As my text got to long for posting it as a comment on Karen's blog I am posting it here (simultaneously being my first post on this blog in English).
First: It was me that asked the questions under the name "Adrian" in the OCLC community forum concerning member libraries opening up their data under a public domain licence. And I definitely agree with Karen saying that the Record Use Policy Committee or any other body representing OCLC has not (yet?) answered my last question in this forum from April 16th 2010. The Committee rather postponed an answer and said that my question would be adressed by a reworking of section 5 of the policy. Albeit, since then this section hasn't changed a bit.
I haven't read the OCLC Motion and so I won't and can't go deeper into this discussion. But it is interesting that somehow related questions popped up in the debate about Karen's assumption that all libraries which do Linked/Open Data aren't bound by the OCLC policy. (See the comments on this post by Karen). I think the commenters attack the wrong person when they accuse Karen of making a statement that isn't fact-based. Maybe one should pose the question why it is so hard to find out the facts about who actually is an OCLC member and thus bound by the policy. In this post, I'll first try by myself to figure out what constitutes an OCLC membership.
Searching for criteria of OCLC membership
At first glance, it seems to be easy to find out the answer to the question who is actually bound by the policy. The policy itself makes clear at the beginning that it refers to the "OCLC cooperative":
"The policy is intended for the OCLC cooperative, which refers collectively to OCLC members, the OCLC governance structure (Board of Trustees and Global and Regional Councils), and the non-profit OCLC corporation."Since July 2009 an OCLC member is defined as follows:
"Institutions worldwide become members of OCLC by contractually agreeing to contribute intellectual content or share resources."Therefore, I might possibly assume that contributing data to WorldCat is a sufficient condition for being an OCLC member. But the membership definition also states:
"OCLC services and products qualifying for membership status are identified annually. These services and products are proposed by OCLC staff, reviewed by the Membership Committee, and any changes are approved by the Global Council and the Board of Trustees."As I understand it, the group of libraries which are OCLC members might change each year with the publication of a list of products and services qualifying for membership status. That seems to me quite strange in itself. Furthermore, the quoted sentence isn't linked to the current list of products and services qualifying for membership status - if such a list really exists. (Anyone?) Batch uploading might - as Karen shows in the case of the British Library which as per OCLC's annual report 2009/2010 (p.15) was the biggest batchloader to WorldCat - not necessarily constitute membership although OCLC's batch processing product page lists the point "Participate as fully as possible in the worldwide OCLC library community" which couldn't sound more like membership to me.
It seems that clear criteria for identifying OCLC members don't exist first and foremost the mentioned list of "services and products qualifying for membership status" doesn't exist openly accessible online. I can't easily find out who's an OCLC member and who's not simply by looking at the services and products they buy from OCLC.
A list of OCLC members?
I might assume that all organizations listed in the search for OCLC libraries are OCLC members and thus are bound by the policy. If this was the case then the British Library would be an OCLC member as it is listed in this index together with Konstanz University Library, the Tübingen University Library and the BSZ (the library service center of the SWB library network both libraries belong to) who also have already opened up their data.
To be clear: I am looking for clarity about OCLC membership, be it clear criteria that have to be met to be a member or a simple index of OCLC members. Contrary to Karen, I would not be the least bit surprised if there actually were OCLC members which are bound by the policy among the libraries doing Open (Linked) Data. It even is most probably the case that Konstanz University Library, Tübingen University Library as well as all other libraries in the SWB library network are OCLC members because the BSZ not only batchloads the union catalog data into WorldCat but they even copy catalog from WorldCat. The BSZ might simply have bargained a contract with OCLC which permits them to do what they want with their data while also being an OCLC member, which would be a very sensible move. (I'd be very happy if somebody from the BSZ might leave a comment here concerning this.)
It's worth to consider that the OCLC corporation might not be the powerful player on the European market as it is in the US and that because of this presumed weaker bargaining position of OCLC in Europe contracts in Europe and the US might differ significantly...
Uncertainty & doubt
The underlying general problem is illustrated very well when Karen says: "I don't think we have a way to know, with one quick search, who is and isn't covered by the agreement". If there exist clear criteria for OCLC membership or a database or list of OCLC member institutions these aren't publicly accessible. In my view, it shouldn't be the case that you have to do weeks-long research and interview several people to get answers to these questions. One might think that a cooperative in the domain of public memory institutions has an open and up-to-date list of its members...
To sum it up: We have some understanding of the policy's intention, what it allows, proscribes and demands. And it is clear that the OCLC policy refers to "OCLC members" but
a) we often don't really know whether an institution is an OCLC member or not or whether it has another relation to OCLC and
b) we don't know how a breach of the policy by one of the members might be sanctioned. There's no document saying what will happen. (But we know what OCLC did to a long-term member which decided to change the record supplier for copy cataloging...)
So, there's some uncertainty and doubt to be dealt with. Hopefully OCLC (members) will work on removing it by providing the relevant information...
 This thread on NGC4LIB about OCLC members and participants displays the same uncertainty about this question.
 A consistent definition of "open" in "Open Data", "Open Content" etc. is the Open Knowledge Definition. I stick to this definition when talking about Open Data which implies that German and French national libraries' data isn't (yet) open. And furthermore - following the Open Knowledge Definition - Linked Data isn't necessarily Open Data while Open Data mustn't be Linked Data, see Linked Open Data star scheme by example for more clarity on this.
 See https://wiki.bsz-bw.de/doku.php?id=v-team:katalogisierung:worldcat.