I believe that Twitter is 2 top-down edicts away from drastically improving their efficiency with fewer employees.
1) Dramatically reducing the size and scope of the speech rules they implement to something akin to "If the speech is legal in a country, then it's legal on Twitter" would reduce technical and operating complexity. Between developing the technical infrastructure to automate restrictions and bans, to having staff trying to judge all sorts of petty bickering between groups trying to take advantage of the rules to report and ban their rivals, I feel like Twitter is wasting a large amount of resources on stuff that they shouldn't even be trying to do. A smaller team could focus on the essentials: removing illegal content and responding quickly to actual issues of safety (stuff like death threats that might pose an imminent harm). Hell, I feel like Twitter is so involved in trying to get involved in Internet drama and disputes that they currently do a mediocre job on actually dealing with genuine safety issues.
2) I realize that the infrastructure to run a site as large as Twitter isn't trivial to maintain, but for a site that has so few real user-facing features, it's kind of surprising that their technical team is so large. If somebody just laser-focused them on a clearly defined set of important features and policies for the business to execute on, it seems like Twitter should be able to move much faster on the handful of things that are core to their business.
There's your problem. You're feeling more than realizing. Twitter's most complex problems aren't infrastruture; it's adhering to various national laws on language while trying to keep hate speech off as well (they are doing a terrible job at this of course). But saying that they should just allow "free speech" is an extreme oversimplification and lacks an understanding of jurisdictions outside the US.
> Dramatically reducing the size and scope of the speech rules they implement to something akin to "If the speech is legal in a country, then it's legal on Twitter" would reduce technical and operating complexity. Between developing the technical infrastructure to automate restrictions and bans, to having staff trying to judge all sorts of petty bickering between groups trying to take advantage of the rules to report and ban their rivals, I feel like Twitter is wasting a large amount of resources on stuff that they shouldn't even be trying to do. A smaller team could focus on the essentials: removing illegal content and responding quickly to actual issues of safety (stuff like death threats that might pose an imminent harm). Hell, I feel like Twitter is so involved in trying to get involved in Internet drama and disputes that they currently do a mediocre job on actually dealing with genuine safety issues.
You say that Twitter can reduce its operational load by doing what exactly what it is already doing. What "Internet drama" is Twitter trying to get involved in?
> What "Internet drama" is Twitter trying to get involved in?
Twitter's rules are weaponized by people that hate each other. Whether it's lone individuals or people using Discord and other social media to organize raids, petty people love reporting their rivals on Twitter for anything and everything purely to harm their account. They can do this because they know that Twitter is trying to regulate speech beyond what is legally required so there's a chance that this tactic can work. Trying to regulate speech beyond legal requirements is an open invitation for this type of behavior.
Detoxifying platforms is a "new" responsibility that a lot of social media has discovered over the last decade and with good reason (again, not saying that Twitter is doing is well).
That responsibility has a few dimensions including financial (a platform that is overly toxic deters users and hence reduces ad revenue), moral (most corporations are a-moral by nature, but there are some attempts to assuage the public and deliver on a few safety features), and regulatory (hate speech is a real thing and can be enforced selectively (as opposed to "free speech" laws)).
Have you ever considered that trying to regulate speech to avoid "toxicity" creates a feedback loop that actually intensifies "toxicity"?
Online debate used to very intense and used plenty of naughty language and what would be considered cancellable slurs, but never felt deeply personal. The winner of the debate was the person who presented the best ideas in the clearest fashion. People of all stripes: whether Dirty Commies or Filthy Libertarians could go to an off-topic board, have pages long debates all together about abstruse topics like who in the hell should build the roads, and then go back to shooting the shit about basketball and computer games without blinking an eye.
People had the chance to freely vent. This is a good thing. It's bad when pressure builds up in a system when it cannot be released.
Today, Internet discourse takes a different form. Partially due to iDevices and social media making the Internet accessible to every low-IQ person out there who has very little business ever sharing any ideas with anybody, petty and unintelligent people debate by trying to bait the other side into rule-breaking so they can appeal to daddy-corporation to hit the other guy with the ban hammer. Every minor disagreement drips with pure resentment and intensity. Just one minor moment of weakness where somebody borderline-arguably breaks a rule becomes a gotcha moment where the other side can feign being outraged so they can try and get the other guy banned.
This modern safe space Internet is the real "toxicity" generator.
Hate speech is not a joke. It predates the Internet. It has led to (and will lead to) massive atrocities. Some of the most well understood atrocities of the 20th century (Rwandan genocide, Nazis, Anti-muslim attacks in India, many others) were engineered via hate speech.
Without regulation, Internet tools will be used to (and are being used to) conduct mass scale human political change including manufacturing consent for war, hate, colonialism, etc.
The world is big, and not all centered around the American take on politics (Commies and Liberatrians). It's also much much bigger than "daddy-corporation" and minor disagreements. The Internet is a tool that governments and political actors use for control.
I would very highly recommend reading up on Hate speech outside the American context of "free speech" and the Internet.
> The Internet is a tool that governments and political actors use for control.
You want to avoid atrocities and genocide. Great! So do I. Isn't it a great thing that we're both on the same page with that? We don't disagree on ends, but we do disagree on means.
The Internet is a tool used for control. We both agree with that. Now think about what you want. You want to protect people from an authoritarian regime, yet you also want to have systems in place for speech to be controlled. These ideas are in conflict. If speech is controllable, it's a tool that will be used by people who actually want genocide.
An atrocity isn't going to come because some edgy teens call each other a bundle of sticks while discussing political ideas online or use some distasteful language while trying to blast each other in some game.
But an atrocity can come if there's systems in place to restrict allowed speech. Do you think the Uyghurs are allowed to freely use the Internet and fully communicate their experiences with the outside world? Or are the systems that China has in place to regulate China's Internet used to control what information gets out and further their abuse?
This is the one semi-reasonable retort presented here. Let's say that Twitter does go to something like "all legal speech is allowed on Twitter" policy. What happens to their business?
Well, nobody can predict the future.
Maybe their business prospects instantly crash and burn.
But I think that Twitter is the unofficial brain of the planet and there's no alternative for all of the grifters out there. Celebrities, journalists, and corporations might talk a big game, but would they walk the walk? They would have to be willing to go without the network of attention they've built up.
The question might become, "Can Musk mentally withstand a few days/weeks of intense derision until something else hits the news cycle"?
Just for actual reference, here's how expenses broke down for '22 Q2 [1]:
Three Months Ended June 30, 2022 - In thousands
-----------------------------------------------
Cost of revenue - $540.676
Research and development - $454,859
Sales and marketing - $308,301
General and administrative - $216,586
---------------------------------------
Total costs and expenses - $1,520,422
So just running Twitter is 35%, R&D is another 30%, sales/marketing is 20%, while general/administrative is 14%.
If running Twitter includes a lot of server/infrastructure costs on top of headcount, and general/administrative includes a lot of rent and overhead, then maybe by totally cutting R&D and sales and marketing you could cut 75% of employees but Twitter would still run?
If you're buying Twitter to maximize value as soon as possible then reducing cost to the bone and maximizing earnings would make sense. It's how many private equity funds would maximize value to sell it at the earliest point.
Musk is different. He wants to maximize value by innovation. It's very risky. I would say that Marketing and Sales are probably the most important right now. R&D is next, since Musk wants to change its course. Reducing them to the bone will probably backfire.
Reducing head count is probably imperative since it reduces cost plus new hires will have a different mindset that's in line with what Musk wants to accomplish. He'll probably replace most top management ASAP. I suspect that we'll be hearing a lot of bad news from Twitter from people that get laid off and feel it was unfair.
He'll also try to fully automate (thru AI) as he tried with robotics in Tesla - but backfired. Full automation is hard there are just too many variables. He'll see that and reverse course. I expect there will be a lot of drama coming from the company but ultimately Twitter will change course. It will be interesting.
He's making sure he has total control over the company so what ever he decides to do will be done without opposition.
People that think he will ruin the company are wrong. It will take time but the investors will get a very good return on their investment.
> It will take time but the investors will get a very good return on their investment.
I think you're underestimating how much money he's paying for Twitter. Just for comparison, Snap's current Market Cap is $12.5B. He's buying Twitter for $44B. Both companies make about the same annual revenue with Twitter losing less money but still losing money.
I seriously doubt that 75% of twitter employees even use twitter, let alone for posting "mean things" about Elon.
Mostly given that twitter "only" has 40mil mDAU in the US[0]. Of course twitter employees probably have higher rates of twitter usage than an average person in the US. But I still find it unlikely (but not impossible) to be anywhere even close to 75% for usage, and, out of those, much less for posting about Musk.
The thought that a company like twitter, that has about 5 features anybody cares about (and pretty much had them for years) spends ONE BILLION PER QUARTER on "R&D" and operations boggles my mind. I mean, 250 people or something, I would understand (but only if there is no major outsourcing involved). But THOUSANDS of developers? How? Why?
There certainly was some effort involved to make Twitter performant if someone with 100m followers posted a tweet. But that seems to have been solved for quite some while and since the number of active users is quite stagnant, I don't see any order of magnitude changes coming which would require a lot of "R&D" on that end.
You're right, it's a lot. But you're also missing something important (like most people here who compare it to products like Cragislist). You're thinking about it as a service that needs to be run. But Twitter is not an open source project. Twitter is a public company where they need to make share holders happy. That means they HAVE to grow Revenue year over year. At that point you start hiring people just for that reason. Most employees were not hired to run the business. (for example, if Twitter wants to grow its current revenue by 20%, next year they need to roughly make $1B more than what they did this year).
BTW, under Musk the pressure to grow revenue is going to be even higher, despite his plans to take the company private. He has plans to triple revenue in like 3 years. Good luck to whoever stays there...
There are tax incentives for doing R&D so everything that could remotely fall in that category is classified as such. There are even consultants that specialize in turning course-of-business development into R&D
Even if you assume that somehow 50%ish of employees do literally nothing at Twitter, the people that remain will still have much more work shoveled on their plate for the same pay. Actions like these are a great way to completely kill a company because no one wants to jump to a ship that's currently on fire.
I would join Twitter after a 75% culling if the stock grants and pay were proportional with the risk taken. Culling 75% from sales and marketing departments is easy. It’ll be harder in engineering.
80% of twitter users are bots. probably half of those are controlled by orgs similar to CIA Mockingbird. i imagine this is elon trying to get twitter staff to revolt and cancel the deal?
Do they have revenue? When they make $5B/year with 40 people we can compare it to Twitter. Most people in tech companies are not there to run the core product.
It's a non profit, and they do sustain themselves with donations at this point. I thought we were comparing team to impact but I agree re: business functions.
Its Wikipedia entry says "By April 2000, there were nine employees working out of Newmark's San Francisco apartment" and says there were 50 employees in 2017.
In the chat logs of the recent lawsuit it seems he explains the reasons why. Compared to other social media companies Twitter gets in less income per employee. The idea is to make the ratio similar to other companies then make it public again.
So firing people and increasing income is the aim. I doubt 75% is accurate however and I think that Twitter blue is a bad idea for income.
The logs suggest video, music etc giving most revenue cut of any platform.
What I see here (especially with the recent Tesla worth 4 trillion statement) is boldfaced Elon going “not for my money.” He’s literally trying to coerce every cow in his gravitational neighborhood to pay for this too keep himself out of pocket. He doesn’t want it. It’s so fucking ridiculous.
Is there any objective way to define and measure productivity at companies like Twitter and estimate how few people they could get away with? The story implies they have 4/3×5600 employees which I could easily be convinced is a lot to do what they do. Likewise with Google and Facebook (who both give the impression of doing way more research and product development so you'd expect more people) - it's certainly plausible that the my could run their core business with way fewer people. Obviously the status quo will push against any changes, it would be interesting to see some information on what a reasonable size is
I don't think Mussk will keep running the status quo. Entire departments and long-standing projects will be axed, and sights set on one or more new major milestones.
Logically and practically speaking, the cuts will be focused on retaining employees perceived or known to be loyal to Master, and then backfill hiring folks who will work "really hard" without questioning too much or complaining.
The relative "openness" Twitter enjoys today will transition to secrecy and proprietary-ness, similar to the way Tesla and Space-X both operate.
just delete everything and charge a fee for keeping ones blog posts.
Do it in waves, oldest stuff first so that people can pay more often.
Make each wave more expensive so that holes appear if one doesn't pay. Newer stuff should be considered more valuable.
Make a desktop client that preserves the tweets for everyone who doesn't want to pay.
Repackage the company so that it makes an attractive purchase.
Make a Tesla Twitter auto and replace the twitter logo with a tiny automobile.
Show videos of spaceX launches and Musk Today at the top of every page. That way you can load the content for each page and only have to pull in the specifics after the audience watched 3-5 minutes of it.
Plenty of time for the slow swarm of desktop clients to find and serve the content.
Make an elaborately advertised but minimal effort to oppress alternative clients and have happy little accidents in the protocol so that undesired content is "accidentally" distributed unless it is top priority not to.
Purchase and merge with Reddit, eventually buy Facebook from meta. Call it TwiReFa.
Dedicated hardware for the desktop client.
Offer a free Tesla auto mesh network internet where internet refers to TwiReFa access only.
Give Tesla's their own TwiReFa account so that they can talk with other autos to negotiate deals.
Elon Musk (is yet again) seeking new ways to withdraw from the deal, this time by convincing share holders/employees that he might wreck even more havoc if the deal goes through.
Well, I could have been more clear: I imagine that Such headline could start mass exodus of Twitter employees looking for new jobs and leaving the company to avoid being fired by Musk if/when recession hits.
If people start leaving to a degree that interfere with Twitter operations, all while the case is still in court (even though Musk offered to dismiss it and settle the deal while Twitter shareholders were wary of it being another trick), I imagine that Musk could try and argue that while he was trying to close the deal, the company has deteriorated and thus he's eligible to withdraw.
This is just a hypothesis based on myself being cynical and obviously not quite familiar with what legal ground such ploy might have.
So you know that the US(?) economy is already in recession right now? While I suppose it's possible, you're saying it's confirmed, would you mind sharing what brought you to that conclusion?
It might be more a play to spook his co-investors into backing out of financing the deal, which then gives him an excuse not to go through with it (while also covering part of whatever penalty fee he can negotiate using whatever penalties his financiers would owe to him).
1) Dramatically reducing the size and scope of the speech rules they implement to something akin to "If the speech is legal in a country, then it's legal on Twitter" would reduce technical and operating complexity. Between developing the technical infrastructure to automate restrictions and bans, to having staff trying to judge all sorts of petty bickering between groups trying to take advantage of the rules to report and ban their rivals, I feel like Twitter is wasting a large amount of resources on stuff that they shouldn't even be trying to do. A smaller team could focus on the essentials: removing illegal content and responding quickly to actual issues of safety (stuff like death threats that might pose an imminent harm). Hell, I feel like Twitter is so involved in trying to get involved in Internet drama and disputes that they currently do a mediocre job on actually dealing with genuine safety issues.
2) I realize that the infrastructure to run a site as large as Twitter isn't trivial to maintain, but for a site that has so few real user-facing features, it's kind of surprising that their technical team is so large. If somebody just laser-focused them on a clearly defined set of important features and policies for the business to execute on, it seems like Twitter should be able to move much faster on the handful of things that are core to their business.