[
Lists Home |
Date Index |
Thread Index
]
5/30/2002 8:41:26 AM, Jonathan Robie <jonathan.robie@datadirect-technologies.com> wrote:
>http://www.webservicesarchitect.com/content/articles/mark04.asp
>
>Is this a good overview of where we are at? Does it leave out anything
>important? Is it balanced?
I think its pretty decent and balanced. A fairly concise summary of the
"conventional wisdom" that web services have been horribly over-hyped,
but that shouldn't obscure their real potential.
I think it downplays one important fact -- web services really ARE being
used "behind the firewall" for application integration to a fairly
significant extent. That avoids the two big problems of security
and interoperability (because enterprises can standardize on one tool
to generate the SOAP/WSDL interface, or at least put in a process to
tweak the generated interfaces until they interoperate).
It misses the big issue in the W3C web services activity, which is
the extent to which web services are really part of "the Web" or merely
a "cuckoo's egg" (to use Edd Dumbill's metaphor) that warms up in
the Web's nest. There is currently a lot of work going on in SOAP 1.2
group to find a middle ground that makes it more Web-friendly
but accepts that there are legitimate use
cases for "tunnelling" SOAP over HTTP; the challenge is to educate users
(and vendors, I suppose) to make sure that they understand how to
make appropriate design decisions. I can understand why the author missed
this, since this has happened pretty fast. [BTW, I see this as real
evidence that some of the changes at the W3C are having a very positive
impact -- the web services groups all do their technical work in public,
and the TAG has played a very useful role in nudging people to think more
about the overall Web rather than just their own tunnel or silo.]
My other main quibble is that the obligatory mis-mash of acronyms is
trotted out (XLANG, WSFL, BPSS, BPML) with no explanation or
analysis of how these will become real standards (in either the
legal or de facto sense), what value they really add, etc. On the other
hand, I did note that the author voiced considerable skepticism
about UDDI rather than the usual inclusion of it
with SOAP and WSDL as part of the trinity of web services standards.
This indicates that the author is reading the tea leaves in the same
way that most people I talk to are reading them.
Overall, a reasonable, but not "must read" article IMHO.
|
