic utility at the scale of the current WorldCat, or, even worse, a flat-out unwillingness to invest in data stewardship but willingness to be a free rider. Skyriver is a very good example to show how difficult it is to maintain a quality bibliographic database. In terms of both quantity and quality of bibliographic records, Skyriver is far behind WorldCat. Skyriver looks like merely a warehouse of bibliographic records instead of a WorldCat-like database that has ongoing maintenance of headings in bibliographic records, accurate holding information, and sophisticated deduplication mechanism (I understand that dedupping in WorldCat does not always work as it is supposed to be; however, I can assure you the probability of finding duplicate records in Skyriver is much much higher than in WorldCat). Sincerely, I hope the reason is more about efficiency than monetary cost-benefits calculation.

Karen Coyle said...


I agree that it would be very difficult for someone else to build up the "asset" that is WorldCat. OCLC has a forty year head start on that. I also agree that WorldCat is not just a bucket of bibliographic data. Considerable effort (both on the part of libraries and OCLC) has gone into the ongoing maintenance of the data over time.

Ideally, that set of data should be considered an asset of the library community that has contributed to it. Were OCLC not in competition with vendors in the library arena, it might be as simple as charging a fee for vendors to make use of the database, to the benefit of all. But as long as WorldCat "belongs" to OCLC, I don't think this will happen. The reason why the solution in the complaint strikes me as "off" is that I can't imagine a court forcing any institution to open its primary asset to competitors.

The best solution, in my mind, would be to determine that WorldCat is not owned by OCLC but by the libraries that contributed to it. WorldCat would become a community asset, silo'd from the services provided by OCLC and other vendors. The companies would compete on quality of service, and the silo'd database would be supported through fees contributed by all users. This would be somewhat like the way that companies are sometimes forced to spin off segments or departments in order to avoid anti-trust violation.

Obviousoly, OCLC would need compensation for this, and I don't know how one would calculate that figure. At the same time, OCLC has been capitalizing on the labor and skills of its contributing libraries for decades, so it's not like they haven't benefited already. Using this solution, the benefits would presumably return to the community rather than to OCLC. Governance would be something of a nightmare -- financial decisions would be simple compared to the bibliographic decisions that would need to be made.

Still, I think that separating WorldCat from OCLC's for-fee services is probably the only way to go.

Jodi Schneider said...

One problem with separating OCLC from the WorldCat database is that while libraries own the data, maintaining it is a separate project that someone needs to take on. That requires resources (time and ideally research money).

Metacomment: Analysis of the case would be a good use of the Commentpress WordPress plugin - we could put up the document and comment inline, paragraph by paragraph.

Karen Coyle said...


I'd love to put it on CommentPress but we've only got a PDF. Any ideas?

As for what it might mean to "spin off" WorldCat, yes, it's not just an inert lump, but a dynamic database that needs quite a lot of care and feeding. Ironically, this solution would lead us to create the organization that I think some people wish OCLC was, and that OCLC claims to be: a non-profit whose main purpose is to take care of the world's largest union catalog for the benefit of libraries.

What we don't know (and the financials that I can access do not reveal) is to what extent the various parts of the OCLC pie are necessary to support WorldCat... or if WorldCat and its services (cataloging and ILL) are supporting the other parts of the pie. It would be fascinating to sketch out a picture of an independent WorldCat and see if it's even economically feasible. Unfortunately, only OCLC has the necessary information that would make this analysis possible.

I would like to see either proof or refutation that the OCLC that we have today is the only way to make this giant catalog work. I would like to understand what options exist for the library community. I do think that massive sharing of data is essential for the future of libraries. I would like for that to be a topic of discussion and brainstorming. It's very frustrating to not have that opportunity.

Anonymous said...

someone should post the worldcat database to wikileaks.

Anonymous said...

I think predatory hiring behavior may refer to Ted Fons (former III ERM Product Manager) leaving III for OCLC a couple of years ago.

K.G. Schneider said...

"I would like to see either proof or refutation that the OCLC that we have today is the only way to make this giant catalog work."

I'm with you. There could be another model. I know what I don't want, which is OCLC dismantled without a solution imminent.

JML said...

Are we sure we need a giant database in the sky? We have fast networks, and cross-platform search capabilities, and consortial arrangements already in place? I think that is exactly why OCLC got rid of the regionals ... they were their own competition.