A difference that makes no difference is no difference (was: XML-RPC and HTTP)

[Russell Turpin]

Mark Baker writes:
>It knows *everything* about the *transfer* semantics of the discussion. It 
>doesn't know what the side effects of any stateful operation may be .., but 
>it knows what kind of transfer occurred ..

Not really. As someone previously noted, there is NO semantic
distinction in HTTP between GET and POST. The only difference
is that POST gives the client one more way to pass data.
In truth, and in how they are actually used, these would
more accurately be called INVOKE-SHORT and INVOKE-LONG, than
GET and POST.

Importantly, HTTP does not prescribe semantics. Not a bit.
It is a transport protocol, not a distributed language. With
a distributed language, such as Linda, one can write a formal
description of the effects of the primitives that must hold
in any correct implementation. In contrast, HTTP prescribes
only the format of a GET, saying nothing about its actual
effect. An application may eat the data passed, make some
important change in state, and steganographically confirm
that transaction through a special 404 page.

Or so it seems to me.

Russell


_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com