Messages longer than 160 characters (the common SMS carrier limit) were split into multiple texts and delivered (somewhat) sequentially. There were other bugs, and a mounting SMS bill. The team decided to place a limit on the number of characters that would go out via SMS for each post. They settled on 140, in order to leave room for the username and the colon in front of the message.

Of course, 140 characters is now part of the lore and essence of Twitter.  It’s as sacrosanct as McDonald’s having two arches (despite starting with only one) or a Swiss Army knife folding up.  (An April Fool’s joke had the company appearing to offer “Twitter Premium” with 160 characters and 50,000 instant followers.)

But there could be good reason to stretch the limit — or allow for a slightly more nuanced set of data behind a tweet — and not just because 140 characters might cut off key information (to be sure, an Onion-ey link) and have only half a thought re-tweeted before it’s fully completed.  In fact, the “retweet” — expressed as “RT @[source] [source's original tweet]” — is a great case study on why.  As danah boyd et al explain, retweeting is one of several “behavioral conventions” arising from Twitter users themselves.  Twitter itself does not have a special “RT” functionality; it’s just two letters that many people have come to use to say that they’re repeating something they saw elsewhere — and giving attribution for it.

With the 140 character limit, though, the “RT” and attribution have to fit too.  An original tweet that’s already near the limit will have to be shortened for it to work.  And if someone retweets further, the cycle continues.  Danah & Co. have some great examples in their draft of ways in which that retweeting can inadvertently distort or even negate the original message.

Twitter has already shown a willingness to adopt users’ conventions.  The use of @___ at the beginning of a tweet to communicate with a user became so common that it made sense for Twitter to put a special link on each user’s home Twitter page to view “@replies” from others.  For retweeting, Twitter could choose not to count RT @___ against the 140 character limit.  That could cause some tweets to be truncated when forwarded to SMS, or divided into two messages, which was an original reason why the limit was adopted.  But that might be worth it at this point.  Or, Twitter could start implementing metadata for tweets.  Already it records a timestamp and source for a tweet (since people can inject tweets into the system in so many different ways, not just at twitter.com.)  Retweets could become part of that metadata, not necessarily fully transmitted as part of the full message itself.  That way, the 140 character limit could be maintained, but people could still follow the genealogy of an idea, right back to its source — just the way that the “in-reply-to” link on twitter.com lets someone unravel an entire conversation with just a click.  And being able to lengthen a tweet in special circumstances could, if applied to URLs, also help avoid the need for the risky phenomenon of URL shorteners. [Update 10 August 2009: tr.im just announced it is folding; shortened links will work for awhile but not forever.]

Twitter is a foundational technology.  By that I don’t mean it’s (necessarily) revolutionary, just that it’s a building block.  Its open APIs allow it to be baked into all sorts of other services, and like other foundational technologies — say, PC operating systems, or Internet protocol — it’s evolving comparatively slowly.  Even MediaWiki, the software behind Wikipedia, hasn’t changed all that quickly.  Too much is built on top of it, both technologically and in users’ practices, to change it hastily.  So that’s one reason just to let it sit as is.  But by giving a little more breathing space for attribution — to let people more readily build on others’ ideas through retweeting — Twitter could help assure an even wider spectrum of use, even if its founders didn’t happen to think it could be used so comprehensively or seriously.