RE: Web Services Architecture going forward (was Re: REST wrap-up )

Hao,

Yes, that is very much the approach that I have been seeing (and following).
In some ways it is nice - decentralized contract specifications and a
decentralized architecture.   All happening "in the wild."

An open spec approach sounds interesting.  Any ideas how to bootstrap this
further?

-bryan

-----Original Message-----
From: He, Hao [mailto:Hao.He@thomson.com.au] 
Sent: Wednesday, February 11, 2004 5:18 PM
To: 'Thompson, Bryan B.'; He, Hao; 'Michael Champion'; '''www-ws-arch@w3.org
' ' '
Subject: RE: Web Services Architecture going forward (was Re: REST wrap-up )


Bryan,

I agree.

It looks like that it is going to take too much efforts to do the REST work
"formally".  What I have in mind is to have a "open spec" approach that
follows the "open source" model. 

Hao

-----Original Message-----
From: Thompson, Bryan B. [mailto:BRYAN.B.THOMPSON@saic.com]
Sent: Thursday, 12 February 2004 8:35
To: He, Hao; 'Michael Champion'; '''www-ws-arch@w3.org ' ' '
Subject: RE: Web Services Architecture going forward (was Re: REST wrap-up )



Hao,

I would definately like to see something emerge here.  There is a lot of
activity in this area, including on many of the things that Mike mentioned
below.  I have been pushing for such an activity within the context of W3C -
and who knows, maybe that will happen now that WSA is closed.

I know of many discussion lists, and many different projects, but nothing
that attempts to setup a framework for coming to agreement on some of the
key issues for interoperability among next generation rest services, some of
which are not even based on HTTP.

In some ways, this is a movement without a center.  Historically, the IETF
and W3C were the centers for these concerns.  But increasingly it looks like
this work is happening out there "in the wild."

-bryan

-----Original Message-----
From: www-ws-arch-request@w3.org [mailto:www-ws-arch-request@w3.org] On
Behalf Of He, Hao
Sent: Tuesday, February 10, 2004 4:19 PM
To: 'Michael Champion'; '''www-ws-arch@w3.org ' ' '
Subject: RE: Web Services Architecture going forward (was Re: REST wrap-up )


hi, Mike,

I thought "minimum interface constraint," was purely technical. 

I think your idea of producing "specs/explanations/examples at the level of
detail and quality of the WS-* specs" is great. Would you like to join?

Hao



-----Original Message-----
From: Michael Champion [mailto:mc@xegesis.org]
Sent: Wednesday, 11 February 2004 0:06
To: '''www-ws-arch@w3.org ' ' '
Subject: Web Services Architecture going forward (was Re: REST wrap-up )




On Feb 10, 2004, at 12:42 AM, He, Hao wrote:

> I guess it does not hurt to have a little technical chat after the 
> group is finished.
>

No one objects to a little technical discussion.  It's the religious 
discussion that some find annoying :-)

I have no say as to what goes on on this list, but a few friendly 
suggestions:

- There are about 100 pages of the WSA Note about to be published that 
don't talk about REST.  How about some critique of that?

- The relationship between the Web architecture (TAG's domain) and the 
web services world is still unclear.  What does that TAG actually say 
(as opposed what some would expect a group with Roy FIelding on it to 
say) about things relevant to services and SOA?  Does anything in the 
Webarch WD conflict with the WSA Note?  Obviously WSA won't be around 
to fix their side, but potentially the TAG should be asked to make 
changes that imply a direction for the Web that doesn't support 
services appropriately.

- If you want to make people angry by discussing politics and religion, 
how about some topic besides the "minimum interface constraint," like 
maybe what to do about the SOAP-based reliable messaging specs?  There 
seems to be no obvious way to break through the political deadlock, is 
there?  Or is something happening behind the scenes?

- Speaking of politics behind the scenes, the "real" web services 
architecture work seems to be done by the folks defining the WS-* 
specs, right?  Maybe someone can tell us how that works, what they 
listen to, how to influence it, etc.  Maybe someone can explain why 
this doesn't happen in public (e.g., you don't want to be distracted by 
RESTifarian ranting???).

- Finally, RESTifarians might want to produce 
specs/explanations/examples at the level of detail and quality of the 
WS-* specs if they want to be taken seriously be the industry.  (E.g., 
how do do transactions, authentication, encryption negotiation, 
reliable messaging across multiple hops, etc. without SOAP).  How about 
you folks discussing how to make that happen, or pointing others to 
material that already exists?

Received on Thursday, 12 February 2004 06:56:03 UTC