Re: [regrep] [Fwd: [regrep-semantic] IBM BI-ICS]

[Anne Thomas Manes <anne@manes.net>]

I seem to have set off a firestorm...

I don't think I ever intended to imply that the *process* by which the 
IBM/MSFT/BEA cartel has defined their myriad specifications is especially 
open. It is slightly more open than other efforts that claim to produce 
standards (e.g., Liberty Alliance, RFID, numerous industry groups, etc). 
The cartel provides open invitations to anyone that wants to participate in 
workshops, and I can join the discussion list simply by signing a legal 
document. I'm not required to pay any fees. But I'm also not permitted to 
join as an author. The cartel is very restrictive about who gets a piece of 
the IPR.

I also want to note that the copyright in the WS-A spec grants IPR to 
implement the WS-A spec -- but it does not grant the right to create 
derivative works. So implementors need to be very cautious to implement it 
as a pure WS-A library.

I recall when they reissued the spec with the new copyright statement. I 
didn't realize that they didn't change the date on the spec, though. That 
does worry me a bit.

Even so, the industry has adopted quite a few specifications that never 
went through a formal standardization process. Just look at SOAP 1.1 and 
WSDL 1.1. These two specs are the foundation for a pretty significant 
market segment. IBM and Microsoft retain the IPR to these specs. Yet we all 
have built products and applications based on these specs, and RegRep 
certainly references them. The copyright statements in these specs are much 
less open than the one in WS-A.

W3C is building standards based on these specs, but that doesn't change the 
IPR issues surrounding use of SOAP 1.1 and WSDL 1.1. SOAP 1.2 is now a W3C 
standard, but I don't know of anyone that's basing products or applications 
on this spec yet. WSDL 2.0 is definitely not ready for prime time. I doubt 
that we'll see people start to adopt the next generation of the Web 
services framework for at least a couple more years.

Regarding DIME, WS-Attachments, WS-Routing, WS-Referral, etc -- these were 
Microsoft solo efforts, not joint collaborations from the cartel. These 
specs are now being superceded by joint efforts -- just as SDL and XLANG 
were superceded by WSDL and BPEL. Note that WS-Addressing replaces 
WS-Routing/Referral.

Also note that MTOM (a product of W3C XMLP) will replace both 
WS-Attachments and SOAP with Attachments. MTOM is a much better model for 
attachments because it includes the attachment in the SOAP message infoset 
-- which enables you to protect its contents using WS-Security. Besides -- 
MTOM is a royalty free product of the W3C.

You should be aware that just because a specification is product of a 
standards effort, that doesn't make it public domain. Look at XrML. You 
need to pay ContentGuard to obtain a license to implement it.

Anne

At 09:40 AM 3/19/2004, Chiusano Joseph wrote:
>+1 on the instability points. Examples that I like to point to are DIME
>and WS-Attachments - which are no longer supported by Microsoft and are
>reported as being superceded by MTOM[1], and WS-Routing whose features
>were reported at one time to be incorporated into WS-ReliableMessaging,
>but are now reported to be out of scope[2]. I have a *hunch* that if
>these specifications (meaning DIME and WS-Routing) were in an open
>standards consortium such as OASIS or W3C, their stability would be more
>assured.
>
>Joe
>
>[1] http://www.w3.org/TR/2004/WD-soap12-mtom-20040209/
>[2] http://xml.coverpages.org/ni2004-03-11-a.html
>(search on "Routing)
>
>Farrukh Najmi wrote:
> >
> > Anne Thomas Manes wrote:
> >
> > > For the most part I prefer to reference standards works as opposed to
> > > private works, but I make exception for some things -- especially when
> > > they provide essential functionality AND when they contain an open
> > > copyright notice. Such is the case with WS-Addressing.
> > >
> > > WS-Addressing provides essential functionality -- a standard mechanism
> > > to reference a Web service endpoint. In my opinion, it's one of the
> > > most critical WS specifications published last year. I am not aware of
> > > any competitive effort that is currently defining this type of mechanism.
> > >
> > > And I suggest you read the copyright notice [1]. It the most open spec
> > > notice I've ever seen.
> > >
> > > [1]
> > > 
> http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnglobspec/html/ws-addressing.asp
> > >
> > >
> > > Anne
> >
> > Anne,
> >
> > I do not see how you call it the most open spec notice you've
> > ever seen. Nowadays copyright notices as pretty open (except in the
> > music industry
> > of course ;)) It's when it comes to patent claims and licenses that
> > things get
> > sticky.
> >
> > On surface this one seems ok there are some disturbing aspects
> > there that should be taken into account. First of all, this same spec,
> > same version, same date (13 March 2003) used to carry a very different
> > notice
> > (and I have a PDF copy that I downloaded from one of their sites in mid
> > 2003 that
> > proves it). The notice in that version said and I quote:
> >
> > "EXCEPT FOR THE COPYRIGHT LICENSE GRANTED ABOVE, THE AUTHORS DO NOT
> > GRANT, EITHER
> > EXPRESSLY OR IMPLIEDLY, A LICENSE TO ANY INTELLECTUAL PROPERTY,
> > INCLUDING PATENTS,
> > THEY OWN OR CONTROL."
> >
> > So yes, they changed above notice to a much better one, no doubt,
> > (although I would
> > really like to know what "commercially reasonable terms and conditions"
> > means
> > and why it replaces the much more common "reasonable and
> > non-discriminatory terms",
> > and why you consider it so amazingly open. It seems that this language
> > still
> > allows them to reserve the right to impose discriminatory terms. Do not
> > you agree?
> >
> > What is most disturbing is that the specification was changed in the
> > middle of the night,
> > as it were, with no notice, no versioning and no date change;
> >
> > *This* is precisely one of the main problems with these type of
> > specifications that masquerade as pseudo-standards: their authors can
> > change them at a
> > moment's notice and they don't have to tell anybody. There is no
> > accountability.
> > There is no control.
> >
> > Oh, and let's not forget that WS-Addressing has a normative reference to
> > WS-Policy,
> > which last time I checked, still carried that no-grant notice either.
> > But wait....
> > maybe they changed it since I last looked. Let me check...
> >
> > Indeed it still says:
> >
> > "EXCEPT FOR THE COPYRIGHT LICENSE GRANTED ABOVE, THE AUTHORS DO NOT
> > GRANT, EITHER
> > EXPRESSLY OR IMPLIEDLY, A LICENSE TO ANY INTELLECTUAL PROPERTY,
> > INCLUDING PATENTS,
> > THEY OWN OR CONTROL."
> >
> > So the notice is still there, but who knows what else has changed since
> > last time we
> > looked at it? And who knows what might change tonight?
> >
> > --
> > Regards,
> > Farrukh
> >
> > To unsubscribe from this mailing list (and be removed from the roster 
> of the OASIS TC), go to 
> http://www.oasis-open.org/apps/org/workgroup/regrep/members/leave_workgroup.php.
>
>To unsubscribe from this mailing list (and be removed from the roster of 
>the OASIS TC), go to 
>http://www.oasis-open.org/apps/org/workgroup/regrep/members/leave_workgroup.php.

~~~~~~~~~~~~~~~~~~
Anne Thomas Manes
VP & Research Director
Burton Group