** Keith Moore wrote: **
> hmmm... one sure-fire way for a working group to get into trouble is
> for it to try to document current practice at the same time it's trying
> to define what should be the practice. people in the group tend to
> get the two confused and the result is usually neither an improvement
> over the current practice, nor an accurate description of current practice,
Which is why documenting current practice is separated from the development
work in the milestones :-)
> But documenting current practice isn't interesting to most WG members -
> they tend to want to do new work.
Some of us are committed to documenting current practice. Others want to work
on new protocols. If those two groups work together, we should get better
results.
One reason to document current practice is to get terminology agreement. Too
much confusion is currently due to lack of "same language".
> > So, if you can accept
> > that we are talking about intermediaries as referring to *all* entities
> > holding replicated copies of an original, then this fits well into your
> > above scenario.
>
> "intermediary" seems like an odd term for what could be called an
> "origin server" in HTTP jargon. I think it would be clearer to
> use the term to describe something that appears in the signal
> path between client and origin server.
In my understanding "intermediary" is "non origin server, non client", exactly
how you suggest the term should be used (for various definitions of signal
path).
Ingrid
This archive was generated by hypermail 2b29 : Thu Nov 18 2004 - 11:21:25 MST