consensus Call: TATWEEL
ebw at abenaki.wabanaki.net
Tue Mar 24 00:47:02 CET 2009
Hyphen and tatweel are very unlike.
Sent from my iPhone, painfully.
On Mar 23, 2009, at 2:06 PM, John C Klensin <klensin at jck.com> wrote:
> --On Tuesday, March 24, 2009 01:06 +0430 Alireza Saleh
> <saleh at nic.ir> wrote:
>> I think this is the registry choice to make it DISALLOW.
>> Would you please tell me the reason of using low-rise or
>> Hyphen in English and why it is allowed both for DNS and IDN ?
> That decision was, for all practical purposes, made around 1971.
> The intention was to have a way to divide mnemonics and
> structure names well before the DNS came along. So, for
> example, we had "MIT-Multics" (note the hyphen) which evolved
> into "Multics.MIT.EDU". The hyphen was then included in the
> DNS, at least in large part, in order to support the transition
> strategy which migrated from "MIT-Multics" (standalone host name
> and in the host table) to MIT-Multics.ARPA" (transitional DNS
> name) to "Multics.MIT.EDU" (target DNS name). I have no idea
> how many people thought of it as transitional in the DNS
> (gradually disappearing as we shifted from host-table names to
> DNS ones) or a permanent features.
> Low-line (underscore) has never been permitted, partially
> because of visual confusion issues with hyphen (there is at
> least one other reason, but it can be easily explained only in
> terms of 1403 print trains :-( ).
> And hyphen is a full character in Latin script, going back
> centuries, and not used for justification, calligraphy, or
> typography only. For some applications of the latter, there are
> other characters to represent the EMdash and ENdash... and they
> are not permitted in IDNs either.
> If there are lessons in the hyphen, they are that one should be
> quite careful to avoid, or at least consider the implications
> of, transition strategies that are likely to last forever.
> Idna-update mailing list
> Idna-update at alvestrand.no
More information about the Idna-update