Archive of UserLand's first discussion group, started October 5, 1998.
New perspectives on XML-RPC
Author: Dave Winer Posted: 1/22/1999; 8:41:07 AM Topic: New perspectives on XML-RPC Msg #: 2383 Prev/Next: 2382 / 2384
As the dust settles on the last round of work in XML-RPC, I've been gaining a new perspective on what we're doing and where it's going.First, it seems mostly to appeal to scripting people. The questions that have been coming up as we work with people in other scripting worlds force us to decide if XML-RPC is itself a scripting environment, a meta-scripting-environment if you will, and it turns out that it is! This is fascinating to me, because I've been playing with scripting environments for 20 years, and now it turns out that the net itself is becoming a scripting environment.
How so?
The nodes can be implemented in any environment on any operating system. The details are hidden behind the XML-RPC interface. That means you could switch out Windows NT with Linux (or vice versa) and as long as you maintained the XML-RPC interface, everything would work.
I know these are the same advantages that CORBA and DCOM are supposed to offer, and at its beginning, Apple Events (everything has been forgotten at Apple, it seems). But the problem with each of those is that they're well-supported on their native platform and sparsely supported on the others.
What's now happening with XML-RPC is that there are flat implementations across several operating systems. We've been encouraging the Python and Perl communities to feel they have equal ownership of XML-RPC for these reasons. No big platform vendor here, no W3C involvement, just a bunch of developers with a bias towards their own environments, working together, for their own reasons.
No platform vendor
To me, that's the promise of the net. No platform vendor. W3C is a consortium of platform vendors, forming a meta platform vendor. Now it's becoming clear why the W3C process isn't working any better than the old system. It *is* the old system, respun so as to appeal to people who bought into the promise of the Internet, that there would be no platform vendor. That's why W3C does not compute, and also, that's why the big platform vendors are doomed if they insist on making us all wait for them. Why should we wait?
In this space, I don't mind if my brothers in Python or Perl-land win. My stake is with the users, and the best way to align with the users is to insist on choice and movement, not lock-in and FUD.
This page was archived on 6/13/2001; 4:47:31 PM.
© Copyright 1998-2001 UserLand Software, Inc.