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

 


Help: OASIS Mailing Lists Help | MarkMail Help

wsrm message

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


Subject: Re: [wsrm] [editorial] change of element names



 [[ sorry the earlier mail was accidentally sent before it was complete]]

 Tony,

 First of all the reason I sent that mail was to point out that our earlier reasoning for changing
 names from RMRequest to Request etc. is flawed as the reasoning at that time was the
 'prefix' uses the word 'rm'  anyway.

 And the 2nd reason being we did discuss briefly about the name change yesterday in the
 con. call and as Tom requested me, I suggested the editorial changes.

 While I agree that this is not an issue wrt the tools & processors, I still think there is some
 value in naming them properly as it will be useful for RM protocol developers and
 for runtime deployment using debugging gateways such as TcpMon to grasp the idea
 looking at the Headers.

 The problem now is that the names are too generic and too vague and some what confusing.

 Either we make it more meaningful or make the names totally vague by starting calling
 them A, B,C etc...  If that's the case, I propose that Request Header  e renamed A, PollRequest
 be renamed B etc... When ever we add a new element or attribute, then lets just pick the
 next available letter and if we come to the end ('Z'), lets start with AA etc...  Atleast we will
 save time avoid arguing about the names in this case.  Please don't misunderstand me, I'm
 serious here and far form being sarcastic.

 So i'm fine with either of the above mentioned approaches, but I don't prefer the current
 pseudo confusing names.

 HTH,
 -Sunil





Sunil Kunisetty wrote:

>  Tony,
>
>  First of all the reason I sent that mail was to point out that our earlier reasoning for changing
>  names fromRMRequest to Request etc.is flawed as the reasoning at that time was the
>  'prefix' uses the word 'rm'  anyway.
>
>  And the 2nd reason being we did discuss briefly about the name change yesterday in the
>  con. call and as Tom requested me, I suggested the editorial changes.
>
>  While I agree that this is not an issue wrt the tools & processors, I still think there is some
>  value in naming them properly as
>
>  The problem now is the
>
> Tony Graham wrote:
>
> > Jacques Durand <JDurand@us.fujitsu.com> wrote at Wed, 25 Feb 2004 12:25:20 -0800:
> > > I think Sunil has a point: the qualifier name (prefix "alias") is
> > > not normative, only the namespace is. E.g. not only we can't prevent
> > > some other SOAP headers to be called "Request", but we can't prevent
> > > someone from using the "wsrm" prefix to resolve to another
> > > namespace. That would be quite confusing to readers (if not to
> > > processors). I am in favor of this renaming.
> >
> > I suggest that adding more to the WS-Reliability element names is
> > unnecessary, for the following reasons:
> >
> > 1. If users have to read even a fraction of one percent of Reliable
> >    messages, then there's something wrong somewhere.
> >
> > 2. If a user who doesn't understand how to read namespaced XML reads a
> >    raw SOAP message, there'll be more that the user doesn't understand
> >    than just the WS-Reliability headers.
> >
> > 3. The choice of prefix is not going to confuse the processors
> >    (although broken processors will increase the incidence of users
> >    reading messages, I admit).
> >
> > 4. With SOAP 1.2, at least, if your RMP puts a "wsrm" prefix on the
> >    WS-Reliability elements, then the prefix will remain the same even
> >    through forwarding intermediaries.
> >
> >    http://www.w3.org/TR/2003/REC-soap12-part1-20030624/#soapinterminfoset
> >    says:
> >
> >    22. All namespace information items in the [in-scope namespaces] of
> >        element information items MUST be preserved. Additional
> >        namespace information items MAY be added.
> >
> >    and http://www.w3.org/TR/2001/REC-xml-infoset-20011024/#infoitem.namespace
> >    says a namespace information item has a prefix property.
> >
> >    (SOAP 1.1 doesn't seem to say anything on the subject.)
> >
> > 5. There's nothing stopping anyone from defining a different SOAP
> >    Header with a local name of "RMRequest".
> >
> > Regards,
> >
> > Tony Graham
> > ------------------------------------------------------------------------
> > Web Products, Technologies and Standards           Phone: +353 1 8199708
> > Sun Microsystems                                              x(70)19708
> > East Point Business Park, Dublin 3, Ireland
> >
> > To unsubscribe from this mailing list (and be removed from the roster of the OASIS TC), go to http://www.oasis-open.org/apps/org/workgroup/wsrm/members/leave_workgroup.php.



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