Sunday, December 30, 2007

Redirection Reflection

Being as I'm moving a few appropriate pages from their original location on my home site over to this blog, I'm doing bit of redirection from the original URLs to blogsitedness (TM Reg. :-) ). Frequently seeing redirects when browsing with Lynx, I felt it was time look into the subject. As usual Wikipedia provides a reasonably thorough overview of the subject at

One of the more interesting aspects of the discussion were all the reasons for the practice, much more than my casual expectation. As typical for the Net some are hardcore technical issues, others in various parts of social phenomenon, some are perfectly innocent reasons, others 'suspect'. In short a microcosm of the factors that make up the WWW.

One of the things I gathered from the article, but didn't seem to be explictly spelled out, was that there seem to be three broad groupings of how URL redirection is achieved.

  • HTTP level methods
    • HTTP refresh header
    • HTTP status codes 3xx
    • Server mechanations, like SSS or special Apache directives
  • HTML level methods, primarily 'Refresh Meta tag'
  • Bogus methods, at least someone out there almost certainly considers them "problematic":
    • Manual redirects (Manual? what do we have technology for?)
    • Javascript (a good general purpose WWW whipping boy)
    • (Yet another) abuse of Frames

The first two HTTP methods grabbed my attention because of their flagrant use of the 'Location: ' HTTP header. I was slightly familiar with this from the way YouTube handles their video URLs. (See my usnatch project at (the soon to be redirected):

You can do it yourself recoding/reconfiguring, or have someone else (web host, special redirection sevice) do the deed, but it is an inevitable necessity of the constantly changing nature of the Internet.


Charles Wyble said...

Awesome stuff. Good link on redirection.

dallas said...

I've created a simple script to generate boilerplate
redirect pages at
Dallas E. Legan II 11 March 2008