Lists Home |
Date Index |
Michael Brennan wrote:
> Well, there seems to be a healthy and thriving XML-RPC community out there.
XML-RPC is XML-RPC. As Amy Lewis has told us, SOAP isn't exactly RPC. It
isn't exactly NOT RPC but it isn't exactly RPC.
> I remember a project I worked on a couple of years ago -- integrating a
> legacy mainframe app with the web -- where we defined an XML messaging
> protocol to wrap the mainframe interface, and it ended up looking quite a
> bit like SOAP. When I started here at Allegis, they had already defined an
> XML messaging approach that looked quite a bit like SOAP (which was not out,
> yet, when they came up with it).
When you say it looked alot like SOAP do you mean to say that it did XML
over HTTP or do you really mean that it was structured in some
substantial way like this:
> There wouldn't be quite the media hoopla out there were it not for IBM and
> Microsoft pushing it so hard, but it -- or something very much like it --
> definitely would be out there and in use, because it's useful and the
> approach it adopted is a very natural progression for those doing web
If you're talking about XML-based RPC then I couldn't agree more. It was
bound to happen and it is probably good that it did. I'm asking about
SOAP in its entirety.
At least it was natural for me, even before I had heard of
> "SOAP". Maybe if I had been aware of REST I would have adopted a different
> approach. But as someone who has done procedural and OO programming for many
> years, who had experience with distributed systems, who had made the
> progression to doing web apps using ASP, JSP, servlets, etc., I naturally
> drifted to a SOAP-like model, as did other developers I've worked with. So I
> have no doubt that we would have something very much like SOAP even if
> Microsoft and IBM had never jumped on the bandwagon.
When you say this, what do you mean by the "SOAP model"? I'm asking you
to cast your eyes on the ink blot and tell me what you see. ;)