A while back I wondered about how "elderly tweets" would fare in discovery in an action. I was concerned that it might be difficult to exact the intended meaning from the brief blurt of yesterday. This concern was premised on the fact that no one (at least no one I know) deletes tweets, and so they stay lodged on Twitter's server.
A recent article on ReadWriteWeb, "10 Ways to Archive Your Tweets," by Sarah Perez, refers to the Twitter API Wiki's Things Every Developer Should Know, where we learn that, indeed, "Twitter still maintains a database of all the tweets sent by a user." But searching through those tweets is not possible:
We also restrict the size of the search index by placing a date limit on the updates we allow you to search. This limit is currently around 1.5 weeks but is dynamic and subject to shrink as the number of tweets per day continues to grow.
So… tweets recede very quickly into the past, shrinking almost to invisibility in a little over a week. This is a problem for some people — as Perez points out, it makes data-mining impossible, unless one is able to solicit the assistance of the Twitter IT people. And, I suppose, it makes tweets an unlikely object of a discovery order.
If, for some reason, you are loath to let your tweets dwindle into insignificance, you have some choices about how to archive them in a way that would make them truly accessible — the point of the ReadWriteWeb article. The simplest of these methods, so far as I'm concerned, would be to subscribe to your Twitter stream's RSS feed in Google Reader, which uses its gigantic storage capacity to hang on to everything it gets its hands on; and, of course, it lets you search all past feed items.
If this appeals to you, take a look at ReadWriteWeb's follow-up post, "How To: Backup And Search All Your Friends' Tweets In Google Reader," which will take you through the process step by step.