How does it solve the problem of dependencies without becoming bloated?

Holy shit why do I only get this crucial info now

But when is RCS provided as an API like text messages, so we can get third party apps?

How does this work against DNS blocks? I assume if you use your ISPs DNS server and they block that domain ECH won’t help

I like the point about LLMs interpolating data while humans extrapolate. I think that’s sums up a key difference in “learning”. It’s also an interesting point that we anthropomorphise ML models by using words such as learning or training, but I wonder if there are other better words to use. Fitting?

Which is also what the article is arguing. As soon as a majority of users are subject to this the websites can discriminate. If only safari does this it’s still bad but not very bad