The article has been super popular and it seems using old machines for a very long time is a quite common practice but one that people also feel insecure about. A cursory reading of comments across the web tells me that part of the attraction to the article (or at least to commenting on it) is the recognition or validation of other people doing the same.

Here's an interesting statistic from the web logs: 10% (~7k unique visitors) of all traffic to the article has used a machine with OS X 10.1 Puma. That is an OS from 2002. I believe that is the last generation of computers, but I might be mistaken. It seems to good to be true though but I can't really think of a reason why this number would be wrong..

Massive use of user agent switching should lead to way more even distribution of other uncommon OSes as well? Does anyone know what could be up?


@rra I heavily doubt there are that many #PowerPC's floating about. The stats seem sketch.


@resynth1943 yeah I'm also doubtful that would be the case. What could be the cause of these statistics though?

Β· Β· Tusky Β· 1 Β· 0 Β· 0

@rra User Agent randomisation extensions would be my first guess.

@rra (they actually worsen fingerprinting, but that's something for a later post.)

@resynth1943 yeah that's what I considered first as well. In that case one would see a more even distribution between different niche/old OS'es but that is not the case. It's quite heavily skewed. Then again it could be that user agent randomizers are not that random :)

Sign in to participate in the conversation

Welcome to, an instance for discussions around cultural freedom, experimental, new media art, net and computational culture, and things like that.