For the best experience on desktop, install the Chrome extension to track your reading on news.ycombinator.com
Hacker Newsnew | past | comments | ask | show | jobs | submit | history | sxyuan's favoritesregister

> Amusingly if you read chapter 1 of Founders at Work quite a big part of what worked at PayPal may have been firing Elon Musk. Max Levchin largely built PayPal tech wise using Unix and then it was merged with Musk's X.com and Musk became CEO and wanted to switch everything to Windows.

I encourage you to read "Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future." Musk talked about PayPal and the Windows switch in detail in Appendix 2:

“As for the technology change, that’s not really well understood. On the face of it, it doesn’t sound like it makes much sense for us to be writing our front-end code in Microsoft C++ instead of Linux. But the reason is that the programming tools for Microsoft and a PC are actually extremely powerful. They’re developed for the gaming industry. I mean, this is going to sound like heresy in a sort of Silicon Valley context, but you can program faster, you can get functionality faster in the PC C++ world. All of the games for the Xbox are written in Microsoft C++. The same goes for games on the PC. They’re incredibly sophisticated, hard things to do, and these great tools have been developed thanks to the gaming industry. There were more smart programmers in the gaming industry than anywhere else. I’m not sure the general public understands this. It was also 2000, and there were not the huge software libraries for Linux that you would find today. Microsoft had huge support libraries. So you could get a DLL that could do anything, but you couldn’t get—you couldn’t get Linux libraries that could do anything.

“Two of the guys that left PayPal went off to Blizzard and helped created World of Warcraft. When you look at the complexity of something like that living on PCs and Microsoft C++, it’s pretty incredible. It blows away any website.

“In retrospect, I should have delayed the brand transition, and I should have spent a lot more time with Max getting him comfortable on the technology. I mean, it was a little difficult because like the Linux system Max had created was called Max Code. So Max has had quite a strong affinity for Max Code. This was a bunch of libraries that Max and his friends had done. But it just made it quite hard to develop new features. And if you look at PayPal today, I mean, part of the reason they haven’t developed any new features is because it’s quite difficult to maintain the old system.

“Ultimately, I didn’t disagree with the board’s decision in the PayPal case, in the sense that with the information that the board had I would have made maybe the same decision. I probably would have, whereas in the case of Zip2 I would not have. I thought they just simply made a terrible decision based on information they had. I don’t think the X.com board made a terrible decision based on the information they had. But it did make me want to be careful about who invested in my companies in the future.

“I’ve thought about trying to get PayPal back. I’ve just been too strung out with other things. Almost no one understands how PayPal actually worked or why it took off when other payment systems before and after it didn’t. Most of the people at PayPal don’t understand this. The reason it worked was because the cost of transactions in PayPal was lower than any other system. And the reason the cost of transactions was lower is because we were able to do an increasing percentage of our transactions as ACH, or automated clearinghouse, electronic transactions, and most importantly, internal transactions. Internal transactions were essentially fraud-free and cost us nothing. An ACH transaction costs, I don’t know, like twenty cents or something. But it was slow, so that was the bad thing. It’s dependent on the bank’s batch processing time. And then the credit card transaction was fast, but expensive in terms of the credit card processing fees and very prone to fraud. That’s the problem Square is having now.

“Square is doing the wrong version of PayPal. The critical thing is to achieve internal transactions. ...


Lots of things. Neither Medicare (1965) nor Medicaid (1965), which provide healthcare to millions of impoverished, disabled, and elderly people, didn't exist in 1950. Nor did NASA (1958) and the US space program. The NSF (1950) which funds most non-biomedical research in the US, was founded in that year, and while the NIH technically existed in 1950, it was mostly just the research institute in Maryland, not the funding agency it also comprises today. DARPA (1957), the defense research and funding agency that among other things created the ancestor of the very network you are using to post here, also didn't exist. You could go on and on.

we've been tracking the deepseek threads extensively in LS. related reads:

- i consider the deepseek v3 paper required preread https://github.com/deepseek-ai/DeepSeek-V3

- R1 + Sonnet > R1 or O1 or R1+R1 or O1+Sonnet or any other combo https://aider.chat/2025/01/24/r1-sonnet.html

- independent repros: 1) https://hkust-nlp.notion.site/simplerl-reason 2) https://buttondown.com/ainews/archive/ainews-tinyzero-reprod... 3) https://x.com/ClementDelangue/status/1883154611348910181

- R1 distillations are going to hit us every few days - because it's ridiculously easy (<$400, <48hrs) to improve any base model with these chains of thought eg with Sky-T1 recipe (writeup https://buttondown.com/ainews/archive/ainews-bespoke-stratos... , 23min interview w team https://www.youtube.com/watch?v=jrf76uNs77k)

i probably have more resources but dont want to spam - seek out the latent space discord if you want the full stream i pulled these notes from


Capillary Action is my favorite Wikipedia article: https://en.wikipedia.org/wiki/Capillary_action

Leonardo da Vinci observed Capillary Action making him remark that the paint brush was a gift from the gods. He rightfully observed paint brush has an unnatural, given our understanding at the time, attraction for paint. If you place the brush in a cup of paint, and come back some time later, the brush will have _pulled_ the paint into itself noticeably above the surface of the paint in the cup.

It wasn't until the early 1800s that we had the early versions of the Young-Laplace equation which attempted to quantitatively explain the phenomenon.

Fast forward to 1900, Albert Einstein's first paper was still exploring Leonardo's paint brush: https://web.archive.org/web/20171025203011/http://gallica.bn...

The wikipedia entry reminds me that this world is full of magic and wonder, everywhere you look. Something as simple as a paintbrush can reveal something peculiar that takes generations of study by the brightest of human minds to try and explain.

Go outside and turn over a rock, any rock, and there is a lifetime of wonder and mystery under it.

And, just because we can explain it now, doesn't mean Leonardo's paint brush isn't magic. It just means that the magic is real.


I am certainly biased but I would like to see the minimum employee comp package include:

* 83b election standard with employee on-boarding paperwork.

* Bonus of up to one year salary to cover 83b early exercise costs.

* Employee equity pool size disclosed along with 409a. Ideally contractual protection for the employee equity pool.

* Offer documents must include a 409a valuation and share price. (I have gotten multiple offers with just a number of shares. And in one case the company was literally peddling a false valuation in my offer).

* 401k with company match, even for a small company. Lottery ticket is not a retirement plan.

Ideally in my mind:

* Employees get convertible debt that is either as senior as investor debt or pays a 5% dividend.

* Double standard employee pool to 20%; keep 30% founder pool size.

* No ISOs or non-quals ever. Extremely tax toxic, and employees can actually get driven into debt from the job. Options should be outright forbidden for comp packages under $1m / year. The tax games are absolutely not worth it.

The main way any positive change can happen is for employees either to negotiate for it during offers or for founders to just start doing it to be more competitive. One might argue that Carta has made 83b easier by chasing a private stock marketplace... perhaps there are other mechanisms for change.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search:

HN For You