Message-Id: <9310172146.AA12667@rs042.scic.intel.com>
Date: Sun, 17 Oct 1993 14:42:57 -0800
To: Terry Allen <terry@ora.com>
From: kevin@scic.intel.com (Kevin Altis)
Subject: Re: <'s for URLs
>So now your HTML text says <URL:foo&rt;. Why not do the
>simple thing and choose delimiters that don't create a widespread
>problem?
Exactly! Maybe a decent set of bracket characters will also allow the URL
to contain common filename characters such as space (%20) without having to
escape it. I don't see how an URL can be considered easily human readable
when many of the characters have to escaped. I guess we're screwed with the
? and " characters, since they also have a special function, but spaces?!!
And we can't use curly quotes or real apostrophes because we're stuck with
7-bit ASCII. This appears to be some plot to make the whole world learn
hexadecimal and ASCII; I can't wait to tell my mom ;-(.
While I'm complaining. I feel it necessary to bring up the fact that URNs
at least should not be case sensitive if you want to be user friendly. God
only knows how many times we'll here "I'm sorry, you should have used
http://domain/Help, not http://domain/help." <-yes the example is an URL. I
realize there is a Unix leaning to many arguments on this list and that for
many existing protocols you have to supply the correct case for the locator
or you won't get a match, but even if we can't rectify this in URLs, we can
for URNs since they don't point directly to a specific location. Maybe we
can finally kill off the sorting notion that capital X,Y,Z... comes before
a,b,c, which comes before a umlaut..., since that is another stupid Unixism
we see in ftp...
ka