Re: Proposal: New Anchor attributes

Daniel W. Connolly (connolly@w3.org)
Tue, 28 May 1996 23:15:58 -0400


Message-Id: <199605290315.XAA23069@anansi.w3.org>
To: marcush@crc.ricoh.com (Marcus E. Hennecke)
Cc: www-html@w3.org, ronnie@driver-aces.com
Subject: Re: Proposal: New Anchor attributes 
In-Reply-To: Your message of "Fri, 24 May 1996 13:27:29 PDT."
             <199605242027.NAA00672@cougar.crc.ricoh.com> 
Date: Tue, 28 May 1996 23:15:58 -0400
From: "Daniel W. Connolly" <connolly@w3.org>

In message <199605242027.NAA00672@cougar.crc.ricoh.com>, Marcus E. Hennecke wri
tes:
>On Thu, 23 May 96 17:57:47 PST, Ron Schnell <ronnie@driver-aces.com> wrote:
>> 1.  ALT=[URL]
>> This attribute would specify an alternate URL should the HREF be
>> unavailable.
>
>It is quite trivial to put in multiple A records for one hostname.
>Suppose the following is in the DNS:
>
>foo.bar.com   IN A    127.1.2.3
>              IN A    254,40.50.60
>
>Then any telnet or ftp client will first try 127.1.2.3 and when it fails
>will try the second, third, etc. address.
>
>I just don't know why no web browser has implemented that as of yet. Are
>any of the browser vendors listening? Why is it that web browsers only
>try the first address and then give up?

Any web browsers derived from the W3C reference library do
The Right Thing here.

Henrik is going to write a short draft on the topic.

For some background on this and related issues, see:

	http://www.w3.org/pub/WWW/Addressing/Activity#dns-problems

Dan

p.s. more on ALT URLs in another message...