Shorten this
zeldman.com
zeldman.com
How much code is a URL-shortening service? Like 50 lines of code + a mysql table? It's like a day's work, maybe two, to handle 90% of the features people need. And it's like a few MB of database and almost no processing power for each request.
And there's like no lock-in. I have no reason to use bit.ly over TinyURL or =>.ws, because they're all nearly the same, and it's not like they'll retro-actively break the links if I use another service.
As a result, they're competing with almost zero differentiating features in a market with no entry costs, no lock-in, and no obvious way to monetize. There's no way most of these services are going to stick around when they realize this.
I'm assuming all, since non-shortened links can't be "monetized". (Feh.)
Personally, I suspect the real source of the objection to URL shorteners is that they break referrer tracking and other sorts of ad-related analysis.
That said, bits like this suggest that it might be slowly dawning on people that in 2009, we don't need mile-long URIs.
* - At this stage in my life, such publicity and content are only hypothetical :(
How many of those TinyURL links lead to pages and sites that no longer exist?
Now, if your site is something that you're serious about keeping up over the long term, including retaining or properly redirecting all the URIs you offer, great - you're ahead of most of the web. However, if you don't want people to shorten your URIs, then you'll want to make short URIs.
It's that adding it's yet another party in the middle that might fail, dissapear, etc at an inconvenient time.
What if your business relied on tr.im, for example, when they dissapeared? What if you'd used them in print so they were easy to type in?
Yes, I get the additional point-of-failure concern. Except, in practice, the points of failure seem to be squarely on the provider end, not the shortener in the middle.