[rights-requirements] RE: [rights] tyranny of the majority but noconsensus

["Radbel, Dmitry" <dmitry.radbel@umusic.com>]

Patrick,

I am sure that out of 100's of e-mails and documents I have missed much more
than a couple of items. Each of us can sift through the "paper trail" and
find something. I believe that Bob and yourself were implying that there was
a pattern of inappropriate behavior. I see no such pattern in the documents
on the website. If Hari at time did not do what you wanted him to do or
allowed a motion to proceed when you did not like the motion - that's not a
justification to question the quality and integrity of his work. If saying
things like "[Hari] has steadily lost the impartiality with which he began
as the TC chairman", "As evidence of that lack of interest, consider that
Hari has attempted to map all the submitted requirements to that initial set
of requirements", "do you believe that Hari would receive a vote of
"confidence" to remain as chairman from a majority of the eligible members?"
is not an attack on him and the work that he's done, then you and I just
disagree on what it means to "attack" someone.

One item that you quote shows that the schedule has been revised in June or
so extended the timeframe for submitting requirements by four weeks. So? As
for the second item quoting unnamed sources - I simply pointed out that,
unless someone has maliciously edited the website, the schedule for
gathering the requirements documents (8/7) and dispositioning them (8/21)
has been posted at least on July 9th and the actual documents from Hari
match that.

Regarding the royalty-free issue - there is a fine distinction between the
language itself being royalty-free and its implementation being
royalty-free. Re: the former, I've heard in the F2F that at least XrML's
contribution will be RF. The latter is a totally different issue and - in my
humble professional opinion - it is highly unlikely that any system
implementation using any rights language will not touch on someone's
patents. Whether this distinction is at the root of the disagreement, I
don't know.

What puzzles me most is the type of debate and arguments that are being
made. This is a TECHNICAL Committee. In my technical training, design issues
and requirements are discussed based on specific, concrete, actionable and
"evaluatable" arguments. I don't know how statements like: "The consistent
pattern has been: We resist any effort to make our "rights-expression
language" actually be expressive of actual rights", "XrML embodies a pinched
view of the world", "the entire committee is a sham", "The time has come to
put to bed the recurrent notion of a "Version 1" of the requirements" can be
constructively evaluated and/or contribute to creating a rights language.

How about a simple exercise? I listed below the "core" requirements from the
latest document dated 9/17. These are language requirements (not system
implementation requirements). There are very general. Can someone point out
expression of which rights being precluded by this set (i.e., can't be added
in extensions)? Can someone point out the negative public-policy
consequences which would result from adopting these "core" requirements?
Perhaps if we concentrate on the specific and constructive issues we can
actually find a ground for consensus.

Respectfully,

DR

List of the core language requirements:

1. Specifying Conditions
2. Specifying Verbs
3. Specifying Nouns
4. Specifying Permissions
5. Extensibility
6. Machine Readable
7. Language Definition
8. Well-defined Semantics
9. Roles, Groups, and Attributes
10. Syntax Does Not Impart Semantics
11. Collections of Conditions, Nouns, and Verbs
12. Ability to Increase Permissions by Increments of One
13. Permission to Permit
14. Permission to Acquire Permissions
15. Security Features
16. Permissions Exercisable Offline [DR: Is not this a system requirement,
not language?]
17. Revocation
18. Revoke
19. The Unauthenticated User
20. Reference to Frequently Used Expression Fragments
21. The Multi-authenticated User
22. Multiple Conditions
23. Multiple Permissions
24. Ready to Be Profiled
25. Permissions Conditional Upon Permissions
26. Unconditional Permissions
 

-----Original Message-----
From: Patrick Durusau [mailto:pdurusau@emory.edu]
Sent: Friday, September 20, 2002 5:16 AM
To: Radbel, Dmitry
Cc: 'Bob Glushko'; rights@lists.oasis-open.org;
rights-requirements-help@lists.oasis-open.org; karl.best@oasis-open.org;
mnemonic@well.com; murray Maloney; liora Alschuler; rachna Dhamija
Subject: Re: [rights] tyranny of the majority but no consensus


Dmitry,

Understand the desire to not base your statements on a couple of 
meetings but your following of the "paper trail" appears to have missed 
a couple of items.

Radbel, Dmitry wrote:

<snip>

>
>I did not want to base my judgement on a couple of meetings. But we do have
a "paper trail", so I went to to RLTC website and reviewed the documents,
the minutes of the meetings (from 5/21, 5/29, 6/12, 6/26, 7/10, 7/24, 8/7,
and 8/21) and the e-mail correspondence. The minutes clearly show that the
schedule that Hari is obligated to uphold - including timelines for draft
requirements gathering and disposition - has been set quite a while ago and
via a democratic process. I found the schedule in Hari's e-mail from 7/9
(there might have been earlier versions) and per minutes of subsequent
meetings it has been reviewed regularly. The "RLTC Requirements" document
indicates multiple iterations - I went to the e-mail threads and Hari posted
all these iterations asking for comments. I have also found multiple threads
which show that when people had specific and actionable comments Hari have
been addressing them. And the requirements did not have to be dispositioned
by 8/7 - Hari's documents clearly show that he co
>
If you review Hari's posting and attached document at:

http://lists.oasis-open.org/archives/rights-requirements/200206/msg00009.htm
l
http://lists.oasis-open.org/archives/rights-requirements/200206/doc00001.doc

You will note that the original date for submitting requirements (recall 
this is a draft posted on June 24th, is July 10th! A rather remarkable 
time frame for developing requirements for a digital rights language. 
The requirements so submitted were to be composed into a requirements 
document by August 7th!

Apparently (I say apparently because the SBL was not at that time a 
participant in the TC) the absurdity of the proposed schedule was 
pointed out by someone, which resulted in the changing of the date for 
submission to August 7, 2002 and the production of a requirements 
document to September 4, 2002.

For this change see:

http://lists.oasis-open.org/archives/rights-requirements/200206/msg00014.htm
l
http://lists.oasis-open.org/archives/rights-requirements/200206/doc00002.doc

I can't cite the minutes of the most recent meetings, which have not 
been posted to the website but it has been reported to me that on 
September 18, 2002 the Rights TC voted to adopt the requirements that 
had been filed and "dispositioned" as of August 7, 2002. You don't find 
it odd that a group would call for submission of requirements only to 
exclude those requirements from consideration? Why make the call in the 
first place if you are not really interested in other requirements? 
Alas, the record does not contain sufficient information to know if it 
was simply ineptitude or some other motivation that is responsible for 
this "yes, no, maybe, not interested" response to additional requirements.

>
> 
>I don't understand the argument that Hari's work is somehow invalid because
requirements were being mapped to the "initial set". Is the set wrong? Is
the mapping wrong? Seems to me that the point here is that not all of the
originally submitted requirements are presented verbatim - but how would one
produce a meaningful requirements document without combining 100's of
overlapping requirements into a concise set? And why is it Hari's fault that
the requirement for a royalty free language implementation has been voted
down? How can RLTC promise anyone royalty free language implementation when
the whole area is full of patents?
> 
>
The mapping to the original set meant that legitimate requirements, such 
as recognizing grants of rights that were not contractual in nature were 
being systematically excluded from the requirements document.

But in point of fact, we are not talking about a whole area that is full 
of patents. What is under discussion is an OASIS TC adopting in full a 
set of requirements and a digital rights language that were complete or 
nearly so, from a particular vendor. That certainly does make the notion 
of a royalty free implementation problematic.

To create a truly general digital rights expression language, the RLTC 
could solicit (with a meaningful period for comments and drafting) 
requirements and then construct its own digital rights language. If 
anyone wanted to assert IP claims in the area they could do so but since 
the language would be solely the product of an OASIS work process, it is 
doubtful that anyone could assert a claim on the language itself. That 
is not in the interest, shall we say, of some parties and it is 
certainly not the present course of the RLTC.


>
>I don't understand why RLTC can't have Version 1 of the requirements and
then have further iterations - most of the requirements documents I have
ever dealt with were iterated. And it does not mean that requirements added
in the next version are some kind of 2nd class citizens. I am reviewing the
requirements and have some comments - but it does not mean that the TC has
to delay its schedule until they are resolved to my satisfaction. I also
don't understand why trying to synchronize with an international standards
body (the goal which has been clearly stated in the minutes of the very
first meeting on 5/21) is seen as some kind of sinister plot. Is not having
common standards a good idea? What is the concern? We are not defining
rights themselves but a language for expressing them.
>
 General standards that are based on general requirements are a good 
idea. Standards that purport to be general but in fact are only specific 
to a particular industry, are a very poor idea.

The time has come to put to bed the recurrent notion of a "Version 1" of 
the requirements. No meaningful set of requirements has been developed 
by the RLTC. All we have at present is the submission from a single 
vendor of what it thinks the requirements should be for a digital rights 
language. That is not the same thing as having requirements that are 
developed from requirement submissions and debate over those 
submissions. The current "Version 1" of the "requirements" is not a 
requirements document at all but one vendor's view of the requirements.

The notion that we are defining only a "language" for expressing rights 
and not the "rights themselves" is a sort of intellectual sleight of 
hand that does not bear  up under examination. The ability to define a 
language includes the ability to define what rights can be expressed. 
Looking at the latest version of XrML, you will note that it is an 
entirely contract based rights language. There is no provision for 
expressing rights that are not contractually set between the parties. 
Despite its claims of neutrality, XrML embodies a pinched view of the 
world of rights and excludes from expression a whole range of commonly 
acknowledged rights.

>
>After reviewing the available documents, I personally think that Hari did
an admirable job of compiling and processing requirements under difficult
conditions. Bob, you are appalled by the votes (would you still be appalled
if they did not go against you?) and you and others blame much of it on Hari
for supposedly being partial, improperly biasing the work process, etc. The
essense of it seems to be that he has an opinion which is different from
yours and that some decisions that he has made are disagreeable with you.
Frankly, I find such a personal attack - as you put it - appalling.
> 
>
If you took Bob's post as a personal attack I suggest that you read it 
and the record of this TC again. Compare the requirements gathering 
process and the recent choice of a closing date for requirements to any 
similar work of the W3C, ISO or OASIS. What is a stake here is the 
process by which a standard is created.

Patrick


>
>Respectfully,
> 
>Dmitry Radbel, VP Advanced Technology 
>Universal Music Group
>2220 Colorado Ave.,  Santa Monica, CA 90404 
>Office 310-865-7801   
>e-mail: dmitry.radbel@umusic.com 
>
>-----Original Message-----
>From: Bob Glushko [mailto:glushko@SIMS.Berkeley.EDU]
>Sent: Thursday, September 19, 2002 10:43 AM
>To: rights@lists.oasis-open.org;
rights-requirements-help@lists.oasis-open.org; karl.best@oasis-open.org
>Cc: mnemonic@well.com; murray Maloney; liora Alschuler; rachna Dhamija
>Subject: [rights] tyranny of the majority but no consensus
>
>
>I am going to respond to the note from Mike Godwin below sent to me and
Hari Reddy since I doubt that Hari will.
>
>I too am appalled by the recent votes at the F2F and on yesterday's
conference call.  But neither surprised me.
>
>In the days before the F2F, I sent Hari several messages urging restraint
in the face of the mounting pressure from the MPEG constituency in the TC to
ignore the fact that the requirements process hadn't reached consensus but
he didn't reply to me. He didn't reply to me afterwards either when I
suggested he take steps to address the bad feelings caused by the votes at
the F2F.  He has steadily lost the impartiality with which he began as the
TC chairman and now predictably sides with the "party line" as defined by
Content Guard and Microsoft, ignoring the fact that a substantial proportion
of the members are opposed to the "damn the requirements process, full speed
ahead" approach they advocate. 
>
>The critical votes were 11 to 10 and 12 to 9 on Thursday and 10 to 8
yesterday.  This bare majority clearly demonstrates there is no consensus
for moving forward at this pace -- and also demonstrates that there is
little chance that a specification will be voted out of the TC for
submission to OASIS, since far more than 1/4 of the membership will oppose
it.  I am puzzled by the persistence of the MPEG side given this arithmetic.
No specification will be voted out of the TC without dealing with the full
set of requirements submitted to it.  Do the math.
>
>I understand that some of the member companies in this TC have strong
business interests to "get something out" but I also believed in its
charter.  The first goal is said to be:
>
>Define the industry standard for a rights language that supports a wide
variety of business models and has an architecture that provides the
flexibility to address the needs of the diverse communities that have
recognized the need for a rights language
>
>It is clear now that any community other than the MPEG is a second-class
citizen whose requirements will be dealt with at some unspecified future
time. It has been disingenuous to call for participation by user
organizations and by people who care about legal and regulatory  issues and
then vote to suppress any meaningful impact of their contributions.
>
>bob glushko
>
>
>
>
>
>
>X-Sender: mnemonic@brillig.panix.com
>Date: Wed, 18 Sep 2002 22:34:46 -0400
>To: "Reddy, Hari" <Hari.Reddy@CONTENTGUARD.COM>,
>   "Mike Godwin (E-mail)" <mnemonic@well.com>
>From: Mike Godwin <mnemonic@well.com>
>
>Gentlemen,
>
>I am astonished to hear that the wishes of experts contributing to the
subcommittee were wholly ignored in the vote this afternoon.
>
>It seems clear to me that certain corporate members attended the meeting
with the intention of circumventing the wishes of those who want to see the
first version of the REL accurately express a full range
intellectual-property rights.
>
>I hereby register my protest. I think this was immensely insensitive on the
part of the corporate members, and am considering whether and how to
publicize this subversion of a purportedly "open" standards process.
>
>This was the last thing I expected, given the representations that had been
made to me about the subcommittee's work. In the time we had allotted this
morning, I believe I demonstrated my willingness to help the committee reach
a first edition of its work in a reasonable amount of time, and I point out
that drawing a line with regard to submissions was my idea.
>
>I feel an immense sense of betrayal, and I imagine that other members do
too.
>
>
>--Mike
>
>
>

-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
pdurusau@emory.edu