nv-vn

12 か月前議論

Then what are they doing at work all day??? If they spend 20% of their week slacking off then they would get paid more if they just worked the whole time and got more done

nv-vn

2 年前議論

Tbf, this is page 13 of the list. If you go to the first page, Saudi Aramco made $230B with 73k employees according to their website. This is still ~3x as much profits/employee as VeriSign. FAANGs are pretty comparable as well. It's definitely a very profitable company _per employee_, but I fail to find a metric where Verisign can be described as the most profitable public company

nv-vn

2 年前議論

>Today, Verisign is the single most profitable company in the stock market, a great example of an economic termite

This is just blatantly false. VeriSign is in fact the 1215th most profitable company in the stock market [1]. By operating margin, it's 155th [2]. I find it hard to give an article much credit when I can't even get past the introduction without having to comb through citationless false/misleading statements.

[1] https://companiesmarketcap.com/most-profitable-companies/pag...

[2] https://companiesmarketcap.com/top-companies-by-operating-ma...

nv-vn

5 年前議論

Sounds like they're already using Rust where performance matters and are looking to switch

nv-vn

5 年前議論

This is a terribly misinformed take. If you throw enough resources at Python then sure, you can probably get adequate throughput. The problem is that in finance a lot of problems require you to think about latency, which is a total non-starter for Python

nv-vn

5 年前議論

Well they only publish trades ~30 days later. There's no real chance of front-running a politician and for all you know they could've closed their position before the 30 day period ended. Plus, if people follow them into the trade (i.e. people buy because Pelosi bought) that only moves the price in the politician's favor. It's not like politicians are running crazy strategies that you can figure out from their trades, so I can't really imagine it creating any incentive to hide trades

nv-vn

5 年前議論

I think maybe the key idea I was missing here is that the type of a variable is not always the same as the type of the scope it was defined in? If I'm understanding correctly, there's an implicit rule that's missing:

    a : la   b : lb   a := b
   --------------------------
             a : lb

So for example if `b : lb`:

   a := b :>>
   doSomething()

need not have a type `>= lb`.

Does that sound vaguely correct?

Btw thanks for clarifying

nv-vn

5 年前議論

I find the explanation for the sequencing rule rather unsatisfying. Why shouldn't the following be true?

         b : lb      d : ld
   -------------------------------
    a := b :>> c := d : max lb ld

Otherwise, it seems like you could leak info by just assigning two variables, where one is always public. E.g.

    x = user.firstName;
    y = random()

would be "safe."

(It's very likely that I'm missing something here)

nv-vn

5 年前議論

I've interviewed with a reputable company (mentioned on HN frequently) that performs background checks before interviewing you on-site

nv-vn

5 年前議論

Doesn't Tesla pay more than GM?

nv-vn

5 年前議論

I think you may be comparing salary to total comp here, though I'll admit I haven't looked for the source for your $105k figure.

nv-vn

5 年前議論

As an American o had the exact opposite impression -- every European company I've applied to has requested one. Interesting to see that it was just an outlier experience (and I suspect your experience was the same)

nv-vn

5 年前議論

I suspect that removing that functionality would take more work than can be accomplished in 1 day, which is why they temporarily suspended it and why they're adding "limited buys" tomorrow

nv-vn

5 年前議論

Either way, it would be illegal to say that they closed their position for the sake of influencing the market. The SEC really doesn't play around with that type of thing in such a high profile case

nv-vn

5 年前議論

That connection is like 3 degrees of separation at best, though. Robinhood sells order flow to a number of market makers, one of which is Citadel Securities [1]. Citadel Securities is part of the Citadel group, which consists of a hedge fund (Citadel LLC) and a market making firm (Citadel Securities LLC) [2]. While both companies are owned by Ken Griffin and share many resources, they are split by a "Chinese Wall" [3] to prevent any conflict of interest between both sides of the firm. They also have different CEOs, management, employees, infrastructure, etc. as a result of this. Now, Citadel (the hedge fund) has invested $2 billion into Melvin Capital, which is one of the firms central to the whole issue [4]. Yet Melvin Capital has already closed its position in GME [5], which means that they no longer have a vested interest in the performance of GME.

There are several glaring problems with the theory that Citadel is pushing for Robinhood to ban buying GME for financial motives:

1. Robinhood has NO relationship with Citadel LLC

2. Everybody knows about the theoretical conflict of interest between Citadel/Citadel Securities so it is incredibly unlikely that Citadel Securities would risk doing anything to help Citadel in such a high profile scenario (i.e. the SEC/FINRA would be able to find out this kind of thing very quickly and very easily)

3. Citadel has no financial motive to stop trading of GME, because Melvin Capital has no financial motive to stop trading of GME

4. If it was as simple as trying to profit off of GME going down, why would they bother halting trades on other securities (AMC, BB, NOK, BBBY, etc.)?

This article does a really good job at explaining the situation (much better than I can do) [6]

[1] https://cdn.robinhood.com/assets/robinhood/legal/RHF%20SEC%2...

[2] https://www.chicagobusiness.com/article/20170203/NEWS01/1702...

[3] https://www.investopedia.com/terms/c/chinesewall.asp

[4] https://www.wsj.com/articles/citadel-point72-to-invest-2-75-...

[5] https://www.reuters.com/article/us-gamestop-melvin-idUSKBN29...

[6] https://finance.yahoo.com/video/heres-why-robinhood-restrict...

nv-vn

6 年前議論

After I stopped using non-stick, I realized it was kind of a cheat method for cooking. Non-stick is super forgiving because you can cook anything too hot/too cold with too much/too little oil and get approximately the same result. With stainless, cast iron, etc. it actually takes some thought to make sure you have adequate time to preheat and are actually using the right heat in the first place, but by the end of it you end up being a lot more precise in the kitchen. Once you get the hang of proper temperature control it seems like your results get way more consistent. Plus you're spot on with the satisfaction thing, it's definitely more work to cook without non-stick but it just feels better for some reason.

nv-vn

6 年前議論

>For whatever reason, papers ... that ... use data to make arguments about productivity, safety, etc. always seem unconvincing to me. ... maybe it's just because the way most programmers write code doesn't matter to me, because most programmers aren't aware of what is possible on the frontier.

As someone involved in PL research I partially agree, but I think this effect happens in both directions. There are plenty of great ideas in PL that haven't been applied to real problems whereas, simultaneously, there are plenty of huge problems for programmers where PL hasn't been applied. That is to say, us researchers often ignore the problems of the masses in favor of ones with more elegant (or more rigorous) solutions because we enjoy writing fancy proofs and theory more than we enjoy pragmatism. I think both sides of the equation are very important here, but it would be great to see some more direct interaction between the two sides. As it stands, a conference like POPL contributes about as much to recreational math as it does to the state of the art in programming.

nv-vn

6 年前議論

>while those of us who worked on crypto without any specific feelings are early retirees

I'm sure all the Enron investors said something similar before they went bust

nv-vn

6 年前議論

That mindset pisses me off so much. Every time someone asks me about advice for an interview and I don't mention doing hours of leetcode studying they accuse me of lying and trying to sabotage them because _obviously_ the only way to pass an interview is to memorize every possible question the interview can ask! It's so crazy to me how people think this and then I'll sit down with them for a mock interview and watch them completely stop communicating or treat me like I'm a binary switch that only comes on when they have the optimal solution.

nv-vn

6 年前議論

I think the 2008ish version of the internet was just a much happier place. Part of why HN feels like a breath of fresh air still

投稿

Getting Started with Functional Programming

コメント