In message <199606272053.AA03700@zephyr.isi.edu>, Bill Manning writes:
I thought we already have one: port 43 for whois services.
We will put the production code on port 43 as soon as the code looks stable for a while.
Please correct me if I missed your point.
David K.
I am sorry that I was not clear. I beleive that, even with a superset of the whois protocol, this is no longer whois as defined and should therefor apply for a new well-known port.
Local conversations with several others here at IETF do not support this view, so I expect that it (the view) will remain a minority opinion. Its something I thought worth mentioning in passing... :)
Of course experimental stuff can use any port for testing. This, being a purely local matter can occur on any port the sysadmin chooses.
-- --bill
Bill, As long as the is backward compatibility no protocol has had to apply for a new port number when it was extended. Curtis