5 replies to this topic

[Randy]

From last week's grab bag of goodies.

Source : Yahoo Search Blog

Today comes a new wave for search engines with the first-ever Beta launch of 'Dynamic URL Rewriting' in Site Explorer. The new feature provides the ability for site owners to alert Yahoo! of the dynamic parameters in URLs that they'd like Yahoo! to ignore, which we'll then automatically rewrite accordingly. Try this out for all the cases where you'd want to use parameters in your URLs that don't affect the content of your page, but that have other important uses.

The two most important uses I can see immediately are being able to tell Yahoo! to strip off affiliate ID's and Session ID's, so as to not only avoid duplicate content but also to merge link popularity more quickly and efficiently. Apparently this will also help them to crawl your site more thoroughly. Beyond that it would also be an easy way for webmasters to see if they have a duplicate content issue being created by certain query strings.

Hopefully Google will offer something similar before too long. It's a good idea IMHO.

[Jill]

Wow sounds great. There are many sites that use tracking codes due to their analytics software. This will be a huge help to them. for yahoo!

[Randy]

Yep, it would alleviate all issues with tracking parameters too. Dunno why I didn't mention that one since it's a huge potential thorn in my side.

It's such an elegantly simple solution to a massive problem too. Dunno who came up with it at Y!, but I wish someone had 10 years ago! IMHO they should also get a raise. :whislte:

[jehochman]

They've done this the wrong way. There needs to be a meta tag, like this:

CODE

<meta name="IgnoreParameters" content="SessionID, SortOrder, Browser" />

It doesn't make sense to have to register with the search engines to make your site work. Like robots.txt and the robots meta tag, we need the search engines to agree on a standard format that they all share.

[Alan Perkins]

IMO it can't be done effectively in a meta tag. It needs to be done at the site level. robots.txt would have been a good place to put it.

CODE

#! robots.txt 2.0
#
#<QueryParams>
#<Ignore><Param>SessionID</Param></Ignore>
#<Ignore><Param>Browser</Param></Ignore>
#<OverRide><Param>SortOrder</Param><Value>Alpha</Param></OverRide>
#</QueryParams>
#

What would be good in a meta tag would be something like:

CODE

<META NAME="robots" CONTENT="permalink=http://url.to.index/path/?query-string">

[Randy]

IMHO the reasoning behind Yahoo! doing it the way they've chosen is because of the relative success of Google's Webmaster Central console. People will sign up for such things, thus becoming less anonymous, if they're useful. And if they (all of the Engines) allow you to control more than one domain from a single account it gives them a lot more control of their own systems with no extra work. eg If they see something nefarious on one site, they can easily check other domains on the same account to see if they're doing the same sort of shady things.

If I were them I'd probably do the same thing, even if things could be made a lot simpler by using robots.txt or metas or whatever in the code itself. If for no other reason than that requiring a login removes some of the anonymity of the web for them.