Yes, that's more or less how it was done. The explanation is here: http://www.spinellis.gr/blog/20090805/index.html
I'd say that's all it took, a lot of automated tweeting with better guesses.
I would imagine posting a bunch of tweets at once with the same number, and deleting the ones which didn't match. That's assuming that deleted tweets completely disappear though.
Re:Self-referential URL
by mhielscher on Monday September 17, 2012 @12:00AM
(#4531284)
It involves multiple redirects. One goes to http://simonsarris.com/recurse, which surely can be modified at any time. So, goo.gl short link to a controlled resource that shortly afterward is made to point back to the tweet.
Re:Self-referential URL
by damncabbage on Monday September 17, 2012 @12:29AM
(#4531332)
The post I was replying to wasn't the goo.gl one; it was the @selfrefer tweet link, which has a link to the tweet at twitter.com in a tweet: http://twitter.com/SelfRefer/status/3128391843
Did it involve basically predicting what the number was likely to be by observing the current posting rate, tweeting a bunch of times, and then stopping when one of the tweets matched? Or am I missing something?