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

 


Help: OASIS Mailing Lists Help | MarkMail Help

sca-policy message

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


Subject: Re: [sca-policy] Action 20080112-02: (Ashok) Raise an issue regarding theplaces where interaction intents can be attached


Let me start by apologizing. I need to back track; propagatesTransaction and suspendsTransaction are interaction intents, as per lines 2262-2263 (rev13). So everything I said before is null and void. This raises a question in my mind. I'm thinking that we should change the spec to make these things implementation intents. It would solve the clunky wire matching problem we have with them. I'll open an issue for this.

Now, to your question. I think it would be more appropriate to say that interaction intents don't _apply_ to implementations. A composite can be an implementation, and we do support the ability to attach intents at the composite level of the XML hierarchy such that they are logically attached to all the applicable child elements, which would be bindings in this case. Remember that we make careful use of the words attach and apply, they are not synonyms.

So to answer your specific question; no, I don't think we should say that interaction intents cannot be attached to implementations.

However, if the question you meant to ask was 'Do we need to say that interaction intents cannot apply to implementations?', then I agree with you. Currently the spec says 'typically interaction intents apply to bindings', which is a fairly weak statement. Can anyone think of use cases where an interaction intent would apply to an implementation? If not, then we should open an issue for this.


As for your question 2 from your original email. I think you should open an issue to tighten up the schema.

And we still need an issue to create/update and consolidate all the intent def'ns into one chapter in the document. Are you going to open this one also?

Dave Booz
STSM, BPM and SCA Architecture
Co-Chair OASIS SCA-Policy TC and SCA-J TC
"Distributed objects first, then world hunger"
Poughkeepsie, NY (845)-435-6093 or 8-295-6093
e-mail:booz@us.ibm.com

Inactive hide details for ashok malhotra ---01/15/2009 11:29:15 AM---Do we need to say that interaction intents cannot be attacashok malhotra ---01/15/2009 11:29:15 AM---Do we need to say that interaction intents cannot be attached to implementations?


From:

ashok malhotra <ashok.malhotra@oracle.com>

To:

David Booz/Poughkeepsie/IBM@IBMUS

Cc:

sca-policy@lists.oasis-open.org

Date:

01/15/2009 11:29 AM

Subject:

Re: [sca-policy] Action 20080112-02: (Ashok) Raise an issue regarding the places where interaction intents can be attached





Do we need to say that interaction intents cannot be attached to
implementations?
All the best, Ashok


David Booz wrote:
>
> From rev13, lines 347-350
>
> Implementation elements often apply to <implementation/> elements, but
> they can also apply to <binding/> elements, where the desire is to
> influence the activity of the binding implementation code and how it
> interacts with the remainder of the runtime code for the implementation.
>
>
> BTW, there is a typo at the start of line 347....it should read
> Implementation intents often apply......
>
> And yes this was done so that the transaction intents
> propagatesTransaction and suspendsTransaction could apply to a service
> or reference binding. Note the preceding paragraph (lines 341-345)
> which says that implementation intents on services and references
> don't affect wire matching. This is the reason we added the @intentType.
>
>
> What we need is for each of the intent Definitions that we have in the
> spec to be updated with the intentType attribute. There are some in
> CH8 and some in CH9. This is what I thought you were going to open the
> issue for. It would be good at the same time to add intent def'ns for
> the intents which don't have def'ns and then put all the def'ns into
> their own chapter.
>
> Dave Booz
> STSM, BPM and SCA Architecture
> Co-Chair OASIS SCA-Policy TC and SCA-J TC
> "Distributed objects first, then world hunger"
> Poughkeepsie, NY (845)-435-6093 or 8-295-6093
> e-mail:booz@us.ibm.com
>
> Inactive hide details for ashok malhotra ---01/14/2009 02:57:39
> PM---Is there an example of that in the Transaction section? Iashok
> malhotra ---01/14/2009 02:57:39 PM---Is there an example of that in
> the Transaction section? I cannot remember why we say that.
>
>
> From:
> ashok malhotra <ashok.malhotra@oracle.com>
>
> To:
> David Booz/Poughkeepsie/IBM@IBMUS
>
> Cc:
> sca-policy@lists.oasis-open.org
>
> Date:
> 01/14/2009 02:57 PM
>
> Subject:
> Re: [sca-policy] Action 20080112-02: (Ashok) Raise an issue regarding
> the places where interaction intents can be attached
>
> ------------------------------------------------------------------------
>
>
>
> Is there an example of that in the Transaction section?  I cannot
> remember why we say that.
> All the best, Ashok
>
>
> David Booz wrote:
> >
> > On your point 1, the spec already says that impl intents can be
> > attached to bindings. Do you want to change that?
> >
> >
> > Dave Booz
> > STSM, BPM and SCA Architecture
> > Co-Chair OASIS SCA-Policy TC and SCA-J TC
> > "Distributed objects first, then world hunger"
> > Poughkeepsie, NY (845)-435-6093 or 8-295-6093
> > e-mail:booz@us.ibm.com
> >
> > Inactive hide details for ashok malhotra ---01/14/2009 02:24:24
> > PM---When we resolved issue 44 we added a attribute to the inteashok
> > malhotra ---01/14/2009 02:24:24 PM---When we resolved issue 44 we
> > added a attribute to the intent definition XSD called intentType which
> >
> >
> > From:
> > ashok malhotra <ashok.malhotra@oracle.com>
> >
> > To:
> > OASIS Policy <sca-policy@lists.oasis-open.org>
> >
> > Date:
> > 01/14/2009 02:24 PM
> >
> > Subject:
> > [sca-policy] Action 20080112-02: (Ashok) Raise an issue regarding the
> > places where interaction intents can be attached
> >
> > ------------------------------------------------------------------------
> >
> >
> >
> > When we resolved issue 44 we added a attribute to the intent definition
> > XSD called intentType which can have its value 'interaction' or
> > 'implementation'.   I would like to raise 2 issues wrt to this.
> >
> > 1. We need to say that implementation intents MUST be attached only to
> > implementations and interaction intents MUST be attached to references,
> > services, bindings, etc.
> > A possible complication is that either type of intent may be associated
> > with 'component' and inherited downwards.
> >
> > 2. The schema could be tightened up to say that only the two values
> > 'interaction' or 'implementation' are allowed for this attribute.
> > --
> > All the best, Ashok
> >
> > ---------------------------------------------------------------------
> > To unsubscribe from this mail list, you must leave the OASIS TC that
> > generates this mail.  Follow this link to all your TCs in OASIS at:
> >
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php 
> >
> >
> >
>
>




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