- From: Umit Yalcinalp <umit.yalcinalp@oracle.com>
- Date: Wed, 03 Mar 2004 02:03:25 -0800
- To: Jonathan Marsh <jmarsh@microsoft.com>
- Cc: WS Description List <www-ws-desc@w3.org>
- Message-ID: <4045AD6D.4020101@oracle.com>
Jonathan Marsh wrote: >Thanks for the work on this Umit. One reason I didn't complete the >action is that I had some doubts about your correction that required >more research. > >My initial recollection was that we were making it possible to reuse ><wsdl:service> in the body of a message as one way to indicate a service >reference. I've been checking some of my personal notes on the topic >and they concur with the minutes as they were posted. > > The logs of the CVS illustrates that this is a typo as I have illustrated and what we were discussing is not the wsdl:service but wsdl:Service. >As I recall, the schema prevented <wsdl:service> from being reused >because it wasn't a global element. > This issue has nothing to do with reusing service as a service reference. This was a separate decision of the f2f. There were two separate resolutions in the meeting. >We made it global to remedy this, >made the rest of the top-level elements global for consistency, and >opened an issue on whether @name should be optional. > The remedy is not about the service references. >This issue >wouldn't make much sense if we were depending on the wsdl:Service type. > It is, because the type itself made the name attribute required. This is already corrected tobe optional in the schema and it allows the wsdl:Service to be reused for extension. It is impossible to reuse the element technically anyway if one wants to extend the definition of the type and fix the interface/bindings by using schema. The whole point of the exercise was to illustrate what we can do with schema without using additional mechanisms in WSDL. >Roberto's proposal was not adopted, but provides some good scenarios >which might be useful in developing the primer topic illustrating the >reuse of <wsdl:service/> > Roberto's proposal was not adopted, that is correct. Roberto proposed using a new service reference type which we have proven to be unnecessary and decided to use wsdl:Service type. I am actually waiting for him to confirm this as well. However, Roberto's proposal depends on type extension and the primer can only show reuse of wsdl:Service. > >It would be more than a typo to change <wsdl:service> (which is clearly >an element) to "wsdl:Service type" in the resolution. However, it >doesn't seem that the illustration of <wsdl:service> in the primer >prevents one from deriving an element from wsdl:ServiceType if that's >what you wanted to do. > > It is a typo as I have illustrated per the records. The issue is not about preventing its use but what we say in the specification to guide people what they can do. >We've got the topic of this correction on the FTF agenda - I hope we can >treat it as a historical question and not as a design review :-). > > Sanjiva's writeup is correct and I don't think opening it is necessary as it illustrates my recollection of the matter and also follows the historical progress of the service definition in the schema that I illustrated in this thread ;-) --umit > > >>-----Original Message----- >>From: www-ws-desc-request@w3.org [mailto:www-ws-desc-request@w3.org] >> >> >On > > >>Behalf Of Umit Yalcinalp >>Sent: Wednesday, February 25, 2004 6:32 PM >>To: WS Description List >>Subject: Correction for the f2f Minutes (Also action 2004-01-30 for >>Jonathan) >> >> >>We have taken the decision to include a paragraph to the Part 1 >>Specification regarding service references. However, there was a typo >> >> >in > > >>the recording of the decision made in [1]. This email is intended to >>clarify this situation for Jonathan [4]. >> >>We have made two decisions as recorded and numbered here for >> >> >reference: > > >>(1) RESOLVED: Add a paragraph in the spec describing that >> >> ><wsdl:service> > > >>can >> show up on the wire as a service reference. Primer will >> >> >have > > >>some >> examples (see Roberto's counterproposal as a basis). >> >>(2) RESOLVED: Make top WSDL elements global in the schema to >> >> >facilitate > > >>reuse. >> >>The first resolution is in error due to a typo. The minutes should >> >> >have > > >>been recorded as wsdl:Service, instead of wsdl:service. >> >>The reason this typo creates a confusion is due to the second >> >> >resolution > > >>that was taken that day. Prior to the f2f meeting, we did >>not have the WSDL components represented as top level elements in the >>schema. The components were only represented via their types. Per >>resolution 2, components became top level elements and schema has been >>changed to reflect this decision. >> >>Prior to implementing Resolutions 1 and 2, the schema did not contain >>the global "wsdl:service" element, only "wsdl:Service" type. This fact >>alone is enough to illustrate the typo. The schema showed that each >> >> >WSDL > > >>component type named with a complexType declaration using the >> >> >convention > > >>in naming, such as "wsdl:Interface", "wsdl:Service", etc. as shown in >> >> >[2]. > > >>Per resolution 2, the names of the complexTypes have been changed to >>"wsdl:InterfaceType", "wsdl:ServiceType", etc. with the addition of >> >> >top > > >>level elements that are defined using these types, such as >>"wsdl:interface", "wsdl:service" etc. as shown in WSDL Schema [3]. >> >>The decision in the minutes must be changed to represent a service >>reference is by using wsdl:Service type (or a type derived from >>wsdl:Service) to reflect the naming in the schema at the time [2] or >> >> >to > > >>reflect the current name, the name of the type that is currently used >>"wsdl:ServiceType". This will avoid confusion. >> >>The whole point of the exercise was to get Schema derivations to >>statically lock in the interface and/or bindings in the type that is >>illustrated by Roberto's proposal. Further, schema derivation can not >> >> >be > > >>achieved using anelement that the typo seems to suggest. Hence it must >>be corrected. >> >>As a matter of fact, those present at the discussions & presentations >>that were relevant to this topic, would remember that wsdl:Service >> >> >type > > >>was frequently referred to as the "Big-S-Service", and created enough >>giggles in the room while Arthur was giving our presentation during >> >> >the > > >>previous f2f [5]. ;-) >> >>Cheers, >> >>--umit >> >>[1] http://lists.w3.org/Archives/Public/www-ws-desc/2003Nov/0061.html >>[2] >> >> >> >http://dev.w3.org/cvsweb/~checkout~/2002/ws/desc/wsdl20/wsdl20.xsd?rev=1 >.3 > > >>&content-type=text/xml >>[3] >> >> >> >http://dev.w3.org/cvsweb/~checkout~/2002/ws/desc/wsdl20/wsdl20.xsd?rev=1 >.5 > > >>&content-type=text/xml >>[4] http://lists.w3.org/Archives/Public/www-ws-desc/2004Feb/0011.html >>[5] http://lists.w3.org/Archives/Public/www-ws-desc/2003Sep/0188.html >> >> >> >> >>-- >>Umit Yalcinalp >>Consulting Member of Technical Staff >>ORACLE >>Phone: +1 650 607 6154 >>Email: umit.yalcinalp@oracle.com >> >> >> >> > > > > -- Umit Yalcinalp Consulting Member of Technical Staff ORACLE Phone: +1 650 607 6154 Email: umit.yalcinalp@oracle.com
Received on Wednesday, 3 March 2004 05:04:21 UTC