If your experience with services not under active development is that they are "breaking all the time," I'm sorry to say that you've worked only at places where the quality bar is very low.
Services that aren't under active development break as the things above them change and they are put into environments they weren't in before. This is very normal and the reason you don't see problems happening because of this is either because you're moving slowly (which is a valid way to do things…but not typical) or your have an army of people keeping it all running smoothly.
Alternative take: I was at Google when they killed Reader. At the time they pulled the plug, Reader had already been destaffed for years and was kept going in one guy's 20% time. The infrastructure effectively ran itself - the underlying Google datacenter teams replaced machines when they died, everything else was automated.
My experience there was pretty consistent with that: if you leave things alone, they won't break. Virtually all breakage was caused by changing things. On a well automated system repairing the physical components, rolling OS updates from time to time and so on, is not a huge amount of work.
The idea that Twitter is ever "not under active development" is ludicrous.
Services with hundreds of millions of users, antagonistic spammers/hackers/countries, and advertisers spending millions are never in maintenance mode.
You sound like you're trying to apply your understanding of lifestyle software businesses to one of the largest real-time communication networks in human history.
I don't believe I claimed Twitter was not under active development. The point that I was making here is that if we were OK with Twitter as-is, it would be possible to maintain it with far fewer people than they employ now. Indeed, 'maintenance' will be defined differently for an indie hacker vs. a popular website with user-generated content. But all software has some split of "keeping the lights on" vs. innovating.
What is your estimate of Twitter's effort split between those two categories?
Exactly. Any one who has ever worked in SaaS industry realizes how many cogs in a wheel it takes to keep it running. Sure, there are always inefficiencies and redundancies. But, I highly doubt Elon figured out in a week that 50% of employees are not necessary. This looks like the vindictive and impulsive action of a man baby who has never been told that he doesn't always know what's best. Time will surely tell. All the best to those who were laid off, but most importantly, good luck to those who are staying back. They are in for a ride.
No one is contesting the merits of infra at scale. There are essential teams that would likely be untouched because their job is operationally critical. However, other teams for less vital jobs would be cut. I think you are making an assumption that the core team who manages the infra is being laid off which is not clear.
