Hashbangs. Yes, again. This is important, dammit!
When the topic first surfaced, prompted by Mike’s post on the subject, there was a lot of discussion. For a great impartial round-up, I highly recommend two posts by James Aylett:
There seems to be a general concensus that hashbang URLs are bad. Even those defending the practice portray them as a necessary evil. That is, once a better solution is viable—like the HTML5 History API—then there will no longer be any need for #! in URLs. I’m certain that it’s a matter of when, not if Twitter switches over.
But even then, that won’t be the end of the story.
Dan Webb has written a superb long-zoom view on the danger that hashbangs pose to the web:
There’s no such thing as a temporary fix when it comes to URLs. If you introduce a change to your URL scheme you are stuck with it for the forseeable future. You may internally change your links to fit your new URL scheme but you have no control over the rest of the web that links to your content.
Therein lies the rub. Even if—nay when—Twitter switch over to proper URLs, there will still be many, many blog posts and other documents linking to individual tweets …and each of those links will contain #!. That means that Twitter must make sure that their home page maintains a client-side routing mechanism for inbound hashbang links (remember, the server sees nothing after the # character—the only way to maintain these redirects is with JavaScript).
As Paul put it in such a wonderfully pictorial way, the web is agreement. Hacks like hashbang URLs—and URL shorteners—weaken that agreement.