[WEB4LIB] RE: 404 page -> hooray for Harvey Mudd College

[Roy Tennant]

I'm glad that Rich pointed out this page, since it's good to see an 
example of a site that really tries to help with its 404 page. But I 
have a couple nits to pick about it as well.

The page lists several URLs that were wrong in some published source, 
and then offers the corrected version. I say if they know about them, 
they should simply be redirecting any calls to those bogus URLs to 
the correct ones. I do this all the time, even with spelling 
variations (and if you don't know what these might be, monitor your 
error log). Why should the user bend over backwards? We should be 
doing that.

Rich mentions leaving the bad URL in the browser's "Location" box, so 
the user can chop back the url, or correct a typo, etc. That is also 
easily done by reflecting the URL in a form field that they can edit 
and then hit the "Go" button to go there. You can see this at 
http://sunsite.berkeley.edu/foo .

Finally, they miss some rather obvious methods by which users can 
locate what they're looking for: site searching and a site index, 
both of which they offer on their site (see http://www.hmc.edu/dir/ 
). So yes, high marks for an innovative way for letting users stroll 
up the hierarchy, but low marks overall for missing some basic 
methods for getting the user where they want to go.
Roy

P.S. By my own informal rule of thumb, if my 404 errors went above 
.5% (half a percent) I paid attention to my error log and did what I 
could to mitigate any error-causing situations. But I could never 
really get the error rate much below that, given the tendency of us 
all to make typing errors.

At 10:26 PM -0700 8/25/00, Richard Wiggins wrote:
>The article George cites has lots of nice examples.  One site they mention
>is Harvey Mudd College, which has a clever dynamic Error 404 page that
>includes clickable versions of the URL in error with each directory level
>"trimmed" from the right.  This is a simple trick but it could be very
>useful for users.  A standard technique for savvy users is doing this by
>hand; it's nice to offer the trimmed URLs as hyperlinks.
>
>Try this URL to see what I mean:
>
>http://www.hmc.edu/fee/fi/fo/fum/foo
>
>(You may have to scroll down a bit.)
>
>By the way, I think it's much better to leave the URL in error in the
>browser Location field, rather than redirecting to another standard page.
>If the "wrong" URL remains, the user can repair minor errors in the URL
>caused by transcription errors (for instance caused by copying too much text
>while copy/pasting).
>
>/rich
>------Original Message------
>From: George Porter <george at library.caltech.edu>
>To: Multiple recipients of list <web4lib at webjunction.org>
>Sent: August 25, 2000 7:53:19 PM GMT
>Subject: [WEB4LIB] RE: 404 page
>
>
>Brian Kelly produced a tour de force article in Ariadne last year on just
>this topic <http://www.ariadne.ac.uk/issue20/404/>.  It is well worth the
>reading, and implementing, for the benefit of your website and its visitors.
>
>George S. Porter
>Sherman Fairchild Library of Engineering & Applied Science
>Caltech, 1-43
>Pasadena, CA  91125-4300
>Telephone (626) 395-3409 Fax (626) 431-2681
>
>Richard Wiggins
>Consulting, Writing & Training on Internet Topics
>www.netfact.com/rww         wiggins at mail.com
>517-349-6919 (home office)  517-353-4955 (work) 
>______________________________________________
>FREE Personalized Email at Mail.com
>Sign up at http://www.mail.com/?sr=signup