Great idea, awful URL:
http://www.nyc.gov/portal/site/nycgov/menuitem.c0935b9a57bb4ef3daf2f1c701c789a0/index.jsp?pageID=mayor_press_release&catID=1194&doc_name=http://www.nyc.gov/html/om/html/2008a/pr229-08.html&cc=unused1978&rc=1194&ndi=1
216 characters. And I thought C|Net was bad. A redundant alphabetic processional:
- portal/site/: in case the server forgets it’s a web site?
- nycgov/: isn’t the nyc.gov domain enough to indicate that?
- c0935b9a57bb4ef3daf2f1c701c789a0: sessions should not appear in the business part of URLs, but as variables. And 32 characters for a unique identifier? My math shows that as 1.2e+24 combinations. That’s a lot.
- Category data (1194) is repeated.
- Three times, it’s HTML. See? See? See?
The separate printer-friendly (grumble) URL uses a different identification scheme:
http://www.nyc.gov/cgi-bin/misc/pfprinter.cgi?action=print&sitename=OM&p=1214012284000
Of course, it’s redundant too. Sigh. I suppose “1214012284000″ is better than a Frankenaddress. But why not:
http://www.nyc.gov/html/om/html/2008a/pr229-08.html
That’s the URL served by some pages on the NYC.gov site (it redirects to the one above). Finally, here’s what we need–just the human readable “pr229-08″ and the “om” so we know it’s the mayor. Heck, throw in a few extra characters for a more readable date and office name, and we could do with:
http://www.nyc.gov/2008a/pr229-200806?for=mayor
Sure, maybe a little longer if there’s some essential data in those variables at the end, or if there’s a specific need to differentiate sites served by a single content management system. But why make it so complex?