And say hello to being increasingly surrounded by Chesterton's fences[0] at every turn.
0: https://en.wikipedia.org/wiki/Wikipedia:Chesterton's_fence
"If the company is left only with 20% of engineers (even the best ones), it will be way deep in survival mode (just barely providing support for an existing product, not building some new exciting things or even improving existing ones)."
He also goes on to talk about how the company has deep pockets and that means they're not paralyzed, as you say, in the long term. Feel free to actually read the article for more details.
"Deep pockets" don't do anything on their own, especially not when no one believes in your company's ability to consistently fulfill its obligations to customers. You can hire engineers, sure, but how are you going to ramp them up when all of your engineering capacity is dedicated to keeping a small group of core services from falling over dead? How quickly can you bring people on board when you're notorious for a culture of overwork and abuse? You can borrow some money, sure, but who's going to lend to you when you're massively in debt from an LBO and you can't provide more than "long term" promises that in six months the company can actually function?
"How quickly can you bring people on board when you're notorious for a culture of overwork and abuse?". I agree that could be a challenge but then again, he's gotten people onboard at his other companies where I've heard the work/life balance is awful as well.
"who's going to lend to you when you're massively in debt from an LBO and you can't provide more than "long term" promises that in six months the company can actually function?". Valid point in today's economic climate, but there have been countless examples of investors making risky bets repeatedly. Uber is a perfect example.
Needless to say, employees will do one of two things. Even very mediocre employees will demand high-end consultancy fees, and "rock-stars" or whatever you want to call them will demand factors of that, or they'll just avoid you like the plague. But frankly, after a relatively short period of time even the high pay won't justify coming back anymore.
And then the time will come to hire and your competitor will be able to hire great people (with intimate knowledge of your company's inner workings), and you won't be able to hire anyone for double the pay. And that's IF you can avoid your company suffering regular sabotage when firing people. People do stupid things when they get fired. Seriously. It's scary.
I'm not one to defend Musk's management practices, but I think the main idea motivating this purge (which may or may not be a correct idea) is that Twitter, specifically, does not need "rockstar" engineers anymore at all.
Sure, he'll take them if they kiss the ring and work for peanuts, but he wants people who follow orders and don't talk back. He's not unique here: all management everywhere always wants this wants this but, in some endeavors you have to tolerate people who aren't this way, either because you don't know the right orders to give, or just because they're the only talent that can get done what you need to get done. Twitter's main corporate priority at this point is keeping the lights on as cheaply as possible, since they have a huge nut of debt they they can't afford to service, and on top of it all revenue is down. Again, I'm not saying this situation is desirable for Twitter employees, users, management, advertisers, or ownership (or even Musk himself), but this is the situation.
https://twitter.com/realGeorgeHotz/status/159527086740295680...
So clearly mr. Musk thinks twitter needs rockstar engineers. I also find it extremely unlikely he paid anything less than a full year's wage for these 12 weeks.
Startup vs mature service codebases can be quite different in conplexity... getting engineers up to speed and ready to contribute on the latter can take much more time.
Using Erlang.
Tiwtter uses Scala that should be much more scalabale, considering the staggering amount of money spent on the JVM compared to the BEAM.
Of course not the same, but the question is, how much harder?
10 times harder?
100 times harder?
Given 1/5 of the users and probably less than 1/200 of the traffic (*) it should be feasible with less than 7,500 employees.
(*) 100 billion messages sent via WhatsApp every day VS 500 million tweets per day
Also, graph algorithm’s likely don’t scale linearly, 10 times more edges may have much much more resource usage.
So you're saying WhatsApp had no critical other teams, but Facebook bought them anyway for a record breaking sum of money at the time, because they looked nice on picture?
When Elon tweets, I'm guessing Twitter doesn't go do 118M write transactions to all his followers. But what it would have to do is flag all 118M accounts as "something as changed". Then the next time one of his followers is on Twitter, which could be a month from now, that flag means "uh oh, better go see what changed", and to do that, all of the follower's "following" accounts have to be accessed to find the change, if there is only 1 change flag per Twitter account.
If Twitter keeps a change flag per "Following" user, that makes finding changes easier, but also means lots more flags to keep updated.
Or, you could go "stateless" and not use flags, but then it seems like you'd be doing a lot of work for users who may not see it for days or months.
Way different problem than 1-on-1 messaging.
if my mom sends a message to the family group it goes to 50 people.
How many billion WhatsApp groups are there?
with the same constraints and problems Twitter has to face (offline users, notifications, keep users messages until they are online again, etc.)
Also, Twitter does real-time data analysis, which is another of their money sources, and takes a huge amount of computing.
It's probably because many Twitter former employees lurk here on HN and can't understand why nobody is rooting for their beloved platform, of course shame on EM for his abusive behaviour, people before money, always, but nobody will miss Twitter if it really dies.
Half of what Twitter does is superfluous and the rest is badly executed.
A normal company would not survive with only 20% of the staff, Twitter might thrive, because honestly, at its core, as many have already pointed out, Twitter is “writing on a bathroom wall”.
Labour productivity is already a per worker measure. I've never heard it used to refer to anything else.
That seems to be what they're referring to when they say they can increase productivity by cutting workers and working the rest much harder.
I'm pretty sure that's bullshit and the sheer amount just daily maintenance and fire fighting will annihilate any possible productivity gains, so I agree with you.
There are plenty of industries where the economic envelope favors cheap human labor. But this in no way implies that you can cut 80% of your staff and get the remaining 20% to do both their own work and everyone else’s.
employees that would do this is unlikely to remain at the company to actually do this. Existing corporate culture or management friction will very quickly force out such an engineer, or make such an engineer ineffective at producing such innovations.
Another example is Tencent, from social network to a mail client, then WeChat, also expanded to games. QQ which is the original social network Tencent was built on, stayed almost the same with or worse with 100x head-counts.
It is always a rare event, but when it happens, it is a 1000x win for a company. It is something worth betting on.
survival doesn't have to be meaningful, the meaning is to survive.
Survival is about not dying, Twitter don't need 7,500 employees to survive.
> forget mitigating large-scale security vulnerabilities(i.e. log4j),
that was simply a matter of upgrading a dependency.
and honestly a remote exploit for that has never been spotted in the wild.
... Across a global infrastructure. ... And finding out if, where you have been compromised. ...And fixing all those systems that was found to have been compromised. ...And handling media questions, PR related to the issue and your organization.
Just saying it was "simply a matter of upgrading a dependency." is so amazingly far from what such an event requires. Even if you wasn't actually affected in a technical sense.
5 people total?
that have too look at the problem for a few days and then go back to their regular duties.
Which is their job, BTW, you don't need "special teams".
I work for a company that has more than 20 thousand employees.
How we fixed it?
Every team checked if their dependencies contained a vulnerable version of log4j, fixed it, ran the tests, submitted the new code, the CI/CD did the rest.
absolutely not.
> How much do they handle media questions, PR related to global IT security events?
You don't need thousand people for that and It's their job anyway!
Listen, I come from what the average American would call "a socialist" country, at least regarding the labour market and organization.
I understand the social part of employing more people than necessary, it's a kind of redistribution that many countries won't or can't do through proper taxation.
I also understand that sometimes it's about resilience, two of everything is better than single points of failure.
But, it's ironic that these observations come from the POV of an American company.
It's ironic that people are leaving Twitter because Musk is asking them to work hard, meaning they aren't actually doing it, according to his standards (he's a terrible boss, no doubt!), but also meaning that Twitter is probably doing so bad because they focused on the wrong thing, like Facebook with the meta fiasco.
The American dream is not about 28 days of paid vacation, maternity leave, life-work balance, it's about making as much money as you can in the fastest way possible.
In its original formulation In such a country the talents, energy and perseverance of a person ... have far greater opportunity to display than in monarchies
You want a better life balance, better working hours, less staying in the office and more paid time to raise your kids?
Come to Europe!
For much less money of course.
Because you can't have both.
So yeah, it might be easier to do something with 3 people than one, but most of the time too many people are enemy of the efficiency.
You know what took the most part of upgrading the log4j incident where I work?
Coordination between too many levels of hierarchy.
That is actually the worst part of working in large companies, that's what actually stifle innovation and paralyzes them.
Twitter was already doing bad, it's been doing bad for years.
I left Twitter (as a user) 5 years ago, because it was already terrible.
I'm no Musk fan at all, but at least he's making things clear: if that's the end result people are used to, we don't need that many people to achieve it, we simply need motivated people (by money).
As SDE employee in Germany you might earn 60k-120k. Where the end of the range is more often found in larger companies with about 10y+ of experience.
But you have health care that never puts you in a position where you have to decide between ruin or death of a family member, what can be done will be done even e.g. if you need brain cancer be burned out with a particle accelerator (no joke). You will never be in a position where you instantly are ruined when you loose your job (1y at 70% income and at least social care later). You are forced to have an extend of retirement solution/insurance, which costs, but at least you cannot mess it up or loose in when you are insolvent.
So this stuff I call the true freedom, not gun ownership or freedom of speech that goes to far. Being free of crushing fear about your family health and future risks, this is freedom in Germany and other European countries. - This freedom also reduces violence and crime you are confronted with a lot, because the motivation or psychic conditions for it are much less common.
So for me Germany is the perfect place for raising a family imo. But maybe yes there might be a reduced chance to become really rich here as a SDE
What? Citation needed.
This spread through Minecraft servers like wildfire. The damage was mostly kids crashing the game on other kids, but that's still a denial of service.
Log4Shell definitely had "in the wild" exploitation. A trivial amount of googling gives evidence.
In most organization management courses, you are taught " you cannot change the culture". The truth is you can and you do it by firing people, firing or pushing out a lot of people.
My point is that Twitter is a bad case study for engineers, its a very standard case for mba though. You cannot and should not view Twitter thru a primarily engineering point of view.
Forget about new products, forget about improvements to existing products, forget infrastructure work/cost optimization, forget not having major data breaches regularly, forget mitigating large-scale security vulnerabilities(i.e. log4j), forget being able to capitalize on market trends on a useful timescale, and forget about the ability to deal with situations that require more engineering capacity that your company uses on an average day.
Entering "survival mode" and keeping only enough engineering capacity on board to keep the service running on a normal day is implicitly a bet that no emergent events are going to demand more capacity from you than is needed on that normal day. The product might be standing now, but you can't do anything useful with a paralyzed company that's going to topple over the first time it faces a stiff breeze.