OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

ws-rx message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: RE: [ws-rx] Proposals for MakeConnection issues summary


I have to take exception to the following statement (from below):

CR33 is still not closed. There is still a potential impact on the RM anon URI. Earlier actions taken in relation to CR33 certainly do not support the notion of other specs defining their own anonymous URIs any longer.

My understanding of the current position of WS-Addressing (and I've been on all the calls and tracked the mailing list) is that WS-Addressing will not make any promises about supporting some other specifications special URI semantics but (and here's the important part) WS-Addressing will make every effort not to interfere with other specifications special URI semantics.

 

The whole debate about "is the RManonURI really anonymous" was only relevant because, at that time, the WS-Addr WSDL Binding Spec's "<wsaw:Anonymous>required</wsaw:Anonymous>" implied some sort of monopoly on the contents of wsa:ReplyTo. Now that we (the WS-Addr WG) have come to see the error of our ways and are working on defining a more extensible mechanism for specifying WS-Addr-related constraints, the whole argument about "whether RManonURI really anonymous" is completely moot!

 

Specifications other that WS-Addr are allowed to define their own special URI's and define what semantics apply when you use those URIs as the value of wsa:Address elements. This has always been true. The only thing that has really changed is that the WS-Addr group has realized that the mechanisms that are used to indicate the support/non-support of WS-Addr-defined special URIs (like wsa:anonymous) can't preclude support for other special URIs.

 

- gp



From: Marc Goodner [mailto:mgoodner@microsoft.com]
Sent: Tuesday, November 14, 2006 6:58 PM
To: ws-rx@lists.oasis-open.org
Subject: [ws-rx] Proposals for MakeConnection issues summary

There has been such a flurry of mails on MakeConnection today (apologies) I thought I try to provide a summary from my point of view. Below I’ve listed the issues related to MakeConnection, how each of the options on the straw poll relates (except the withdrawn proposal 1) and any relevant notes. In particular I’m calling out where each side agrees that something further needs to be done.

 

I apologize that there seem to be two new issues in the table below. I view them more as open questions that may or may not have issues behind their answers. I am calling these out as proposal 2 does not have any open questions related to those points. The open questions are only in relation to the current spec.

 

Issue

Close No Action

Proposal 2

Notes

1

Solved

Solved

 

15

Open

Open

Both sides seem to agree that an assertion for MakeConnection is an idea worth exploring. No proposals on the table yet.

27

Open

Open

Both sides seem to doubt there is an issue here. This might be close with no action either way. If it isn’t there may only be clarifying text required.

28

Open

Closed

No proposal if 1 is closed with no action

29

Open

Closed

No proposal if 1 is closed with no action

30

Open

Closed

No proposal if 1 is closed with no action

31

Open

Closed

No proposal if 1 is closed with no action

WSA WG dependency*

Open

Closed

CR33 is still not closed. There is still a potential impact on the RM anon URI. Earlier actions taken in relation to CR33 certainly do not support the notion of other specs defining their own anonymous URIs any longer.

 

Proposal 2 does not have any WSA WG dependencies as it does not define its own special URI requiring support from that WG’s outputs.

Potential security composition issue**

Open

Closed

There has been an open question in the implementation subcommittee for a month on how to relate the proper token to a sequence created through the use of MakeConnection. There may not be an issue with the current spec, but there is the open question.

 

Proposal 2 has advice that a new security token should be issued for any new sequence created through MakeConnection so there isn’t an open question there.

 

References follow.

 

Regards,

Marc g

 

References:

Current straw poll ballot for what to do about MakeConnection:

http://www.oasis-open.org/apps/org/workgroup/ws-rx/ballot.php?id=1147

 

*http://www.w3.org/2002/ws/addr/6/11/06-ws-addr-minutes.html

The WG decided here to issue an errata to strike the text in the WSA SOAP binding suggesting other specs define their own anonymous URIs.

 

** http://www.oasis-open.org/archives/ws-rx-implement/200610/msg00010.html

The open thread on how to relate a security token to a sequence created through the use of MakeConnection.

 

Current MakeConnection issue threads (links are to beginning of thread):

1 – I’m not sure which one to point to for this, there’s a bunch. Nothing as fresh as today.

Description: http://docs.oasis-open.org/ws-rx/issues/pr/Issues.xml#i001

 

15 – no thread yet, the idea of an assertion has been discussed under some of the other issue threads.

Description: http://docs.oasis-open.org/ws-rx/issues/pr/Issues.xml#i015

 

27 - http://www.oasis-open.org/archives/ws-rx/200611/msg00125.html

Description: http://docs.oasis-open.org/ws-rx/issues/pr/Issues.xml#i027

 

28 - http://www.oasis-open.org/archives/ws-rx/200611/msg00120.html

Description: http://docs.oasis-open.org/ws-rx/issues/pr/Issues.xml#i028

 

29 - http://www.oasis-open.org/archives/ws-rx/200611/msg00123.html

Description: http://docs.oasis-open.org/ws-rx/issues/pr/Issues.xml#i029

 

30 - http://www.oasis-open.org/archives/ws-rx/200611/msg00122.html

Description: http://docs.oasis-open.org/ws-rx/issues/pr/Issues.xml#i030

 

31 - http://www.oasis-open.org/archives/ws-rx/200611/msg00121.html

Description: http://docs.oasis-open.org/ws-rx/issues/pr/Issues.xml#i031

 

smime.p7s



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]