Hacker News Reader

Comment by danans | Hacker News Reader

> So, he's destroying all of the institutional knowledge, the brand image, and the existing business relationships, while pivoting to a radically different market.

Sounds about right. Also, I don't think he values the institutional knowledge much, and thinks he can replace it quickly, because he thinks Twitter employees were idiots and he makes reusable space rockets something something.

Now he has to find a way out of his predicament. First step, fire a lot of people to stop the monetary bleeding he caused. Second step, find others to blame - activists organizing boycotts - or have his buddy the founder take public blame for him. Third, bring in his "A-team" of super rocket EV tunnel software engineers to fix it all up to work better with 1/5th the people. I don't think he cares if he has to do institutional and human damage to achieve it.

View on HN

I'd like to hear your thoughts on what kinds of unique institutional knowledge you think they've built up and what sets the Twitter employees apart from their peers at other companies. My main guess would be scaling caches and services, but even that isn't a super rare these days compared to say 2006. I'm not sure what about Twitter specifically, the app and its functions, is that complex to pass on. Coming into codebases that large is very challenging, but it's doable. Isn't the current calculus ~ lowering headcount to stay solvent vs. the risk of more lost knowledge and the other 50% of the company losing their jobs?

Well, for one, Twitter is under a 20 year consent decree with the FTC and all of the relevant leaders in charge of this resigned this week and individual twitter engineers now need to "self certify" that they're complying with the decree.

People who have intimate experience with just exactly the degree of latitude the FTC will allow for twitter and have a personal relationship that allows them to negotiate with the FTC isn't exactly a skill you can just find on the street.

For details, there are at least 2 issues with the FTC.

in 2011 there was a settlement about Failure to Safeguard Personal Information [1].

"Under the terms of the settlement, Twitter will be barred for 20 years from misleading consumers about the extent to which it protects the security, privacy, and confidentiality of nonpublic consumer information, including the measures it takes to prevent unauthorized access to nonpublic information and honor the privacy choices made by consumers.

Then in 2022 FTC charged Twitter for using account security data for targeted advertising. They have a $150M fine and some other conditions [2].

[1] https://www.ftc.gov/news-events/news/press-releases/2011/03/...

[2] https://www.ftc.gov/news-events/news/press-releases/2022/05/...

> what kinds of unique institutional knowledge you think they've built up and what sets the Twitter employees apart from their peers at other companies

Literally knowing how twitter codebase works. Knowing about existing issues and workarounds for them. You wont find this knowledge in other companies.

> Coming into codebases that large is very challenging, but it's doable.

Sure, but it takes awful lot of time if there is no one around to explain you stuff. By awful time, I mean really a lot of additional time.

I have followed Tesla closely over the years and this is par for the course of how he operates(similar things have happened at SpaceX but I know more Tesla specifics).

The issue is that in the case of Tesla, automakers have tried a lot of stupid things Musk have done and failed beforehand. Musk ignores those prior learnings and repeats all the mistakes that incumbents made decades ago.

What has saved his butt time and time again is the fact that he has been able to conjure up enough financial runway to re-learn the mistakes and recover and the fact that sometimes mistakes made by others in the past are no longer mistakes given today's environment and end up becoming competitive advantages. So he wins some and he loses some.

One example of him losing is when he attempted to automate all the assembly in the factory in 2017. He messed up and hired the wrong people to design the body of the Model 3 and it ended up too expensive to manufacture. In response he attempted to use robots in assembly to cut costs. The technology for what he wanted to do was still not ready just like it was not ready in the 70s when GM tried to do the same thing. Robots are Blind, one armed, and stupid. They cannot handle finesse when dealing with a noisy real world.

On the other hand, GM and others were adamant to design their cooling systems with multiple loops because that was the way things were done and it was a reliable way to accomplish the task. As a result this thinking was ingrained in all the OEMs. They have cooling departments that defend their approaches and it causes stagnation. Musk's approach of repeating all the things that OEMs already did led to him discarding beliefs that no longer are the best way of doing things and led a much better and cheaper design using one cooling loop and adjusting between heat/cold using software. When every penny counts as it does in automotive this was a competitive advantage that now competitors are copying.

As long as he has enough runway I wouldn't count him out as he seems to running the same playbook as Tesla and SpaceX.

Reply on news.ycombinator.com