David Schaub retweeted
Iβve survived Perl, SOAP, XML, the microservices hype cycle, and four distinct eras of JavaScript frameworks. You cannot intimidate me with a new build tool.
Iβm not sure people understand the context of this product.
There is a huge array of aftermarket kits to give original ipods all of these features and more. They require skills to install and cost at least double the 250$ for this.
Fully modded ipods hit 1k+$ on ebay.
David Schaub retweeted
Reading all 1030 of Fujitsu patents to complete my analysis of Bubble System, a project I've been working on for six years, and collecting dozens of obscure patents to determine the logic cell topology of the Konami Firebeat GCU (350nm) would have been impossible without LLMs.
despite how how often people spend a lot of money on random stuff, money remains the most underused lever for positively changing peopleβs own lives
I've been watching a lot of slop and it hurts my heart a little realizing that video has fallen. This time next year 99% of the new content on youtube will be slop. I plan to profit from it, but deep in the cockles of my heart it's a little sad.
Video will eventually be so good that it will be unwatchable. Every video you see will be the best thing you've ever watched. You will fall into a trancelike state as the subliminals penetrate your mind.
The shared experience of a big movie or show coming out and everybody watching it at the same time will become increasingly atomized. We'll lose context of eachother. How am I supposed to explain Chinese Emily singing Paula Abdul with Roger from American Dad in a cat costume? If you weren't in the microniche that created it, there's no way it can make sense.
I think it will be for the best, but not everybody will adapt. It will feel like a flood.
Learn to swim.
Investing in everyone youβre fucking is not a strategy that works. Thatβs why he went bankrupt.
BREAKING: If Sam Bankman-Fried's FTX hadn't gone bankrupt, its portfolio would be worth approximately $206 billion today.
That includes a 1,500,000% gain on Cursor and 34,000% on Anthropic, making him one of the best investors of the decade.
I wish the day was here where 3d printers were fully autonomous thing dispensers. But so far even the bambu h2d needs constant support.
But it does seem to be the best option and is very repairable. At least it doesnβt need glue sticks.
David Schaub retweeted
TLDR of the job market rn:
No one bets on anyone. Everyone wants big-name people from flashy startups, but when it comes to early-stage experience, suddenly youβre βtoo risky.β They want the track record without taking the risk that creates it.
David Schaub retweeted
i owe almost all success i've had in my life to one video game: Ultima Online (1998 era).
UO existed at a unique time period in gaming where games were made by nerds, not MBA types.
there were no instructions, no youtube, no influencers, no guides.
it was difficult to advance and it was unforgiving. a misstep could set you back days, weeks, or months worth of effort.
there was no pay to win. the dopamine hits were more consistent with reality.
it taught me about human psychology, markets, fiat, inflation, economics, strategy, exploitation, entrepreneurship, efficiency, programming, how to communicate/type fast, how to use things in ways they weren't intended to be used to gain an edge... the list goes on.
it makes me sad that this style of game is mostly lost for kids today.
David Schaub retweeted
I remember when you would install Anaconda and it would fuck up your whole laptop
I instinctively didnβt install X money because it seemed like it would lead to some weird hack.
I was right, but at what cost?
David Schaub retweeted
Serious answer: yes. Before joining Microsoft, I knew maybe five people who used Copilot. I lived in a San Francisco tech bubble, and all of us worked at companies with fewer than 5,000 employees.
Microsoft 365 Copilot now has over 30 million paid seats and is growing fast. It's also a significant market penetration if you look at the number of seats / knowledge workers in Enterprises
Working at MSFT has made me understand why. Large companies need AI that works with their existing data and meets their privacy and compliance requirements, often across multiple countries. Add governments and regulated industries, and it gets even more complex. Copilot can handle that.
So people use Copilot for real work, and yes, most people love it. I use it every day now: Copilot CLI for coding (with access to models from Gemini, OpenAI, xAI and Anthropic!) + Copilot Cowork for regular chat Etc. Iβve been happy with it.
I think the perception gap on this website comes from the San Francisco tech bubble. If youβre a tech bro like me, you mostly know people at smaller companies. You donβt see much of the enterprise world where Copilot is used.
David Schaub retweeted
It was the worst mistake of my coding career.
The boss called me into the office. He said "Hillel, did you commit this code?"
It was my first job after college. I knew all the patterns, the big O notation, the UMLs.
I knew my SOLID and my clean code. I knew I wrote good software
The boss was angry. He said "take a close look at what you did".
And I saw my mistake. I had forgotten to ask for code review, which means nobody noticed these lines of code:
```
user = UserFactory.createUser();
user.add_to_group(group);
πππ
ππ€πππ€ππππ»π
πππ³π
±π€ππππππππ
ππ²
πππππ»ππΏππππ»π
π»π
ππππππππππ€ππππππ§ππ
```
I had accidentally committed the entire Egyptian Book of the Dead!
We tried to roll it back, but it was too late. Anubis had risen from the underworld to weigh our codebase against his feather.
Our codebase was too sinful (no GDPR-compliance) and so the crocodile goddess Ammut devoured the souls of all 1,000 of our production servers.
I paid dearly for this mistake. I had to buy the devops team donuts for a whole month.
And those donuts taught me a lesson I would never forget: always get a code review before committing, even if you use SOLID and clean code.
Because the real world is not school.
In the real world, Anubis is real.