Decoding URLs

Why is it that some web-addresses carry a long tail while other do not? How much does a web-address reveal?

For instance, a Yahoo mail-page has the following code:

http://us.mc369.mail.yahoo.com/mc/welcome?.gx=1&.tm=1296694680&.rand=e6tpkm8umv6mv#_pg=showFolder&fid=Inbox&order=down&tt=3571&pSize=100&.rand=850448145&hash=fa5c91801af9f69d3189c024f2592cd2&.jsrand=8243756

While  Facebook pages come up with something a lot simpler, like-

http://www.facebook.com/home.php#!/

Even though most web-addresses (aka URLs - universal resource locators) follow the format, << transfer protocol://servername.domain/directory/subdirectory/filename.filetype>>, they contain clues that tell you a number of interesting things:

• <~> - a tidle: Usually indicates a personal folder -- perhaps the customer of a Web host, or a student at a university, etc.
• <?> - a question-mark: Typically means that behind the scenes, a script will call information from the server or a database. Eg.,
• <=> - the equal-to mark: Indicates use of stylesheets

Some gurus recommend the use of CURL-tools to dig out more information from URLs...

-------------------------------------

LINKS / REFERENCES

• Decoding URLs http://www.library.jhu.edu/researchhelp/general/evaluating/url.html
  How to analyze a web-address - http://archive.virtualchase.justia.com/researchskills/quality_decodeurl.html
• How can I unwrap or decode Web URL redirects? - http://www.askdavetaylor.com/how_to_unwrap_decode_web_url_redirects.html
• What CURL can do - http://curl.haxx.se/docs/features.html
• Curl - A Gentle Slope Language for the Web - http://groups.csail.mit.edu/cag/curl/ (MIT White Paper)