Re: [ubl-dev] ebXML approval retrospective

["Fraser Goffin" <goffinf@googlemail.com>]

Fulton/David,

agree with your points although I don't paint such a black picture (as
a pointer - it doesn't help that ebXML proponents often seem to be
their own worst critics). Yes SOA/SOAP has greatly under-performed
against the ridiculous hype that fan-fared its arrival, but IMO most
designers and technical architects have realised this for quite some
time and have focussed attention on the subset of the available and
genuinely useful specifications and standards to build real service
based solutions (that is certainly what I have been up to for the last
few years and it has been more than possible to implement
sophisticated and secure services, including workflow and BPM (not the
same thing - but that's another debate :-).

Timing is the critical issue (as per my earlier comment about ebXML -
maybe I tried too early !), and whereas that does operate as a
constraint it is can be of assitance to organisations that need time
to develop and mature their understanding of what it takes to put
together and manage a services portfolio (deciding on granularity,
governance, versioning, managing service registry, canonical data
models et al).

IT is a practical profession and those of us that are practitioners
can see through the hype-curve. Hopefully we are coming out of the
trough of disillusionment and are moving into the slope of
enlightement :-)

Regards

Fraser.

On 15/05/06, David RR Webber (XML) <david@drrw.info> wrote:
> Fulton,
>
> General Inertia indeed holds the key.
>
> During the height of the .com thrash corporates were willing to
> throw $ at anything to do with XML out of fear and panic.
> Once they realized that instead of a tidalwave they
> were facing a slow tidal swell - the rush was off - and
> in-place EDI systems continued to offer hard to beat
> ROI migration v Maintenance in-situ metrics.  AS2 also
> is a factor - offering low-cost internet delivery with
> old EDI payloads - bad limitations - but known factors.
>
> Now - we see new deployments using ebXML because
> of the better infrastructure this gives you - but - yes
> its not just ebXML - its across the board - that
> just-write-code approaches are proving remarkably
> resilient - for a lot of reasons we have noted earlier.
>
> Once however the newer infrastructure has some
> proven metrics and success stories - then pressure
> will once again be on corporates to get caught up
> in the IT technology race.
>
> Peter Fingar has thrown out this piece - but we are
> yet to see this translate into aggressive adoption
> as opposed to just-in-time adoption!
>
>  http://www.mkpress.com/extreme/
>
> DW
>
>  -------- Original Message --------
> Subject: RE: [ubl-dev] ebXML approval retrospective
> From: Fulton Wilcox <fulton.wilcox@coltsnecksolutions.com>
> Date: Mon, May 15, 2006 9:42 am
> To: 'Fraser Goffin' <goffinf@googlemail.com>
> Cc: ubl-dev@lists.oasis-open.org, xml-dev@lists.xml.org
>
> Fraser,
>
> From a five-year retrospective viewpoint, I think it is accurate to say
> that
> SOA/SOAP has greatly underperformed both "expectations" and, more
> importantly, its practical potential. If you scanned a large company's
> environment and relationships to identify where SOA/SOAP would be better
> than what is being used (often not a high hurdle because what is being
> used
> is still fax, phone calls, etc), my guess is that the glass is about 2%
> full, even if you toss in variants like REST and pure push such as RSS.
>
> Whereas ebXML has tended to be overlooked, SOA/Soap has been smothered
> in
> "marketing love." The resulting cloud of FUD and positioning messages
> from
> both leading edge and trailing edge suppliers and consultants, and some
> IT
> quarreling over, say, differences in numeric precision, cost two or
> three
> years out of the past five.
>
> Besides SOA and ebXML, one can also consider the languishing state of
> Business Process Management (BPM), which tends to be stuck in corporate
> niches and pilots, and, a further indignity, has had its three letter
> acronym hijacked by the "business performance measurement" crowd. The
> underwhelming adoption of BPM makes it somewhere between rare and
> infinitesimally unlikely that even a medium sized corporation has
> implemented an effective trans-functional, widely used workflow process,
> and
> that of streamlined workflow helps retard the ROI and adoption of
> inter-enterprise solutions.
>
> Given the synergies among SOA, ebXML and BPM, lags in adoption tend to
> be
> mutually reinforcing. Indeed, if one looks at the IT landscape, the
> number
> of deserving, but languishing IT advances resembles the air traffic
> pattern
> at a major airport during dark and stormy weather - backed up for 500
> kilometers.
>
> Out in the great world beyond IT standards or even IT folk, we are
> competing
> with the delaying tactics perfected by that military genius, General
> Inertia. The General tends to be quite content with offerings of
> yesteryear
> and can employ divide and conquer techniques to hold off the newbies for
> decades.
>
> Where all this becomes a worry is when, for example, we try to
> streamline
> the services economy, notably healthcare, because General Inertia stands
> in
> the way and, quite accurately, points out the migration costs of moving
> ahead.
>
>
> Regards,
>
> Fulton Wilcox
> Colts Neck Solutions LLC
>
>
>
>
>