@dalias @Edent Your question presupposes a premise I don't agree with.

I think it's well established that a massive increase in aggressive scraping is causing serious infrastructure problems.

From the perspective of someone running a website that is buckling under load, blocking implausibly old user agent strings is extremely appealing. It's extremely cheap to implement, gets rid of a lot of bot traffic, and results in miniscule collateral blocking.

This is not like IE. Browsers automatically update frequently now, and this makes old ones stick out. Many people running bots do not regularly update their UA strings.

Regarding fingerprinting, this is a nuanced topic, but for the narrow purpose of "is this a browser or a scraper pretending to be one (badly)", no, I don't think it's a problem that it's possible to do that.

Also, if someone were actually implementing a browser whitelist, I would agree with you that they're assholes. There's a difference between blocking anything unrecognized and blocking things recognized as sus.