[thesite] anchors and 6 possible URL formats

[Daniel J. Cody]

jeff wrote:
> that's the problem though.  matt has brought up a valid issue of there being
> too many url formats to get to an article.  i think that's a valid concern.

I'm not introducing any new URL formats. Sure its a valid concern, but
its got nothing to do with the redirecting.

> on top of that, people using the old urls should be encouraged to update
> their bookmarks.  with some other solutions that we're exploring and setting

Maybe you missed isaacs post about doing a search of the evolt site only
for pretty much any content on google? All that showed up in the 20
pages of results were 'Update your bookmarks'. Obviously the whole
bookmark thing was being made clear, but because spiders dont follow
meta-refresh tags, and weren't following the link on the page, shit
wasn't getting indexed. In the last 2 weeks since I put the 301 in there
we've been getting indexed like crazy. Check out:

http://www.google.com/search?as_q=browser&num=10&btnG=Google+Search&as_epq=&as_oq=&as_eq=&as_occt=any&lr=&as_dt=i&as_sitesearch=www.evolt.org&safe=off
(wrap)

> the old-style url redirect back to an intermediate page would get us back to
> a single, consistent url when viewing articles.

Again, there seems to be the impression that the code I used is fucking
people up with a new URL scheme. As the whole keyword thread can attest,
I'm the #1 enemy of having any more URL schemes than our current scheme
since evolt.2.0.

To be clear, someone comes in with an old(evolt 1.0) URL like
/index.cfm?menu=8&cid=444 - walker and you put that script together that
would query the database, put together a new URL and display it on a
page telling people to update their bookmarks. The problem with that
whole scenerio though is that spiders weren't updating their
'bookmarks', hence the problem.

To correct the problem, I have Cold Fusion return a 301 status code
which stands for "Moved Permanently" along with a "Location" field
informing the client where the document can now be found. 

This is all the way it *should* be done, according to the HTTP/1.0 and
HTTP/1.1 RFC's which are available for everyones viewing pleasure at:
http://www.ietf.org/rfc/rfc1945.txt
http://www.w3.org/Protocols/rfc2068/rfc2068 (section 10.3)

I'd request any action on this issue to be held off untill I return from
the easter holiday on Monday.. 

Thanks

.djc.