At 06:26 AM 1/20/2003 -0500, Christopher B Ferris wrote:

Mark,

Please see my comments below.

Cheers,

Christopher Ferris
Architect, Emerging e-business Industry Architecture
email: chrisfer@us.ibm.com
phone: +1 508 234 3624


xml-dist-app-request@w3.org wrote on 01/15/2003 12:10:13 PM:

>
> Here is another list of requirements following the ARTF (Attachment
> Requirements Task Force) telcon held from 11am-noon EST on Jan. 15,
> 2003.  A couple of former draft requirements have been moved forward
> to "Considerations" instead.  Those requirements marked "DR" have not
> yet been discussed in detail by the task force; those requirements
> marked "R" have achieved consensus within the task force.
>
> Mark A. Jones
> AT&T Labs -- Strategic Standards Division
> Shannon Laboratory
> Room 2A02
> 180 Park Ave.
> Florham Park, NJ  07932-0971
>
> email: jones@research.att.com
> phone: (973) 360-8326
>   fax: (973) 236-6453
>
> ================================================================
>
> Concrete Attachment Feature Requirements
> ----------------------------------------
>
> Considerations
>
> * The specification should not invent a packaging scheme.


Agreed.

Perhaps I don't quite understand the meaning of "packaging scheme" but the way I
interpret this is "the ARTF is going to pick between SwA and DIME", which isn't
truly possible since neither are sound enough.  Perhaps you mean
   The specification should resemble existing packaging schemes.
?

 

> * The specification should aid debugging with simple tools.


This has me baffled. What is it that you have in mind in the way
of tools, and more specifically, are you suggesting that the specification
would define said tools or that the specification would define
a concrete binding that had as a design consideration that an implementation
would be inherently debuggable? Further, what manner of "debugging" are
we talking about here?

I think this one was added at my suggestion.  I would word it:
  The specification should rely on plain ASCII headers.
Plain ASCII (no not internationalized) makes debugging message systems
considerably easier.  Compare anyone's experience in working with HTTP on
the one hand and RPC/Jini/Corba on the other.  Please note that ASCII does
not mean unformatted ASCII.  Processing many small messages (~packet
size) would benefit from fixed formats.


John J. Barton          email:  John_Barton@hpl.hp.com
http://www.hpl.hp.com/personal/John_Barton/index.htm
MS 1U-17  Hewlett-Packard Labs                                   
1501 Page Mill Road              phone: (650)-236-2888
Palo Alto CA  94304-1126         FAX:   (650)-857-5100