Review of comments on the requirements document.

[ian.c.jones@bt.com]

Conference call participants

	I am unlikely to make today's call.  I decided to start to address
the review cycle as I suggested in my previous post so I humbly offer the
following start.

######
Comments on the comments received from Anders Grangard of Edifrance original
document from Eric Jarry of software AG France June 10th.

All numbers refer to the original comments from Eric,

I support comments 1,3,4(I think this is either Document exchange or
Choreography),10,11,12,13 without issue.

Specific comments

2 - I believe this may be an "English" i.e. language question I re-read what
we agreed and I see his point although I also think what we agreed is
correct to Quote the document 
		"In outline the working group will develop deliverables
that: ..... 
		8) define the information required that describes how to
interact with a service
		9) Provide a default method of usage that enables
bootstrapping of services"
I am not sure how to addresses this one.
EDITORIAL I thought we now had to use project team not work group.

5 - I am not sure what he is asking

6 - I suppose this comes down to if this is an online or printed document -
perhaps the steering committee need to address the online or printed and
colour issues!

7 - Are we trying to very general or is this a language and definition
question?

8 - Disagree. We should not tie our selves directly to IDL.

9 - Either the original or suggested change are O.K. I cannot see the real
difference.

14 - Disagree.  I think what is written is correct, his interpretation is
not what was intended  I do not know how we could make this clearer without
a lengthy definition and examples.

15 - I think this is a good comment alternate delivery addresses were talked
about repeatedly in Boston so I think we need to put the requirement.

16 - I understand his point but feel we need to finish our discussion on TPA
before we can put this one to bed.  I think a static QOS is possible in an
open environment it depends on the definition of QOS used.  I think he may
be thinking a little too strict, as in the QOS FedEx offers re guaranteed
delivery by 10 am compared to ordinary mail which will guarantee to tell you
it was delivered at some point in the future.  Both offer a QOS but are very
different, FedEx is a closed network point to point, but mail is not
especially if international.  I hope you all get the analogy it is a bit
contrived but I think it is difficult to prove that QOS can be defined over
an open network when people are used to only having QOS over closed
networks. As usual it is all about definition and shared understanding.

#####

I hope it starts the ball rolling,

Ian Jones

PP E1B
84-85 Adam Street, Cardiff, CF24 2XF
Tel:  +44 (0)29 2072 4063
Fax:  +44 (0)29 2046 1752
Email: ian.c.jones@bt.com