[
Lists Home |
Date Index |
Thread Index
]
You are correct in all particulars as usual, Robin,
OASIS rules don't specify interoperability tests per se. That is, it
doesn't require that three member companies use each others'
implementations/applications with their own
implementations/applications nor that each others'
implementations/applications can use any of the components or
products produced by those other implementations/applications within
their own applications.
OASIS only says that three member companies must be willing to
certify that they are using the specification successfully. And your
characterization of what that means is all too accurate.
That said, if a specification as complex and comprehensive as WSRP is
attested to by at least three member companies, and in this case by
more than double that number, then a certain level of
interoperability can be said to attain to applications which use the
same specification with the same terms, operations, signatures,
response mechanisms, etc, at least to the extent that terms and
datatypes specified will behave the same in different applications
or, in the case of WSRP, portlets built on different platforms with
different tools will look very much the same in the end user's
browser regardless of how they come to be aggregated onto a page nor
by what tools those aggregations are produced, as long as they are
aggregated according to the specification. Now, I can say that
because it also happens to be true that the WSRP TC has conducted
fairly extensive interoperability and conformance testing, quite more
than would be strictly necessary to simply abide by OASIS rules, but
since the objective all along has been to produce a system with
genuine, demonstrable interoperability, the rules have simply been
guidelines which in this case have been surpassed in favor of more
comprehensive testing, which, is ongoing, btw.
One important point needs to be made though. It is the products of
standard-compliant or conformant applications that are interoperable,
not the standard. Standards don't interoperate, they specify the
things that interoperate. A lot of us, me in particular, seem to
forget this regularly when writing about the topic of
interoperability and regularly write about interoperable standards
instead of standards aimed to produce interoperability. Mea Culpa.
I'll try to do better.
While WSRP did not hold public demonstrations of the interoperability
of applications using it, it was demonstrated in a session at XML
2003. It was also demonstrated again in March at the XML.gov Working
Group Meeting.
CAP, for another example, which has not been tested as thoroughly as
WSRP, and is not a web service standard although it can be
implemented through a web service, had public tests in September at
the Global Homeland Security Conference and again March 11, at a
Congressional demonstration, the ComCARE-sponsored Intreroperability
Now! workshop, holding the demonstration in the Rayburn House Office
Building that evening.
I apologize if I am spreading misconceptions, and like I said, I will
try to do better.
Ciao,
Rex
At 12:30 PM -0600 4/3/04, Robin Cover wrote:
>Re:
>
>> You seem to have missed my post that OASIS approval process for a
>> standard requires three member companies to vouch for using the spec
>> successfully, which also happens to mean interoperably, as in it
>> being read, loaded and displaying each company's web services
>> portlets.
>
>"... which also happens to mean interoperably, as in it
> being read, loaded and displaying each company's web services
> portlets"
>
>Hmmm, Rex, could you clarify? I'm imperfectly informed about the
>OASIS rules, and endeavor never to engage in casuistry, which is
>typically done by Karl Best, but I'm not aware that the OASIS
>process has any interoperability requirement. Your source?
>You could intend to describe that the OASIS WSRP TC has done [1]
>but that's not what the quoted paragraph asserts.
>
>At the canonical location [as of 2004-04-03 11:52 CST]
>
>http://www.oasis-open.org/committees/process.php#standard
>
>sub 'Member Approval' we have the requirement:
>
> "Certification by at least three OASIS member organizations
> that they are successfully using the specification consistently
> with the OASIS IPR Policy"
>
>Within the range of my (fallible) memory and (incomplete) tracking,
>the notion of "successfully using" has been interpreted very
>loosely to mean (simply) ""successfully using". Nothing is said
>about the kind of software, whether it's private/internal, or
>any such thing. In one memorable case, one or more members
>made a declaration that it(s company) had "successfully
>implemented" a Committee Specification. Subsequently, a very late
>patent claim was made upon the technology (possibly) necessary to
>"implement" the specification, with RAND terms. So the declaration
>was changed to meet the nominal requirement, as "implementation"
>might have suggested particular code implementation and resulted in
>a demand for a hefty royalty payment. The new declaration simply said,
>as is required:
>
> "successfully using the specification"
>
>Who knows? Maybe they printed the spec and successfully used it
>as a doorstop, violating no patents.
>
>That said (and facetiously in the preceding sentence), many OASIS
>Technical Committees work very hard at achieving and demonstrating
>interoperability. OASIS interop events are routine [1], held at
>major XML conferences, RSA conferences, Burton Catalyst conferences,
>etc. Sometimes these events are held prior to a Committee Specification
>being balloted for approval as an OASIS Standard (to reveal spec
>ambiguities and errors); sometimes these events are held after
>spec approval in order to establish rules for best practice
>implementations.
>
>OASIS standards directly address "interoperability within and between
>marketplaces" but (as far as I know) there are no formal requirements
>for interop as part of the OASIS process.
>
>===============
>
>[1] WSRP Interop
>
>http://xml.coverpages.org/WSRP-InteropDemo2004.html
>
>[2] OASIS interop events
>
>http://lists.oasis-open.org/archives/announce/200312/msg00004.html
>"OASIS Interop Demos Showcase ebXML, SAML, UBL, WS-Reliability, and XACML"
>
>http://xml.coverpages.org/ni2004-02-19-a.html
>"OASIS SAML Interoperability Event Demonstrates Single
>Sign-On at RSA Conference"
>
>http://xml.coverpages.org/ni2003-06-27-b.html
>"OASIS Member Companies Host SPML Identity Management
>Interoperability Event"
>
>http://xml.coverpages.org/ni2002-07-15-a.html
>"Burton Group's Catalyst Conference Features SAML Interoperability Event"
>
>
>-----------------------------------------------------
>Robin Cover
>XML Cover Pages
>WWW: http://xml.coverpages.org
>Newsletter: http://xml.coverpages.org/newsletter.html
>ISOGEN: rcover@innodata-isogen.com
>OASIS: robin.cover@oasis-open.org
>
>--------------------------
>
>
>On Sat, 3 Apr 2004, Rex Brooks wrote:
>
>> You seem to have missed my post that OASIS approval process for a
>> standard requires three member companies to vouch for using the spec
>> successfully, which also happens to mean interoperably, as in it
>> being read, loaded and displaying each company's web services
>> portlets. For WSRP that included: * BEA * Citrix * Fujitsu * IBM *
>> Oracle * Plumtree * Sun * Vignette. For a loooong set of example
>> portlets from Oracle see:
>>
>>
>>http://psprovider.oracle.com/jsp/catalog/search/search_results.jsp?sort_order=ch.COMPANY_NAME&sort_direction=ASC&portlet_keyword=&portlet_category=&provider_type=&portal_version=&supported_language=&company_category=&company_name=Oracle9%3Ci%3Ei%3C%2Fi%3EAS
>>
>> To view Oracle's interoperability test site, inlcuding samples of
>> WSRP/JSR168 compatible portlets visit:
>>
>> http://portalstandards.oracle.com
>>
>> For IBM's WSRP test kit:
>>
>> http://www.alphaworks.ibm.com/tech/wsrptk
>>
>> For the Portlet Open Source Trading Post (for wsrp
>> -conformant/compliant portlets) started by Ross Fubini of Plumtree:
>>
>> http://sourceforge.net/projects/portlet-opensrc/
>>
>> Citrix also has resources for .Net.
> >
>> Obviously, I only speak to WSRP interoperability, though I believe
>> WS-A is fairly well accepted for the run of that WS-I's specs.
>> WS-Eventing could prove worthy, but I hate the names Source and
>> Sink--like voodoo hocuspocus when its just another pub-sub model. As
>> for all the frameworks, I'll reserve judgement, but I like WSBPEL's
>> chances of providing substantial interoperable value for business
>> rules within web services and web apps running inside web services.
>>
>> Ciao,
>> Rex
>>
>> At 4:51 PM +0100 4/3/04, Paul Sumner Downey wrote:
>> >What seems to be lost in all this "flower blooming" is
>> >the notion of interoperability - the only reason to even
>> >look at Web services.
>> >
>> >No one is going to use *any* of these specs, regardless of
>> >how great they are or who publishes them if they don't
>> >interoperate - and i mean in product shipped by a wide
>> >variety of vendors, not just on paper or in some one-off
>> >fest held behind closed doors.
>> >
>> >Paul
>> >
>> >--
>> >Paul Sumner Downey
>> >http://blog.whatfettle.com
>> >
>> >
>> >-----Original Message-----
>> >From: Tim Bray [mailto:tbray@textuality.com]
>> >Sent: 02 April 2004 21:04
>> >To: xml-dev@lists.xml.org DEV
>> >Subject: [xml-dev] WS-Emperor naked?
>> >
>> >
>> >Would anyone here like to argue that the list found in
>> >
>> > http://www.tbray.org/ongoing/When/200x/2004/04/01/WS-Mumble
>> >
>> >is coherent, or sensible, or viable, or generally that the parrot is
>> >not dead?
>> >
>> > -Tim
>> >
>> >
>> >
>> >-----------------------------------------------------------------
>> >The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
>> >initiative of OASIS <http://www.oasis-open.org>
>> >
>> >The list archives are at http://lists.xml.org/archives/xml-dev/
>> >
>> >To subscribe or unsubscribe from this list use the subscription
>> >manager: <http://www.oasis-open.org/mlmanage/index.php>
>>
>>
>> --
>> Rex Brooks
>> GeoAddress: 1361-A Addison, Berkeley, CA, 94702 USA, Earth
>> W3Address: http://www.starbourne.com
>> Email: rexb@starbourne.com
>> Tel: 510-849-2309
>> Fax: By Request
>
>
>-----------------------------------------------------------------
>The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
>initiative of OASIS <http://www.oasis-open.org>
>
>The list archives are at http://lists.xml.org/archives/xml-dev/
>
>To subscribe or unsubscribe from this list use the subscription
>manager: <http://www.oasis-open.org/mlmanage/index.php>
--
Rex Brooks
GeoAddress: 1361-A Addison, Berkeley, CA, 94702 USA, Earth
W3Address: http://www.starbourne.com
Email: rexb@starbourne.com
Tel: 510-849-2309
Fax: By Request
|
