Article+Title+Here
July 7, 2005 3:50 PM   Subscribe

Stupid technical question: Why does CNet put the title of their articles in the URL for said articles?

For example:

http://news.com.com/Microsofts+personnel+puzzle/2100-1022_3-5770771.html

Removing the article's title from the URL does nothing (that I can tell) to the page, it certainly doesn't break the link.

Is it just for SEO reasons? If so, the URL has more weight than I thought.
posted by o2b to Technology (9 answers total)
 
but then why not just use the title alone as the url?

i had a similar problem in the system i'm currently working on - we wanted human-readable uris, but that led to a security hole. so in our case you have things like
ivo://nsa.noao.edu/proposal-10/2005-06-87/45:5fe45a
which would be the 45th image taken that night for proposal 10. the junk at the end is a mac (message authentication code) that guarantees (to some quantifiable extent) that the name is correct (ie exists in the system). but i don't see how that makes sense in the case of cnet (and if you don't give the extra bit to our system, it will still work - it's just slower as it has to do extra security checking; cnet gives an error if you give it just the title).

hmmm. maybe it makes server log analysis easy? never underestimate the influence of a dirty hack.
posted by andrew cooke at 4:10 PM on July 7, 2005


The Boston Globe does the same thing (e.g. http://www.boston.com/business/technology/articles/2005/07/07/boston_data_firm_loses_calif_bank_tapes).
posted by ericb at 4:15 PM on July 7, 2005


Google looks very very kindly upon words that appear in URLs. So it's a simple hack to boost search ranking.
posted by cillit bang at 4:17 PM on July 7, 2005


Keep in mind that we are talking about a company that decided news.com.com is preferable to news.com. I'm not sure it's worth reading too much into it.
posted by Galvatron at 5:05 PM on July 7, 2005


Yup, SEO.
posted by blag at 5:14 PM on July 7, 2005


It's also good if you forward the URL to other people or if you see it on del.icio.us etc.. you get to see some useful information before you even hit the page. I think it's a good idea.
posted by wackybrit at 5:48 PM on July 7, 2005


Google really likes it! Also, if you're typing news.com.com into your address bar looking for an article you already read, you'll be able to figure out very rapidly which one it was. That's the reason I use it on my blog. It's more usable, it gains search engine juice, what's not to like?
posted by evariste at 8:15 PM on July 7, 2005


CNET's using the content management system that was commercialized as Vignette's StoryServer. It's trying for the best trade-off between dynamic content and fast delivery. The URLs used to end with 4 comma-separated integers (you can still see this on other sites.) These represented story ID, whether or not it could come from the cache, or had to be built dynamically, and other stuff I could have once told you off the top of my head. Obviously, the scheme's changed, but the point is that there's useful and discrete pieces of info embedded in the ugly numbers.

If they used just the story name in the URL, they'd have to do a database lookup of the rest of the info for every page request. For a zillion page requests, that would add up.

(I worked for CNET way back in '97 when Snap! was going to take over the world.)
posted by Zed_Lopez at 10:17 PM on July 7, 2005


I'll second the 'Yep SEO'.

Aside from Google PageRank ( the equation that rates your page in relation to others), Google's index also weights the title tag, early H[1-6] tags and words in the URL highly. For instance http://www.petstuff.com/dogfood/index.html will rank higher for 'dog food' than www.petfood.com/canine/ if their PageRanks are the same because of it's placement in the URL.

most blog engines also add the title to the url (aka 'The Slug')
posted by oliyoung at 10:30 PM on July 7, 2005


« Older What would happen in a train crash?   |   Makeup for oily skin Newer »
This thread is closed to new comments.