So why can't we just build our own infrastructure? It's not as though you need spectacular sums of money to run a search engine or a social media platform. You need a few decent engineers for sure and you probably won't get rich from it, but most of this is demonstrably doable even without clouds.
Well Parler tried to start their own social media and that didn't turn out to them. Three of the largest tech companies kicked them out. How do you compete when you aren't allowed to compete?
You pretty much have to deal with big tech to have an app and no social media site can really be big without it. (I know side loading exists on android).
Parler hosting on Amazon was an incredibly dumb decision though.
The problem with a website is your cookies get deleted every once in a while. This is is inconvenient since you have to re login. Also, push notifications for a regular website isn't great (pwas might be better?)
Cookies get deleted when you explicitly delete them (most users don't know how to do that), or when you've set them to expire (which can decades into the future), or when you break your own code (which should be never).
Cookies should be set to session not a date. Many sites like HN don't do that presumably for convenience. I assume most social media sites set a date though.
go work for few years for a facilities-based last mile and middle mile ISP that runs things at the OSI layer 1 and 2 level of the internet and then tell me if you think your statement is accurate.
I don't mean we should build a parallel internet, I mean we should build enough redundancy into the Internet's public services that they become effectively impossible to exert control over.
Barring a complete shut-down of the Internet, simply making these applications cheap enough to operate would make them virtually impossible to control or stamp out.
They've been trying to shut down BitTorrent almost two decades, without much failures. They nailed the TPB guys to a cross, but that did all of nothing to actually shut bit-torrent itself down.
