[rfc-i] URL checking

[Paul Hoffman]

At 10:16 PM -0700 10/19/05, Bob Braden wrote:
>I apologize forf not responding earlier.  The answer is definitely maybe. ;)

Good. That sounds more hopeful than Joe's response.

>Seriously, checking URLs is part of our normal procedure, and 
>sometimes we heve
>to go back to authors about a bad URL.

Good to hear that! I'm glad you don't find it has prohibitive as Joe does.

>I was not clear from your message
>whether we had completed prrocessing on the particular document with
>a bad URL, but there is always a possibility that we missed one.

Nope: it's still in your queue. I was using it to find the right URL 
for a response to a mailing list when I discovered that it was wrong 
(a minor copy-paste-o that dropped the last character). The authors 
have promised to fix it in AUTH48.

>More generally, we try our best to avoid URLs that do not look like they
>have a reasonable chance of being stable over a reasonable time
>period.

Yep, this is a good policy. FWIW, the URL that sparked this 
discussion is a W3C URL, one that the RFC Editor has already allowed 
in the normative references of other standards-track RFCs.

At 5:29 PM -0700 10/19/05, Joe Touch wrote:
>Paul Hoffman wrote:
>  > Meaning, about one additional minute per RFC. Maybe this would not be
>  > prohibitive (or even measurable)
>
>One minute to check that a page comes up.

Boy, ISI should have a better internet connection than a 28Kbaud modem. :-)

>More than one to check that it is a sensible page.

Fully disagree. In the references section, a typical URL in a 
reference has a title and maybe an author, and the URL itself. 
Finding those items on the first screen of the web page should not 
take more than a minute, if that.

>Don't even count the minutes to see if each citation matches the
>appropriate page.

I don't. It doesn't matter. I highly doubt the RFC Editor currently 
checks that for citations to RFCs in the body of RFCs-to-be.

Fortunately, from what Bob says above, your negative view of the 
possibility of this is moot.

--Paul Hoffman, Director
--VPN Consortium