RE: D-AG004: Ensures platform independence

Sorry, I didn't mean to send the mail below to the whole list.
-Suresh

-----Original Message-----
From: Damodaran, Suresh 
Sent: Thursday, March 21, 2002 9:28 AM
To: 'Abbie Barbir'
Cc: www-ws-arch@w3.org
Subject: RE: D-AG004: Ensures platform independence


Hi Abbie,
 
I don't think anybody answered my question here.
May be it was lost in the melee.
Will appreciate your thoughts.
 
Regards,
-Suresh

-----Original Message-----
From: Damodaran, Suresh 
Sent: Tuesday, March 19, 2002 2:16 PM
To: 'Abbie Barbir'; Anne Thomas Manes; Vinoski, Stephen
Cc: www-ws-arch@w3.org
Subject: RE: D-AG004: Ensures platform independence


Abbie, 
I am doing what I can to add to the traffic volume:-)
 
In general I agree with the statement that WS should be language independent
as others have broadly stated earlier. Well, what exactly does this
"independence" mean?
I would like to throw in the issue of object serialization as a test case.
 
When SOAP is used for RPC like calls, parameters are passed to/from using
serialized objects. To be platform independent, WS can either ignore this
issue
completely (saying other solutions exist), or address it by a normalized
serialization format.
Note that there is no restriction that the binding should be an OO language.
Care to comment on what "language independence" means to WS w.r.t. this
issue?
 
Regards,
-Suresh
 

-----Original Message-----
From: Abbie Barbir [mailto:abbieb@nortelnetworks.com]
Sent: Tuesday, March 19, 2002 12:57 PM
To: Anne Thomas Manes; Vinoski, Stephen
Cc: www-ws-arch@w3.org
Subject: RE: D-AG004: Ensures platform independence


Please note that JAVA was used only as an example !!!!!!!
 
It is not the only way.
 
ps: AT least I am getting some traffic now, that is good ??????
 
abbie
 

-----Original Message-----
From: Anne Thomas Manes [mailto:anne@manes.net]
Sent: Tuesday, March 19, 2002 12:29 PM
To: Barbir, Abbie [CAR:1A00:EXCH]; Vinoski, Stephen
Cc: www-ws-arch@w3.org
Subject: RE: D-AG004: Ensures platform independence


I agree with Steve. I think that we need to avoid requiring the details or
object model of Java or any language.
 
Web services are platform independent and language independent simply
because we don't attempt to specify the mappings between XML and language
semantics.
 
All Web service specifications are defined using XML, and all data types are
specified as XML types. It's up to the implementors to map those types to
language types. For the Java community, the JCP is defining standard type
mappings as part of JAX-RPC. Microsoft is defining the type mappings for the
.NET community. I recommend that we continue to let the language communities
define their own language type mappings.
 
Anne

-----Original Message-----
From: www-ws-arch-request@w3.org [mailto:www-ws-arch-request@w3.org]On
Behalf Of Abbie Barbir
Sent: Tuesday, March 19, 2002 11:30 AM
To: Vinoski, Stephen
Cc: www-ws-arch@w3.org
Subject: RE: D-AG004: Ensures platform independence


steve,
 
this why the work like JAVA, as opposed just JAVA.
 
cheers
 

-----Original Message-----
From: Vinoski, Stephen [mailto:VINOSKI@iona.com]
Sent: Tuesday, March 19, 2002 11:28 AM
To: Barbir, Abbie [CAR:1A00:EXCH]
Cc: www-ws-arch@w3.org
Subject: RE: D-AG004: Ensures platform independence


Supporting heterogeneity requires supporting multiple programming and
scripting languages. Stating that we need to "adopt the use of platform
independent languages such as Java" to help ensure that we allow for the use
of multiple programming models is a bit of an oxymoron. For true platform
and language independence, we must avoid requiring the details or object
model of Java or any other programming language to "show through" at the Web
Services level.
 
--steve

-----Original Message-----
From: Abbie Barbir [mailto:abbieb@nortelnetworks.com]  
Sent: Tuesday, March 19, 2002 10:49 AM
To: michael.mahan@nokia.com; Austin.D@ic.grainger.com
Cc: www-ws-arch@w3.org; Christopher Ferris
Subject: RE: D-AG004: Ensures platform independence


Mike and all,
 
Per the last teleconference call D-AG004 is still open and must be closed by
the next call.
In this e-mail I attempt to close the issues regarding the design Goal.
The goal as stated 
" 
D-AG0004 
ensures platform independence of web services components in a way that does
not preclude any programming model nor assume any particular mode of
communication between the individual components 

"

In my original e-mail I have riased some issues regarding the goal, but
unfortunatly I did not get good track.
 
So here are my $.002,
 
Discussion:
------------------
 
In order to ensure platform independence, we need the following:
1. Adopt the use of platform independent languages such as JAVA. 
2. Have very well defined interfaces to web components that are general and
could be invoked using appropriate protocols.
3. Try to use XML and other markup techniques that ensures the ability to
invoke web components using variaty of protocols.
 
 
The design goal can coexist with D-AG0003: is sufficiently extensible and
with D-AG0005:provides simplicity and ease-of-use that does not impose high
barriers to entry for users of web services. In my opnion, the appropriate
designs for platform independence interfaces can help the above two stated
golas.
 
 
Furtheremore, Some issues were raised regarding the interdependecne of
D-AG004 and D-AG011. Here D-AG011 is restated:
 
CURRENT GOAL STATEMENT: 
Web Services are consistent with the existing web and its heterogenous
environment and distributed architecture to the greatest extent possible.
 
Discussion -2:
-------------------
 
I really do think that the requirement of platform independence does
contradict the above design goal. In my view designing for platform
independence must be consistant with D-AG011. So in this regard, D-AG011 can
feed into D-AG004.
 
Crtitical Success Factors:
----------------------------------
 
The design goal can be achieved through the insistance on using platform
independent development tools and languages such as XML and JAVA. The
interfaces to web components must also be properly designed.
 
Derived Requirements:
--------------------------------
1. Consistent definition of web components
2. Good definition of web components Interfaces.
3. Use XML based protocols for invoking the web components.
 
 
Current Recommendation:
--------------------------------------
The goal should stay as written
 
 
cheers
abbie
 
 



-----Original Message-----
From: michael.mahan@nokia.com [mailto:michael.mahan@nokia.com]
Sent: Thursday, March 14, 2002 10:48 AM
To: Austin.D@ic.grainger.com; Barbir, Abbie [CAR:1A00:EXCH]
Cc: www-ws-arch@w3.org
Subject: RE: D-AG004: Ensures platform independence


Hi Abbie,
 
In message [1], I questioned whether platform independence goal is separate
from D-AG0011. Can you comment on that.
 
Thx, Mike
 
 
[1]  <http://lists.w3.org/Archives/Public/www-ws-arch/2002Mar/0181.html>
http://lists.w3.org/Archives/Public/www-ws-arch/2002Mar/0181.html

-----Original Message-----
From: ext Austin, Daniel [mailto:Austin.D@ic.grainger.com]
Sent: March 12, 2002 05:41 PM
To: 'Abbie Barbir'
Cc: 'www-ws-arch@w3.org'
Subject: RE: D-AG004: Ensures platform independence


Hi Abbie,
 
    The text of this goal was quoted directly from the group charter, so we
have to be careful about amending it. But I think these are good questions.
What are the answers?
 
Regards,
 
D-

-----Original Message-----
From: Abbie Barbir [mailto:abbieb@nortelnetworks.com]
Sent: Tuesday, March 12, 2002 1:19 PM
To: www-ws-arch@w3.org
Cc: Abbie Barbir
Subject: D-AG004: Ensures platform independence




This message kicks off discussion on goal 4, "ensures platform independence
of web services" 

Please comment on goal wording, success factors, relation to other goals,
etc. 

The goal as stated 
" 
D-AG0004 
ensures platform independence of web services components in a way that does
not preclude any programming model nor assume any particular mode of
communication between the individual components 

" 

Discussion: 
-------------- 
I think that the goal as stated provides a nobel idea that software
designers would like to comply with. However, it seems to me that 

1. A proper definition of the term components must be provided, where the
line would be drawn on where and how the platform idependence could be
achieved.

2. In addition, does plateform independence means the independence from the
actual operating system and/or programming language. 

3. In order to be able not to preclude any programming model, what kind of
interface do we assume for web services components?

4. In order not to assume any mode of communications between the individual
components, how would data be exchanged (if need be ?)


Other issues: 
------------- 

5. Is this redundant with D-AG0003: is sufficiently extensible to allow for
future evolution of technology and of business goals 

 when combined with D-AG0005:provides simplicity and ease-of-use that does
not impose high barriers to entry for users of web services 

Critical success factors 
------------------------ 

How platform independence could be measured, verified and achieved? and who
will verify that? 

cheers 
abbie