URL Shorteners VS GET Variables

You know, I was just looking at a shortened URL the other day, and it was something like "domain.com/1jx820xu". Then I started thinking about the whole "SEO friendly" URL craze, which in turn lead to good old GET variables (?id=abcdef). Why not just go back to using GET variables?

Get people to stop using Twitter and either way will be fine.

Using GETs as part of SEO can help, but one of the factors in SEO is density, therefore, the part of the GET that isn't optimized will work against you.

Compare

/?search=keyword
vs
/keyword

a-z0-9 = 36^8 = 2,821,109,907,456 = 2.8 Trillion possible keys

It took Google years to index 4 Billion URLs, how quickly are people exhausting all of the possible keys here, and more importantly, are any of the domains being used for URL shortening staying in business anywhere near long enough to exhaust those keys ?

I want to know how these shorteners are interpreting the keys in the background, are they using something like mod_rewrite to convert to a GET variable anyways so that existing technology can work with it, are they using purpose built HTTP servers that simply look the URI up in a hash table and either redirect or return a 404 ?



You mean search engines are treating the entire URL like a string of text, and not parsing it into pieces? That would make me think that all of the <h2> elements in my HTML are diluting the density of my articles keywords.

I'm sure Google is doing some parsing with URLs, because there's a lot of options in Google's tools such as Webmaster Tools and Analytics that make it possible for us to specify how Google should handle specific GET variables.

--

Also, there's actually a pretty interesting read about URL shorteners at Wikipedia.
http://en.wikipedia.org/wiki/URL_shortening

The problem with obfuscated urls? HUGE security risk. I understand why people use them to save screen space but memorise that and I will be impressed! Get back on the friendly urls, I mean they are, well (ironically) friendly