W3C home > Mailing lists > Public > www-talk@w3.org > November to December 1996

Re: CGIs to manage backlink information

From: Benjamin Franz <snowhare@netimages.com>
Date: Tue, 17 Dec 1996 05:15:23 -0800 (PST)
To: Alejandro Rivero <rivero@sol.unizar.es>
cc: www-talk@w3.org
Message-ID: <Pine.LNX.3.95.961217050326.9045A-100000@ns.viet.net>
On Mon, 16 Dec 1996, Alejandro Rivero wrote:

> Hi again,
> After my experience with the script to capture REFERRER information
> (WWW pages pointing to pages on my server), I took a bit more
> of time to be able both to capture and serve such "backlinks" 
> within each html page. It could be a good reason
> for browser to implement finally the LINK tag.

Perhaps. But my experience (I've been recording and analyzing REFERRER for
about a year and a half now) indicates that even a moderately popular page
could be swamped with the number of backlinks - it really should have at
least one level of indirection -> pointing to another URL with the list
of backlinks. 

> It works fine on our CERN httpd 3.0 server. You can find
> the "report" in the same page that before,
>    http://dftuz.unizar.es/wwwlab/links.html
> Pointers to the new script appear near the end of the page.

A probably more practical (less server loading) solution for those running
Apache or NCSA is a log analysis tool such as RefStats
(<URL:http://www.netimages.com/~snowhare/utilities/refstats.html>) since
no CGI is required on those servers to record the referrer. I believe the
CERN server can generate the Combined/Extended format log as well: 
RefStats can extract the referrer information from that as well.

Benjamin Franz
Received on Tuesday, 17 December 1996 08:15:44 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:32:59 UTC