Sun, 26 Aug 2001 23:26:37 -0700
Mostly the decisions have to do with scaling. For things that are accessed
frequently and don't change much we do stating renderings. Sometimes
dynamic. When we're creating interfaces we use XML-RPC, and in the last few
months as our SOAP implementation started to interop with others, we use
----- Original Message -----
From: "Paul Prescod" <paulp@ActiveState.com>
To: "Dave Winer" <firstname.lastname@example.org>; <FoRK@xent.com>
Sent: Sunday, August 26, 2001 6:42 PM
Subject: Re: Cross-post
> Dave Winer wrote:
> > > As you have recently decided to do both GET/XML-style services and
> > > XML-RPC services, maybe you could offer your opinion on the
> > > circumstance for each.
> > You drew an incorrect conclusion from that post.
> > I have been creating systems around static and dynamic GET requests on
> > for years.
> > http://backend.userland.com/
> Does that imply you HAVEN'T decided to use GET/URI recently and
> therefore cannot express an opinion on when that style is appropriate?
> The URL you cite says: " But much of the same content is available in
> XML and in snippets of HTML, and thru XML-RPC and soon thru SOAP."
> How do you decide when to make information available in GET-accessible
> XML, in a custom XML vocabulary, in XML-RPC and SOAP? Do you just always
> build all interfaces to all information "just in case"?
> Take a recipe. Leave a recipe.
> Python Cookbook! http://www.ActiveState.com/pythoncookbook