This is especially a problem for a site like Twitter, which can have unexpected spikes in user traffic and interest at random, Krueger says. Krueger compares Twitter to online retail sites, where companies can prepare for big traffic events like Black Friday with some predictability. “When it comes to Twitter, they have the possibility of having Black Friday on any given day and at any time of the day,” Krueger says. “On any given day, there can be some news events that can have a huge impact on the conversation.” This is hard to do when up to 80% of your SREs are laid off — a number that Krueger says has been manipulated within the industry but that MIT Technology Review has not been able to confirm. The engineer agreed that the ratio seemed “reasonable.”
The current Twitter engineer doesn’t see a way out of the problem – other than reversing the layoffs (which the company has done He reportedly already tried to back off somewhat.) “If we were to push at an accelerated pace, things would break,” he says. “There is no way around that. We are accumulating technical debt much faster than before – almost as fast as financial debt.”
The list goes on
It presents a dystopian future where problems pile up as the backlog of maintenance tasks and repairs gets longer and longer. “Things will break. Things will be broken more often. Things will be broken for longer periods of time. Things will break in harsher ways,” he says. “Everything will pile up until the end, it is unusable.”
The engineer says Twitter’s collapse into unusable wreckage has been some time, but clear signs of process rot are already there. It starts with the little things: “The bugs are in whatever part of any client they use; whatever back-end service they are trying to use,” the engineer says. “It’s going to be a small inconvenience at first, but as backend repairs get delayed, things will pile up until people eventually give up.”
Krueger says Twitter won’t blink from life, but we’ll start seeing more tweets that don’t load, and accounts popping up and going out of existence seemingly on a whim. “I would expect anything that writes data to the backend to be slow, timeout, and more subtle types of failures,” Krueger says. “But they are often more cunning. They also generally put in more effort to track down and solve the problem. If you don’t have enough engineers, that is going to be a big problem.”
Manual retweeting of judgment and declining follower count are indications that this is already happening. Twitter engineers have designed security tools that the platform can fall back on so that functionality doesn’t stop working entirely, but instead provides shortened versions — and that’s what we’re seeing, says Krueger.
Besides minor crashes, the Twitter engineer also believes there will be major outages on the horizon, thanks in part to Musk’s cost-cutting drive to reduce Twitter’s cloud computing server load in an effort to recover up to $3 million per day in infrastructure costs. Reuters reports This project, which came from Musk’s war room, is called the “Deep Cuts Plan.” One Reuters source described the plans as “fake”, while University of Surrey cybersecurity professor Alan Woodward said that “unless they significantly over-engineer the existing system, the risk of poor capabilities and availability appears to be a logical consequence”.
Meanwhile, when things go wrong, there is no longer any internal institutional knowledge to quickly fix issues as they arise. “A lot of the people I’ve seen leave after Friday have been there nine, 10 or 11 years, which is absurd for a tech company,” says the Twitter engineer. As these individuals moved out of Twitter’s offices, decades of knowledge about how its systems worked vanished with them. (Those within Twitter, watching from the sidelines, have previously argued that the Twitter knowledge base is Overly focused in the minds of a handful Of the programmers, some of them were fired.)