Date: Thu, 21 Oct 93 12:55:59 -0500
Message-Id: <9310211755.AA09523@joe.uwex.edu>
To: uri@bunyip.com
From: hoymand@joe.uwex.edu (Dirk Herr-Hoyman)
Subject: Re: <> One Last Time
Marc Andreessen writes:
>Terry Allen writes:
>> Re draft-ietf-uri-resource-names-01.txt, just announced, we still
>> have specified < > as wrappers for URNs in plain text. I'm going to
>> say it one more time, then shut up: This is the worst possible
>> choice, as plain text will often be upgraded to HTML
>
>...and (to try to separate this out from what may be seen as
>proprietary interests of the World Wide Web project) other SGML
>formats.
>
Gee Terry, as someone who encouraged you to speak your mind, I hope that
you aren't bothered if I do a little flip-flop :-). This is a problem for
SGML, but it's not relegated to just this use. ANY <>& needs to be
translated (as has been pointed out). HTML is but the first case of SGML
showing up as part of a UI that I am aware of, but more is coming. For
example, the COSE (which is a proposed standard for X GUIs) uses SGML for
help messages.
John Kunze writes
>Don't know if I can articulate why exactly relying on implied translations
>goes against all my instincts. More words won't help here anyway. I'm
>also very suspicious of solutions that involve statements like "oh, just
>use emacs", or "there will be more clever text editors soon".
>
I was a bit skeptical too, and then just yesterday someone told me about
HTML editing extensions for a popular (and free) Mac text editor BBedit.
Among these is one to translate <>&. So, as the collect fire power of the
net turns its attension to HTML/SGML, we can indeed expect this type of
support to be there.
>From what I have seen, lot's of folks are ALREADY using <> as delimiters
for URL, especially in mail/news. Perhaps this is because of the e-mail
address use. Whatever, we do have evidence that <> are indeed excepted.
If we were to change the proposal, I would support the use of {}. These
are not widely used in plain text. The only problem, is that I'm afraid
the horse may be out of the barn already on this one. And, I think one
poor standard is better than 2 good ones ;-)