Another aspect is the social graph. It’s targeted for normies to easily switch to.

Very few people want to install a communication app, open the compose screen for the first time, and be met by an empty list of who they can communicate with.

https://signal.org/blog/private-contact-discovery/

By using phone numbers, you can message your friends without needing to have them all register usernames and tell them to you. It also means Signal doesn’t need to keep a copy of your contact list on their servers, everyone has their local contact list.

This means private messages for loads of people, their goal.

Hey, we know this account sent this message and you have to give us everything you have about this account

It’s a bit backwards, since your account is your phone number, the agency would be asking “give us everything you have from this number”. They’ve already IDed you at that point.

To me I’d consider Linux not standardized since anything outside the kernel can be swapped out. Want a GUI? There are competing standards, X vs Wayland, with multiple implementations with different feature sets. Want audio? There’s ALSA or OSS, then on top of those there is pulse audio, or jack, or pipewire. Multiple desktop environments, which don’t just change the look and feel but also how apps need to be written. Heck there are even multiple C/POSIX libraries that can be used.

It certainly can be a strength for flexibility, and distros attempt to create a stable and reliable setup of one set of systems.

The only problem I run into is sites that use Bluetooth or USB APIs to talk to a local device. Both Firefox and Safari don’t implement them due to security concerns.

I’ve dabbled in a couple recently, xTTS v2 in Coqui sounds pretty good and is pretty quick. Especially if you use one of the built in voices. Has a weird “only for non commercial use” licence if that bothers you.