Reading USAToday.com over a bowl of cereal this morning, I clicked on story about the Columbia shuttle disaster only to get an error about a non-existent domain. I checked the link:
http:// cms-preview-site2-t.usatin.usatoday.com/ tech/ news/ 2003-06-04-colu
Looks like a “CMS preview” link accidently made it’s way into production.
I always get intrigued when I see things like this because I’m interested to know what content management systems the really big sites use. I viewed source on USA Today and found some code in HTML comments that I Googled, but I didn’t turn anything up. Too bad. I’d be interested to know.
Much easier is a site like ComputerWorld.com. View their source and right at the top you get:
Which leads me to another point: content management systems have URL signatures. You can tell a Vignette-powered site a mile away because the file name has about six commas in it (actually, it’s probably not a file name as much as it’s a key to a database record, but….).
Movable Type sites almost always have their individual entry pages in a folder called “archives” because that’s the default and a lot of people (myself included) never change it. Additionally, the individual pages are a number padded with zeros to make six digits. Look at sites like Gizmodo and ReachCustomersOnline — neither of them say they use MT, but you can figure it out after looking at a few URLs.
When I was working with Documentum last year, their WebCache product — which compiled and published pages to a Web server — included an HTML comment that promoted Documentum as the content management system of the site. (I just noticed that none of the HTML source on Documentum’s own site has the comment…) We shut it off, but I wonder if you could wheel-and-deal with a CMS vendor to leave it on in exchange for concesssions? “We’ll leave that HTML comment on so geeks like Deane Barker can see it in exchange for…”
Look at Wired — they have a Vignette-style URL, but no HTML comment. Maybe they didn’t get a free shirt or something?