ietf
[Top] [All Lists]

Re: What's an experiment?

2006-02-18 10:14:47
Hallam-Baker, Phillip wrote:

One of the most important uses of Experimental is to document
the minority position when a working group gets its cranium
embedded in its posterior. In a significant number of cases
the minority position has turned out to be right.

So here you'd say "let the minority have an experimental RfC",
in the spirit of "better document than not publish", besides
the minority might otherwise decide to block everything.

What if the solutions simply can't coexist, one of them harming
the other ?  What I said was "intentionally disrupting" ... "is
dubious".  If both sides get their way, but it cannot work in
practice, e.g. different semantics for the same bit, agents
don't know if it's a minority bit or a majority bit, then it's
wrong.

And if a simple solution how both sides could coexist without
worldwide upgrade stunts is possible, but one side refuses to
consider this, then it's malice.

Given the impressive lack of success of BEEP vs SOAP it would
be much better for the IESG to formally recognize that this
attempt to ratify a 'me too' protocol in 9 months has
backfired and is now harming IETF Web Services efforts.

Are they related ?  I had the vague impression that BEEP is
a way to multiplex one TCP-connection, and SOAP is a kind of
RPC with XML over http-post (and other bindings).  Not the
same layer, different problems => different solutions (?)

Both no experiments and not harming each other, or are they ?

                        Bye, Frank



_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf

<Prev in Thread] Current Thread [Next in Thread>