W3C home > Mailing lists > Public > public-ws-resource-access@w3.org > December 2009

RE: issue 8200: new proposal

From: Ram Jeyaraman <Ram.Jeyaraman@microsoft.com>
Date: Mon, 14 Dec 2009 19:33:47 +0000
To: Doug Davis <dug@us.ibm.com>
CC: "public-ws-resource-access@w3.org" <public-ws-resource-access@w3.org>
Message-ID: <503546C5699C1144BDEA0D0DFFE7F881181AB5A6@TK5EX14MBXC119.redmond.corp.microsoft.com>
  Sounds ok to me. So the complete proposal for 8200 would then be:
- remove definition/mention of ws-mex-all dialect URI from spec (2 tables in section 5, 1 paragraph in section 6)
- modify the definition of "[Body]/mex:GetMetadata/mex:Dialect " so instead of including:
    When this element is not present, the endpoint SHOULD return all the types of metadata that it deems necessary to communicate with it.
it says:
    When this element is not present, the endpoint MUST return all available metadata.

This sounds fine to me.


  I just noticed that the spec includes this definition of the ws-mex URI:
[Body]/mex:GetMetadata/mex:Dialect@URI='http://www.w3.org/2009/09/ws-mex/Dialects/ws-mex'
    This value returns no metadata sections.

  which isn't true.  It does return a metadata section, but its for a mex metadata doc.  I propose we remove this definition since this IRI is no different than the WSDL IRI - there's no reason to call it out for special treatment.

This is related to issue 8297. I am researching this part to find out why the current statement (noted above) was introduced.

Thanks.

From: public-ws-resource-access-request@w3.org [mailto:public-ws-resource-access-request@w3.org] On Behalf Of Doug Davis
Sent: Monday, December 14, 2009 7:34 AM
To: Ram Jeyaraman
Cc: public-ws-resource-access@w3.org
Subject: RE: issue 8200: new proposal


Sounds ok to me. So the complete proposal for 8200 would then be:
- remove definition/mention of ws-mex-all dialect URI from spec (2 tables in section 5, 1 paragraph in section 6)
- modify the definition of "[Body]/mex:GetMetadata/mex:Dialect " so instead of including:
    When this element is not present, the endpoint SHOULD return all the types of metadata that it deems necessary to communicate with it.
it says:
    When this element is not present, the endpoint MUST return all available metadata.

I just noticed that the spec includes this definition of the ws-mex URI:
[Body]/mex:GetMetadata/mex:Dialect@URI='http://www.w3.org/2009/09/ws-mex/Dialects/ws-mex'
    This value returns no metadata sections.

which isn't true.  It does return a metadata section, but its for a mex metadata doc.  I propose we remove this definition since this IRI is no different than the WSDL IRI - there's no reason to call it out for special treatment.

thanks
-Doug
______________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com<mailto:dug@us.ibm.com>
The more I'm around some people, the more I like my dog.

Ram Jeyaraman <Ram.Jeyaraman@microsoft.com<mailto:Ram.Jeyaraman@microsoft.com>>
Sent by: public-ws-resource-access-request@w3.org<mailto:public-ws-resource-access-request@w3.org>

12/11/2009 08:10 PM

To

Doug Davis/Raleigh/IBM@IBMUS

cc

"public-ws-resource-access@w3.org<mailto:public-ws-resource-access@w3.org>" <public-ws-resource-access@w3.org<mailto:public-ws-resource-access@w3.org>>

Subject

RE: issue 8200: new proposal







>  I'd pefer something like:
       When this element is not present, the endpoint MUST return all available metadata accessable to this particular sender.

The part "When this element is not present, the endpoint MUST return all available metadata" seems sufficient, since in practice, the service will only return metadata that the user is privileged to know. Thanks.

From: Doug Davis [mailto:dug@us.ibm.com]<mailto:[mailto:dug@us.ibm.com]>
Sent: Friday, December 11, 2009 10:16 AM
To: Ram Jeyaraman
Cc: public-ws-resource-access@w3.org<mailto:public-ws-resource-access@w3.org>
Subject: RE: issue 8200: new proposal


the problem with this wording is that it still allows a service to return zero metadata sections and still be spec legal - but not very interoperable.
I'd pefer something like:
       When this element is not present, the endpoint MUST return all available metadata accessable to this particular sender.

thanks
-Doug
______________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com<mailto:dug@us.ibm.com>
The more I'm around some people, the more I like my dog.
Ram Jeyaraman <Ram.Jeyaraman@microsoft.com<mailto:Ram.Jeyaraman@microsoft.com>>

12/10/2009 05:44 PM


To

Doug Davis/Raleigh/IBM@IBMUS

cc

"public-ws-resource-access@w3.org<mailto:public-ws-resource-access@w3.org>" <public-ws-resource-access@w3.org<mailto:public-ws-resource-access@w3.org>>

Subject

RE: issue 8200: new proposal











How about if the current definition of 'absence' implies the following [2]?

Essentially, I have grafted the ws-mex-all definition into the definition of the 'absence' behavior.

Thanks.

[2]

[Body]/mex:GetMetadata/mex:Dialect
When this repeating OPTIONAL element is present, the response MUST include only Metadata Sections corresponding to the Dialect specified by the IRI attribute; if the receiver does not have any Metadata Sections of the indicated Dialect, the response MUST include zero Metadata Sections.
When this element is not present, the endpoint SHOULD return all known metadata. The endpoint has the option to optimize the format in which it returns the data.
From: Doug Davis [mailto:dug@us.ibm.com]<mailto:[mailto:dug@us.ibm.com]>
Sent: Wednesday, December 09, 2009 3:56 PM
To: Ram Jeyaraman
Cc: public-ws-resource-access@w3.org<mailto:public-ws-resource-access@w3.org>
Subject: RE: issue 8200: new proposal


I can go for deleting it if absence == all.  What's there now isn't "all" - its "all that the service thinks is important" - there's a difference.  IMO, the current defn is useless.

thanks
-Doug
______________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com<mailto:dug@us.ibm.com>
The more I'm around some people, the more I like my dog.
Ram Jeyaraman <Ram.Jeyaraman@microsoft.com<mailto:Ram.Jeyaraman@microsoft.com>>

12/09/2009 06:21 PM




To

Doug Davis/Raleigh/IBM@IBMUS, "public-ws-resource-access@w3.org<mailto:public-ws-resource-access@w3.org>" <public-ws-resource-access@w3.org<mailto:public-ws-resource-access@w3.org>>

cc

Subject

RE: issue 8200: new proposal














We also need to mention that when the "ws-mex-all" Dialect URI is used in a GetMetadata request, the Identifier and Content attributes MUST be absent.

Thinking a bit more about this, why not just entirely delete the "ws-mex-all" Dialect URI, since the absence of a Dialect in a request already implies retrieving all [1] metadata?

Thanks.

[1]

[Body]/mex:GetMetadata/mex:Dialect
When this repeating OPTIONAL element is present, the response MUST include only Metadata Sections corresponding to the Dialect specified by the IRI attribute; if the receiver does not have any Metadata Sections of the indicated Dialect, the response MUST include zero Metadata Sections.
When this element is not present, the endpoint SHOULD return all the types of metadata that it deems necessary to communicate with it.
From: public-ws-resource-access-request@w3.org<mailto:public-ws-resource-access-request@w3.org> [mailto:public-ws-resource-access-request@w3.org]<mailto:[mailto:public-ws-resource-access-request@w3.org]> On Behalf Of Doug Davis
Sent: Tuesday, December 08, 2009 9:33 PM
To: public-ws-resource-access@w3.org<mailto:public-ws-resource-access@w3.org>
Subject: issue 8200: new proposal


Per my AI, here's an amended proposal for 8200:
- remove the "...-mex-all" Dialect URI from the Identifier table and the Metadata Format table
- for the definition of the ..-mex-all dialect, replace:
This value returns all known metadata. The endpoint has the option to optimize the format in which it returns the data.
with:
This special dialect IRI indicates that all known metadata dialects are to be returned.  This Dialect IRI is meant to only be used in GetMetadata request messages, not as a Dialect IRI in a mex:MetadataSection element, as there is no actual metadata format associated with it.  The endpoint has the option to optimize the format in which it returns the data.

thanks
-Doug
______________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com<mailto:dug@us.ibm.com>
The more I'm around some people, the more I like my dog.
Received on Monday, 14 December 2009 19:34:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 18 December 2010 18:18:16 GMT