Date: Tue, 6 Nov 2001 11:15:10 +0100
From: "Jeroen Massar" <[EMAIL PROTECTED]>
Message-ID: <000701c166ab$eb9d2630$[EMAIL PROTECTED]>
Aside: I'm not on [EMAIL PROTECTED] so the copies
of messages I'm sending to that just bounce. I'm leaving it there so
replies others send won't drop the list by accident. But do be aware
that readers of that list will be seeing only a one sided exchange.
| With "numberic URL's" they probably mean the
| http://[2001:6e0::1]/example/ addressing...
That's what I assumed.
| Wellps... from a developers & network point of view one would like to
| see that feature though
Yes, as I said, the IETF is over represented with people who can perhaps
make reasonable use of it, and should know when to, and when not to.
But the much wider Internet community is not.
| As one could quickly check if a site is running without adding dns
| entries...
If it is running without DNS entries, what use is it? Especially if
a popular browser doesn't support finding it any other way? It might
just as well be down.
The supposed justification for those things is for accessing things that
aren't web sites but just happen to use http as a config tool, and similar.
| And it surely 'saves' the surfing crowd from the IP hassle...
It isn't the hassle - it is that if these things work, they will get
embedded in web pages. Perhaps by some web page designer who has
calculated that their web pages load 70ms faster (avoiding 3 or 4 DNS
lookups) when embedded addresses are used.
But that then makes it almost impossible to ever change those addresses,
if you fear that some other site might have links to your web pages
using numeric addresses, and you want to keep attracting visitors, rather
than discouraging them, then you have to keep the same address forever.
That's not what we want.
The best way to avoid that is to simply make this fail, and if that is
what Microsoft have done, then I applaud their foresight.
Note: I don't mind as much if "numeric URLs" work when some user types
them into a locator box - that's much less of a problem. Where they
really must not work, is when they're extracted from HTML. (Of course,
typically, one parser will handle all cases, so breaking them for one
breaks them for all.)
kre
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
