Re: web services uber alles? [was] Re: [ebxml-dev] RE: ebXMLdeliveries

[James Bryce Clark <jamie.clark@mmiec.com>]

Hi Mike.  Cheer up, we're just in a thick
patch of stupid, competitive smoke & mirror fog at the
moment.   

> Just being a curmudgeon I
> guess ;^).  But please allow me to be a bit grumpy on a Monday after
> having invested 18 months and considerable personal resources in ebXML
> and having it wind up like it has.  (I read two comprehensive
> articles on "Web Services" in recent trade journals today and
> there were no mentions of ebXML.)

Of course there weren't.   Unlike
software vendors, the owners of the "ebXML" name aren't paying
PR firms big bucks to exploit the latest buzzword.  I am mildly
disappointed that the OASIS side of the house is going its own,
proprietary-laden way, and that CEFACT has taken so long to pull out of a
recursive navel-gazing loop.  Neither is really minding the open
standards store or the ebXML brand very well at the moment.  But
that optic flaw does not affect the merits of the work.

> I can probably find a way to
> say that my dogs offer web services.  * * * 

If you start an OASIS WoofML TC, I will join
it.  And I see no reason why it should fare any worse, or fit in any
less, than all the other web service offerings or TCs.  The
aggregate vector sum of OASIS work product seems to be small, as various
initiatives cancel each other out.   There are many bright
spots, and much pushing and pulling, but little aggregate movement. 
Sort of reminds me of litigation, or my physics teacher's demonstration
that if you push against a wall for a long time, and it does not move,
you have exerted Force but accomplished no Work.  It may be that the
OASIS vendor-straddling power structure does not really permit much
movement in any one direction.  I hope they will be able to evolve so
that it makes more sense for nonaligned people to unreservedly donate
their efforts there.

> I guess my mention in the
> particular context about the BPSS was inspired by your mention of
> "140" change requests to fix the BPSS 1.0 spec.   If
> 1.0 was that off the mark  * * *

Well, it wasn't.  The last printed change log we have from from BPSS 1.0 is dated 4/27 (not counting a set of post-Vienna typo edits)>  It has 137 items representing every public comment thru that date, meritless or not.   Many of them were resolved in the 1.01 final print, or combined, or properly discarded.   I expect that most of the immediate work is reconcilation of the BP standard with other related artifacts that were developed in parallel --- CPA, MSG, UMM.  Unfortunately politics and big firm economics rushed the 1.0 ship date, and there were fewer hours reserved for cross-spec coordination than would have been useful.

That still leaves us with a doable refinement to a best-of-breed, open system, with a clear potential for compatibility all the way from one-off EDI messages to long-running modeled collaborations.  Best of all, there are now newcomers kibitzing the work, finding flaws and options that the authors were too close to see. 

What's not to like?  I mean, other than that it's a Monday.   Jamie

James Bryce Clark   
VP and General Counsel, McLure-Moynihan Inc.
Chair, ABA Business Law Subcommittee on Electronic Commerce  
1 818 597 9475    jamie.clark@mmiec.com    jbc@lawyer.com