FrontPage SiteMap RecentChanges RecentNearChanges HowTo Blog

Matching Pages:

RSS

Referrers

We’d like to know who links to our pages. Several solutions for AutomaticLinkBack have been proposed such as a TrackBack Ping, or a WikiPing (which never took off).

For Oddmuse, I designed a different system that did not rely on authors to run specific code to tell another site “I have created a link to your page” – instead I wanted to rely on the behaviour of readers. This is possible as RFC 2616 specifies that the HTTP request “allows the client to specify, for the server’s benefit, the address (URI) of the resource from which the Request-URI was obtained”.

Thus, if a visitor comes from elsewhere and reaches my site, I can get the URL of this “elsewhere” and do something with it.

The benefit is that I only record links if somebody actually follows the link to arrive at my site.

As with everything on the Internet, the question of spam arises, since the link can be faked. Therefore Oddmuse will fetch the source page and scan for the URL of my site. Only if this URL is in fact found does the link get registered.

Theoretically bots spidering the web can also generate these links. Usually, however, they don’t provide referrer URLs, therefore this doesn’t interfere with our data gathering.

A drawback is that search sites will link to our site, and requesting the search URL will in fact reveal a link to our site, generating a lot of useless data. That’s why there is an option to filter these URLs. If you discard any URL containing the term “search”, for example, you can solve this problem. On the other hand this is an indirect way of telling you what search terms people are using to find your site.

One last element of the solution is the decay: In order to reflect the current situation we discard older data. Per default, only data gathered in the last 24h is kept.

To see a list of all referrers to this site, see Self:action=refer

EditNearLinks: MeatballWiki AutomaticLinkBack TrackBack PatrickAnderson

Languages: