This is especially problematic, Krueger says, for a site like Twitter, which can have unexpected spikes in traffic and user interest at random. Krueger compares Twitter to online retail sites, where businesses can prepare for big traffic events like Black Friday with some predictability. “When it comes to Twitter, they have the ability to have a Black Friday any day at any time of the day,” says Krueger. “On any given day, some news can happen that can have a significant impact on the conversation.” That’s harder to do when you’re laying off as many as 80% of your SREs, a number that Krueger says has been dismissed in the industry, but which MIT Technology Review has been unable to confirm. The engineer agreed that the percentage sounded “plausible.”
Twitter’s current engineer sees no way out of the problem other than reversing the layoffs (which the company has reportedly already tried to back out kind of.) “If we’re going to move at breakneck speed, things are going to break,” he says. “There’s no way around this. We’re accumulating technical debt much faster than before, almost as fast as we’re accumulating financial debt.”
The list goes on
It presents a dystopian future where problems pile up as the backlog of maintenance and fixes grows ever larger. “Things will break. Things will break more often. Things will break for longer periods of time. Things will break in more severe ways,” he says. “Everything will compound until eventually it can’t be used.”
Twitter’s collapse into an unusable wreck is a free time, the engineer says, but the telltale signs of process rot are already there. It starts with the little things: “Errors in whatever part of the client they’re using; whatever backend service they’re trying to use,” says the engineer. “It’s going to be small annoyances to begin with, but as backend solutions get delayed, things will pile up until people finally give up.”
Krueger says that Twitter won’t blink to life, but that we’ll start to see an increased number of tweets that don’t load and accounts appear and disappear seemingly on a whim. “I expect that anything that’s writing data to the backend has possibly much more subtle slowness, timeouts, and failure conditions,” Krueger says. “But they’re often more insidious. And they also generally require a lot more effort to track down and solve. If you don’t have enough engineers, that’s going to be a big problem.”
The stuttering manual retweets and faltering follower count are signs that this is already happening. Twitter engineers have designed security systems that the platform can fall back on so that the functionality doesn’t go completely offline, but instead provides scaled-down versions, which is what we’re seeing, Krueger says.
In addition to the minor glitches, the Twitter engineer also believes there will be major disruptions on the horizon, thanks in part to Musk’s cost-cutting to reduce the load on Twitter’s cloud computing server as an attempt to recover up to $3 million a day in infrastructure costs. Reuters reports that the project, which emerged from Musk’s war room, is called the “Deep Cuts Plan.” One of Reuters’ sources called the plans “delusional”, while University of Surrey cyber security professor Alan Woodward says that “unless they have over-engineered the current system, the risk of reduced capacity and availability seems a logical conclusion.”
brain drain
Meanwhile, when things are going well, the institutional knowledge is no longer there to quickly fix problems as they arise. “A lot of the people I saw leaving after Friday have been there nine, 10, 11 years, which is ridiculous for a tech company,” says the Twitter engineer. When these people left Twitter’s offices, decades of knowledge about how its systems worked disappeared with them. (Those at Twitter, and those watching from the sidelines, have previously argued that Twitter’s knowledge base is too concentrated in the minds of a handful of programmers, some of whom have been fired.)
