Re: [DotGNU]Web Services: It's So Crazy, It Just Might Not Work

[Rhys Weatherley]

Bill Lance wrote:

> One quote from the article gets right at one of my own
> major questions about the whole WebService idea.
>
> "In contrast to the presentation of public and
> universally interoperable services, Web Services are
> actually best at automating private, previously
> negotiated conversations. The idea of unknown but
> perfectly described capabilities existing out there in
> the cloud -- at once unfamiliar enough to need to be
> discovered, tailored and reliable enough to build a
> business on, and not so critical that they need to be
> hosted locally -- describes a small and fairly trivial
> set of possible services. The ASP business ran aground
> on this issue, and there is no sign that the Web
> Services solution will work any better."

This is pretty much accurate.  If the two parties,
client and server, agree on the protocol to use to
communicate, then it all works out fine.  But the
goal of self-describing services that just magically
work, even if the client or server has never seen
the protocol before, is a myth.

The terms "Web Service", "ASP", "Web-enabled",
"Network Computing", etc, are all marketing
gobbledegook at the end of the day.  No one has
really been able to define them, because it simply
isn't possible to do so.

Every vendor for the last 8 or so years in this space
has claimed to be the main purveyor of *insert this
week's buzzword*.  All of their solutions have been
appropriate in the narrow market they targetted, but
were never general-purpose.

The ASP market was full of proprietry systems
running hacked-up CGI scripts with no common
standards involved.  These vendors used the term
"ASP" to give their systems an air of legitimacy when
they were no less proprietry than any other solution.
The current crop of "Web service" vendors are the
same-old proprietry ASP guys with a new buzzword.

Microsoft's .NET strategy is yet another variant of
this.  Is it a runtime platform?  An execution daemon?
An authentication service?  A programming language?
An XML schema?  Depending on who you ask, you
get different answers.  But there is no common thread
running through the marketroid-speak.  Microsoft
likes confusing the market, and they've done a good
job with .NET.

For DotGNU to be successful, I think we should stop
focusing on marketroid-terms like "Web Service",
and instead focus on technical solutions to actual
problems.

Having a runtime platform to execute IL and/or Java
code is a useful solution to the "code portability and
distribution" problem.  Having a distributed auth system
is a useful solution to the "I'm not at my home computer
right now, but I need to access my data" problem and
perhaps the "I'm sick of memorising 300 passwords"
problem.

Let's not fall into the trap of trying to apply marketing
labels to technical solutions to specific problems.
Solve the problem and let the market decide what to
call it after the fact.

Cheers,

Rhys.