W3C home > Mailing lists > Public > www-ws-arch@w3.org > April 2002

RE: Summary: D-AG0011

From: <michael.mahan@nokia.com>
Date: Tue, 2 Apr 2002 11:20:54 -0500
Message-ID: <5C76D29CD0FA3143896D08BB1743296A072C4E@bsebe001.NOE.Nokia.com>
To: <www-ws-arch@w3.org>
Summary of the Summary

I didn't realize that we have a race condition between wsawg and
the TAG, otherwise I might not have volunteered to be this goal's
champion ;-) 

I tried [1] to separate compliance with the existing web into:
1) accepted Web design principals
2) accepted Web technologies

and formulate these into 2 CSFs. These seem to me to be sufficent 
domains to express an architecture.

David seems to have ageed, with some modifications, with the first
CSF (CF1), but not with its subgoals due to overlap and potential
skew with TAG results [2] and also due to the time it would take 
for us all to agree on the set of architectual principals. 

The second CSF (CF2) has not been commented on. Note that I believe
the subgoals I raised for CF2 suffer from the same critisism as CF1.
I would like to see if the list has issues with CF2 by itself (that is,
without its subgoals). For context, here again are the 2 CSFs:

CF1) The Web Services reference architecture complies with the 
architectural principals and design goals of the existing web.

CF2) The Web Services reference architecture recommends the use
of existing web technologies which adhere to the above principals 
and which provide clear functional coverage of the responsibilities 
and constraints for a component identified in the reference
architecture.


As I understand the thread, these are the current expressed options 
on what to do with the subgoals:

Opt1. Keep the subgoals, possibly with some tinkering. Substitute 
      these with the TAG's work products whenever then appear - 
      hopefully by summer.
Opt2. Replace the subgoals with one subgoal that in essence says "be 
      compliant with the TAG's work products when they are available"
Opt3. Replace the subgoals with the subgoal described by Opt2 plus 
      another subgoal which in essence says "influence the TAG by 
      making them aware of WS perspective / use cases."

In the defense of the first option: in producing the subgoals, I tried 
not to assert anything radical, or even controversial - but what is 
easily mined from W3C design principal and REST documents. I was hoping 
the asserted subgoals could be modified to a LCD consensus 
principals/technologies without too much dissention. 

We can hash this out at the F2F. Personally, I am flexible but still
prefer Opt1. However, if a fierce debate does transpire, then I would 
quickly punt and move for Opt2/3.


Mike Mahan, Nokia


[1] http://lists.w3.org/Archives/Public/www-ws-arch/2002Mar/0471.html
[2] http://lists.w3.org/Archives/Public/www-ws-arch/2002Mar/0475.html


>-----Original Message-----
>From: ext David Orchard [mailto:dorchard@bea.com]
>Sent: April 01, 2002 03:54 PM
>To: 'Champion, Mike'; www-ws-arch@w3.org
>Subject: RE: Summary: D-AG0011
>
>
>Mike,
>
>Agreed we need to move asap and can't wait.  But we've got 
>this lovely goal
>that says the wsarch has to work with the web architecture.  
>My argument is
>that we shouldn't try to define the web architecture in order for us to
>figure out how we fit into it.
>
>Heck, I'd almost be glad with a CSF that says "We are successful at
>complying with the web architecture until the TAG directly 
>says the wsarch
>isn't compliant" ;-)  Let the compliance *stuff* flow uphill 
>so to speak.
>
>I'm trying to look for a way for us to do less work, rather 
>than more.  I
>feel strangely compelled to very closely examine each of the 
>CSFs currently
>listed as I'm afeared of how they could be used, and I *KNOW* that will
>spark heaps of debate.  And I don't think I can live with the 
>interpretation
>as it stands because of the potential interpretations :-(
>
>Dave
>
>> -----Original Message-----
>> From: www-ws-arch-request@w3.org 
>[mailto:www-ws-arch-request@w3.org]On
>> Behalf Of Champion, Mike
>> Sent: Monday, April 01, 2002 12:26 PM
>> To: www-ws-arch@w3.org
>> Subject: RE: Summary: D-AG0011
>>
>>
>>
>>
>> > -----Original Message-----
>> > From: David Orchard [mailto:dorchard@bea.com]
>> > Sent: Monday, April 01, 2002 2:39 PM
>> > To: 'Mark Baker'
>> > Cc: michael.mahan@nokia.com; www-ws-arch@w3.org
>> > Subject: RE: Summary: D-AG0011
>> >
>> >
>> > There's a lot of
>> > "gospel reading" and intepreting that we should avoid.
>>
>> A-men, brother!
>> >
>> >
>> > If the TAG has excluded something from the web architectural
>> > principles, then I really don't think the wsawg ought to
>> define that
>> > *thing* has being part of the web architecture.  That's 
>not our job.
>>
>> I see our job as to contribute a web services perspective to the TAG
>> discussions to ensure that the official "web architecture"
>> covers our use
>> cases.
>>
>> My attitude toward this issue is similar to my attitude
>> toward the SW issue:
>> in the long run, we all hope that there is one architecture
>> for the "web",
>> the "services web" and the "semantic web."  Unfortunately (or
>> fortunately,
>> since it is fun to live in "interesting times") we don't have
>> the luxury of
>> figuring this all out in the lab before it gets released to
>> the world.  The
>> world is out there frantically experimenting with all this
>> stuff, and our
>> job is to help sort it out.  We can try to figure out and 
>convince one
>> another what the One True Web/SW/Services Architecture is
>> (and I personally
>> tend to agree with most of what Mark Baker says on this
>> subject), but we
>> have to do SOMETHING long before that has been specified
>> (because that's
>> what the member companies setup  this  WG to do).
>>
>>
>
>
Received on Tuesday, 2 April 2002 11:21:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 July 2007 12:24:57 GMT