On Mon, 12 May 2003, Marshall Rose wrote:
There are several examples where working groups picked BEEP as
transport. I assume none of those groups regret their choice now. Can
you confirm/deny that they are all satisfied with BEEP, based on your
personal information/perception?
alexey - it would be inappropriate of me to comment regarding the
emotional state of ietf working groups.
If you say so. Personally, I do not see a problem with such comments,
especially if they just summarize the facts like "I have seen
virtually no public regrets or complaints about BEEP transport since
it was chosen" or "I am aware of such and such problems that have been
successfully resolved". I assume you have had exposure to those
groups/issues and was simply asking for your personal opinion.
if you want to talk to people, i suggest you contact the chairs of
the calsh and syslog working groups, along with andy bierman of the
xmlconf effort (which is not yet) a working group. alternatively,
you can send an email to the mailing list associated with each of
these three.
Calsh and xmlconf groups seems to deal with non-performance-intensive
applications. Syslog has some corner-case performance issues, and it
solves them by aggregating application messages (small log lines)
together, inside one BEEP message, something we cannot do in OCP. In
other words, it seems to me that these groups are/were solving a very
different set of problems compared to OCP, where performance was not a
major concern. Is my understanding correct? Are there any groups that
focus on performance a lot and are using BEEP?
And perhaps more importantly, do you know of any group (dead or alive)
that have chosen a custom transport after seriously considering and
rejecting BEEP? I wonder if we can learn from others mistakes here...
it would be inappropriate for me to comment on the ability of this
group to learn from the mistakes of others.
If you say so. IMO, constructive comments, especially negative ones,
may help. Not sure what ethical norms such comments would violate.
Thanks,
Alex.