Seeking Closure on D-AG006

Hi All,

   In order to come to closure on D-AG006 (see [1] for details), I will
start the final round of debate for each of the sub-CSFs: AC008 (see [3]),
AC012 (see [4]), AC015 (see [5]) for which agreement has not been achieved.
For details on the voting for these ACs see [2]. In an attempt to resolve
issues surrounding the CSFs and requirements, I have made proposals for
each. Without further delay:

AC008 [3]:

1. D-AC008.2 Architecture supports the concepts used in commonly accepted
design patterns.
Proposal: remove.
Status: This AC was an issue for a number of voters. Issues include clarity
and which design patterns the CSF is referring to. 

2. D-AC008.5 There shall not be wildly different means to achieve the same
ends in the architecture.
Proposal: remove. It seems unlikely that we will reach consensus.
Status: When we voted on this CSF, we seem to have a fundamental
disagreement on whether or not it is okay to have "wildly different" means
to achieve the same ends in the architecture. 

3. D-AC008.6 The definition and use of the components is consistent within
the Web Service Architecture.
Status: This CSF was added after the voting period.

AC012 [4]:

4. D-AR012.4 - usage scenarios and use cases must be referencable via
URI(reference)
Proposal: reword to "usage scenarios and use cases must be given a URI"
Status: We had two dissenting votes on 12.4. The issues were that it was
"purely editorial" and exactly what this implied ("giving URI name" vs.
"publishing at URI"). 

5. D-AC012.7 The Web Service Architecture must be validated against Web
Service Architecture use cases.
Status: This AC was added after voting and has never been voted on. 


AC015 [5]: While there seems to be consensus that time-to-market is a team
goal, there isn't consensus on most of the sub-CSFs. I have recommended
removing all of the CSFs that have not yet been agreed on which leaves AC015
sparse. We have concensus on time-to-market as a team goal and, as we have
seen recently (with the security WG discussion), time-to-market manifest
itself as the ability to balance the work to extend the existing
architecture (currently in use) with the work to create the overall (future)
architecture. However, I don't want to throw a hitch in things with a
complete re-write. So on with the recommendations:

6. D-AC015.1 Is the Web Services Activity a center for Web Services
standards specification, that is the community able to start new working
groups in a manner that is usable by the community?
Proposal: remove.
Status: There were two issues with the CSF: its intent and its wording as a
question. It is unclear that attempting to reword this as an imperative will
solve the issue of clarity. Further, as worded, the requirement is on the
web services activity, not the architecture or the WSAWG. It seems,
therefore out of scope to me.

7. D-AC015.3 Is the WSA document perceived as usable and referenceable in
time for products? New/revised products would be able to reference this
WSArch doc if it was delivered in time for their products.
Proposal: remove.
Status: This CSF also had two issues: wording and whether or not the CSF was
reasonable or achievable. It seems unlikely that we can agree on a
definition for exactly what "products" we are referring to and further, if
we could it would be difficult to determine that we were meeting this CSF.

8. D-AC015.5 Is the architecture document regularly revised?

Proposal: remove.
Status: The issue with this CSF is that it speaks of a longer time frame
than the life of the WSAWG.

9. D-AC015.6 Is the architecture document regularly referenced by other
specifications, including but not limited to W3C specifications?

Proposal: remove.
Status: The issue with this CSF was that it was worded as a question. It
could be reworded as such "The architecture document is regularly referenced
by other specifications" in the imperative. However, we now run into the
same issue as with AC0015.5, in that it speaks about a future time when the
WSAWG no longer exists. How could we meet this reworded CSF?


10. D-AC015.7 Is there a lack of press/developer commentary that refers to
time-to-market problems with WSA? To paraphrase, no press is good press on
this issue.

Proposal: remove.
Status: There were two issues with the CSF: wording and clarity. Even if we
manage to reword this in an acceptable and clear fashion, is it reasonable
to hold ourselves to and measure this CSF?


Comments?
Zulah

[1] http://www.w3.org/2002/ws/arch/2/06/wd-wsa-reqs-20020605.html#IDAPQ2OB
[2] http://lists.w3.org/Archives/Member/w3c-ws-arch/2002May/0225.html
[3] http://www.w3.org/2002/ws/arch/2/06/wd-wsa-reqs-20020605.html#AC008
[4] http://www.w3.org/2002/ws/arch/2/06/wd-wsa-reqs-20020605.html#AC012
[5] http://www.w3.org/2002/ws/arch/2/06/wd-wsa-reqs-20020605.html#AC015

Received on Monday, 15 July 2002 18:11:55 UTC