2nd try: clarification for "remote procedure" facet

Earlier I wrote a possible clarification for the "remote procedure" facet,
<http://lists.w3.org/Archives/Public/xml-dist-app/2000Mar/0072.html>

But I think I've got an even more clear description.  The current
wording is:

    [Remote procedure] may mean the ability to pass generic procedures
    and have the other side have some mechanism for giving a
    best-guess response, or it may mean that there is a way to have
    the other side do something for you, ie. protocol.

Most of the protocols have some way to "have the other side do
something for you".  The distinction is in whether those things to be
done are defined by some application that uses the protocol or defined
in the protocol itself.  Another distinction is whether the protocol
explicitly supprts remote procedures or is just a carrier protocol.
"Remote procedure" should only apply to the former.

Here's a possible rewording:

    The protocol explicitly supports sending requests to a remote
    system to execute a designated function, method, or procedure
    defined by an application using the protocol rather than functions
    defined in or by the protocol itself.

This is specifically intended not to try to distinguish messaging from
RPC (if there is a difference), only whether functions are defined by
the protocol or by applications using the protocol.

  Spec                 Remote procedure
  ------------------   --------------
  BizTalk              no
  ebXML                can't tell from the doc linked on the matrix
  eCo                  can't tell from the doc linked on the matrix
  ICE                  no
  IOTP                *no
  Jabber               no
  LDO                  yes
  LOTP                 yes
  SOAP                *yes
  Userland's XML-RPC  *yes
  WDDX                 no
  Wf-XML              *no
  XMI                 *yes
  XMOP                 can't tell from the doc linked on the matrix
  XP                   no

Those marked (*) would change on the matrix if the above definition is
used.

  -- Ken

Received on Sunday, 23 April 2000 10:36:54 UTC