[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
denoting IPv6 addresses and ports
- Subject: denoting IPv6 addresses and ports
- From: [email protected] (Bill Manning)
- Date: Thu, 15 Feb 2001 09:49:27 -0800 (PST)
- In-reply-to: <[email protected]> from "Antonio Querubin" at Feb 15, 2001 02:25:06 AM
% On Thu, 15 Feb 2001, Robert Elz wrote:
% > Literal (hex/binary) IPv6 addresses should be seen by almost
% > nobody (there should never have been a standard way of putting
% > them in URLs, that is a truly dumb idea).
% I suspect there's a fair number of folks working in the tech support field
% who would disagree with you. Perhaps you don't do spend a significant
% amount of time troubleshooting combined httpd/DNS problems for customers.
% On a number of occasions I've had a need to explicitly specify an IP
% address and/or port in a URL. Ensuring that one can continue to do so in
% an unambiguous way even with an IPv6 address only helps win acceptance of
% IPv6 by reducing the annoyance level. While this standardization might
% seem dumb to you, I think it's important enough to at least some of us.
Actually, I beleive that many folks in the tech support field would
agree w/ Robert that "tech support" == "nobody"
Robert has been plinking w/ DNS for far too many years to be naive on this
subject. I understand the human factors considerations and would to
"deity" that we could abandon address literals from the lexicon of the
common UI. Under the covers, they have to exist and periodically either
agents or people will have to muck w/ them. For the good folks that
think IPv6 address literals are "difficult", you may wish to review the
next thing over the horizon... BitString lables. Mix Bitstrings with
DNAMES to build a composite IPv6 literal and then come back and tell me
that this is something you want my mother to use... Double dog dare you.