Seeing parallels
Steve Lawson wrote this wonderful piece for his blog See also..., reprinted here (by permission) in full:
There is a large organization whose main business isn’t producing information, but instead hosting and aggregating information for many thousands of users on the web. Users upload content, and use the service to make that content public worldwide, and, likewise, to find other users’ content. Then one day the large organization decides to change the rules about how that information is shared, giving the organization more rights–to the point where it sounds to some people like the organization is trying to claim ownership of the users’ content, rather than simply hosting it and making it available on the web.Perfect, just perfect.
A small but vocal and influential group of users object to the policy change. The organization protests that it isn’t their intent to fundamentally change their relationship with their users and that legal documents tend to sound scarier than they really are. Most customers are either unaware or unconcerned by the change in policy, but the outcry continues until the organization backs down a bit, sticking with the old policy for the time being. The future, though, is up in the air.
Facebook? Or OCLC?
Labels: facebook, oclc, open data, steve lawson
posted by Tim at 9:52 PM
17 Comments:
"Legal documents tend to sound scarier than they actually are." "Sound" is not involved, here. The new Facebook TOS was plain and unequivocal. Facebook was flatly asserting that it had ownership rights over user content, in perpetuity and irrevocably. That doesn't just "sound" scary; that is scary, and, more to the point, it's morally wrong. (Shrinkwrap licenses, asserting that you can change the license without notification, etc. See xkcd, http://xkcd.com/501/.)
Agreed, Jonathan. But both organizations tried to go that way, I'd say. I'll point you to Mark Zuckerberg saying "we think that a lot of the language in our terms is overly formal and protective so we don't plan to leave it there for long," and Karen Calhoun's "The updated policy is a legal document. Being a player on the Web, working on behalf of libraries, requires that the policy be a legal document. I know that some librarians will be uncomfortable with that, but it is necessary."
So Tim,
What happens when someone makes you an offer you can't refuse? Additionally, what percentage ownership do you have in LT? Unless it is 51%, one of the other owners could always buy out the others, at which point, you're a minor player. I you are impervious to offers of the first kind. I hope the second situation does not obtain. I would hate to think all the work that has collectively been put into this enterprise would suddenly become privatized. Or that we would suddenly be deluged with a score of ugly advertisements cluttering up our sites.
Right. That's why you want *licenses*. For user data, the LibraryThing license is non-exclusive. You can take your data off LibraryThing and put it somewhere else. The OCLC license isn't like that. Once it's on OCLC it's theirs forever and anywhere that data goes must use it under the same terms—a permanent, viral lock-down.
As regards non-user data, much of it is released as Creative Commons, which means it is virally free—it cannot be locked down by anyone.
Tim said:
"Once it's on OCLC it's theirs forever and anywhere that data goes must use it under the same terms—a permanent, viral lock-down."
I disagree. OCLC data can be used non-commercially - that is explicit in the proposed agreement. It can be used commercially by forging an agreement. It just can't be taken by a commercial company or for commercial interests without agreement.
To Steve- the Facebook agreement is not about allowing anyone to use their data non-commercially while restricting commercial use to agreement so the analogy is inaccurate, I'm afraid. While I can understand the tendency of some to wish to draw a superficial analogy, I say no - since you ask. ;-)
I disagree. OCLC data can be used non-commercially - that is explicit in the proposed agreement. It can be used commercially by forging an agreement. It just can't be taken by a commercial company or for commercial interests without agreement.
"OCLC data"—which is as valid a concept as "Facebook data"—is subject to a permanent, viral license. Non-commercial use must follow the agreement, and pass on the agreement, forever.
This license by no means allows unrestricted noncommercial use. For example, OCLC members may only transfer records that are of their "own Holdings"—just what you physically own, and such transfer must not have the effect of "discouraging the contribution" of records to OCLC, or being similar in "function, purpose or size" to WorldCat. Thus, libraries can give their data to OCLC, but any attempt by libraries or anyone else to form or involve a service similar in "function, purpose or size" is forbidden—and forbidden always and virally. This language is why it's unclear whether libraries can form state-wide catalogs, or other record sharing that doesn't explicitly involve OCLC. It would certainly preclude giving your records to OpenLibrary which, although a non-profit and indeed a registered library, is similar in size and function to OCLC. It is, as noted, fortunate that OCLC ate RLG before the terms came out, as it would have illegalized record sharing with RLG libraries and museums.
It's also worth noting that "non-commercial" is a *defined* term in the contract, meaning "research, teaching, scholarship or private study." That's not non-commercial use, but a defined subset with no possibility of threatening OCLC.
http://openlibrary.org/about/help/oclc
From OpenLibrary's website:
"The OCLC record usage guidelines state:
Each member and nonmember library may use records without restriction, and may transfer records of its own holdings without restriction to other libraries.
In addition to transfers to libraries, each member and nonmember library may transfer records of its own holdings without restriction, to (a) member networks, (b) state and multi-state library agencies and (c) all other noncommercial firms.
The Internet Archive (the nonprofit that's operating the Open Library) is both a registered library in the state of California and a registered nonprofit organization."
Commercial entities don't get a free ride to use the collective work (1.1 billion records) of the thousands of member libraries without an agreement with OCLC. Is this a problem?
Right. That's the Guideline. The policy ends the Guideline, and institutes the Policy. That's what we're talking about, and why there's any fuss.
There's a minimal level of understanding of an issue necessary to be a provocateur on that issue. I'm afraid that you do not possess it.
[Enjoys watching Tim bitchslap people[
I'm frankly surprised about all this "new TOS" kerfluffle over at Facebook--I'd been hearing for a long time that the old TOS definitely allowed them to hold onto "deleted" content and to assert rights and so on and so forth. Believe me, I would have gone off Facebook a long time ago if I planned on running for office or if I believed in copyright. I'm actually concerned that all the uproar over the "new" TOS is just going to let Facebook revert to the old rules, where they had the right to all your material but kids didn't realize that's what it said.
Bitchslapped said "Commercial entities don't get a free ride to use the collective work (1.1 billion records) of the thousands of member libraries without an agreement with OCLC. Is this a problem?"
Well, I responded to the main point—which revolved around simple ignorance of the issue.
To your point, yes, it's a problem. Most of the records were created by government institutions, predominantly the federal government, for the benefit of all. Their records are legally impossible to copyright. So OCLC can only lock them down by instituting a bizarre license scheme that attempts to make third-parties "agree" to something they never agreed to.
More generally, I fail to see why, if library data is to be free, commercial companies may not take advantage of it. Libraries make their collections open to commercial and non-commercial interests alike. For the thousands of Carnegie Libraries, for example, the "progress of industry" wasn't a side effect but a *stated goal*!
It should be added that the issue is a bugaboo and a distraction. Commercial companies challenging this policy have all embraced open data. Nobody is trying to appropriate library data and sell it—other than OCLC. Commercial companies, like LibLime and Talis release all their records for unrestricted use—and permanently bind them to remain unrestricted.
It's also quite doubtful that OCLC is locking down its records and creating a permanent monopoly out of a desire to squeeze a few pennies out of LibraryThing. As the Policy makes perfectly clear, their real fear is Open Library—a non-profit and library. Raising the specter of evil companies is just a way to distract people from the truth—hosting and sending tiny text files other people create is no longer hard, and certainly doesn't justify the hundreds of millions of dollars libraries give OCLC.
Tim sez "s the Policy makes perfectly clear, their real fear is Open Library—a non-profit and library. Raising the specter of evil companies is just a way to distract people from the truth—hosting and sending tiny text files other people create is no longer hard, and certainly doesn't justify the hundreds of millions of dollars libraries give OCLC."
Bitchsplapped replies: Ignorance is bliss.
Tim -
If you're looking for history, things are not good. I'd point you to CDDB (which started out as user-contributed data, got bought by Gracenote, and now is part of Sony) and also to IMDB (a Usenet project that grew into a company and is now part of Amazon).
OCLC owns all that data in much the same way that you own my data that I typed in (with the bloody stumps of my bloody fingers). As an organization it is dependent on making use of what it has to continue its own existence. Just the way it goes.
Ed,
I agree there are some bad precedents. Two points though:
LibraryThing does have a non-exclusive right to use your data pretty much as we want (some privacy protections excepted). But you actually entered the data into LibraryThing, and when you did so you knew what you were doing. OCLC members thought they were sharing records with *each other*, and now find the pipe is the master. And if LT can use your data, it's control is not exclusive and it does not try to bind others downstream--you can give your data to Goodreads, and Goodreads is not suddenly violating some license of ours.
Can I give away Edward's LT data also?
If Edward wants to share his data with you, sure. The OCLC Policy prevents sharing data, except as regulated by the Policy, among other things preventing all commercial sharing and any sharing with an entity that is similar in size, purpose or function to WorldCat. It also effects publicly-funded data, which is legally required to be shared.
If OCLC allowed non-public libraries to indicate that their data couldn't be shared, then at least that part that is covered by copyright (the minority of the data, but some of the more interesting stuff), should be unshare-able. LibraryThing of course would pull the link to any library data source (open z39.50) if the library asked. They never do—indeed they send us new Z39.50s unsolicited). Libraries want to share. It's their new master who wants to stop it.
Post a Comment
<< Home