Cost of taking shortcuts

Joshua Schachter recently provided an excellent account on the problems of URL shorteners. URL shorteners are web sites which provide much abbreviated versions of web addresses. For example, the original address of this post ( could be shortened as below (I used TinyURL, currently the most popular URL shortener):

Services like these dramatically reduce the length of web addresses that when one cannot spare a lot of space writing down a long-winding URL (like jotting it down on a sticky note, or linking it on Twitter), one is forced to find methods to avoid the hassle. Here, a 65-character address has been slashed to a 25-character shortcut. Presumably there are much longer addresses on the web that you cannot bear to even copy and paste on a document, let alone writing it down with a pen.

As much convenient as they are, however, I have been sensing the same problem with URL shorteners that Schachter kindly described:

A new and potentially unreliable middleman now sits between the link and its destination. And the long-term archivability of the hyperlink now depends on the health of a third party. The shortener may decide a link is a Terms Of Service violation and delete it. If the shortener accidentally erases a database, forgets to renew its domain, or just disappears, the link will break. If a top-level domain changes its policy on commercial use, the link will break. If the shortener gets hacked, every link becomes a potential phishing attack.

“[P]otentially unreliable” may be an understatement to some. I have personally lost a whole set of useful links when one of these middlemen disappeared. Most of (previously accessible but) unreachable resources on the Web seem to be case that the original source has been deleted or moved, but once the location of the resource has been conveniently abbreviated by one of those URL shorteners, now the users have to worry about the longetivity of their databases too. Most of the web sites hardly last a number of years, and with a plethora of URL shorteners, many of those will disappear and leave a lot of people with invalid links to whatever precious stuff they found on the Internet, in a matter of time.

Yet, it is difficult to simply ditch URL shortening altogether and use web addresses as-is. Further complicated is the problem with a lot of Korean web addresses (I don’t know if the same thing happens to other 2-byte systems like Chinese and Japanese web sites) which convert Korean characters to Unicode. You would find it almost agonizing to share an URL on email or instant messengers like the following (the URL was randomly selected to demonstrate the staggering length of Korean-to-Unicode URLs):

The address above has a length of 449 characters, which, depending on language, may be as long as a full paragraph, making it impossible to paste it on a 140-character-limited Twitter post, let alone sharing it on an SMS (which is limited by 80 bytes in Korea). A few clicks on TinyURL would yield a much more elegant result:

Given the horror of a paragraph-long web address, it can be unthinkable to completely avoid any use of URL shorteners. It must be noted, still, these services are effective solutions to certain constraints as in Twitter or SMS, and it is always best practice to link the original, unabridged URL if possible. As for the owners of original web resources, especially bloggers in Korea and other countries that use 2-byte language systems, should avoid using 2-byte characters in web addresses, instead use numbers ( or alphabetic letters. Shortcuts do exist for certain convenience, but it should be acknowledged that using them, in turn, costs one the authenticity of the original path.


About this entry