I used to, but that’s not enough and that’s only for me anyway. To few people are doing it, so you can be easily identified because you do stuff like this. (See the TOR browser approach and their recommendations about changing the defaults) Many sites don’t like it so most people aren’t willing to do that or rather don’t care or know how to.

By Bowdlerizing/randomizing the data for each call (or site/session probably) instead, the data will still be ‘valid’, but not as traceable and would negate the need for such hacks that almost noone does and which also makes you stand out like a sore thumb.

It probably isn’t possible or at least not easy on Windows/Apple, so it would be a Linux only thing. Which is a problem too and opens up to Linux blocking or subtler “upgrade your browser to…” type errors.

There are certainly issues and problems with this I’ve not considered or mentioned. It is not a popular idea among the tech giants for one, and they’d do whatever they can to nip something like this in the bud.