throwawaymath

That's a really effective framing for any company in a highly publicized legal dispute. If it happens to be true for Google in this particular dispute, that's nice but not necessary for the PR speak to be advantageous.

I am much closer to the floor (new grad) than I am to the ceiling, as far as FAANG salaries are concerned.

To begin with, the cost of basically anything you buy online from Amazon, Walmart, Apple, Best Buy, etc is the same no matter where you are in the country. Likewise for digital goods. That's a point in favor of the high COL areas.

Of course it's not that simple. You're right that there are plenty of things which cost more money in higher cost of living areas; namely entertainment, cinema, service-oriented experiences like restaurants, bespoke labor, groceries and housing.

In most of those cases the absolute cost raises significantly but the relative cost to your increased salary is still tiny; for example, I spend $6 - $8 for a half gallon of milk, but since I earn well over $300k/year that doesn't really matter. Similarly movie tickets are ~$18 but again, that doesn't scale enough to make much of a dent relative to a competitive engineering salary here.

On the other hand, some cost increases are significant even relative to competitive salaries. This mostly and primarily applies to housing, but it does also apply to restaurants and entertainment somewhat. But despite the fact that I spend over $4000/month for a luxury condo and another ~$2500/month on fun "stuff", I'm also saving over $100k/year on top of maxing out my 401k. That simply blows out any combination of lifestyle and savings I could enjoy in a meaningfully cheaper area.

Finally there is (unfortunately) an opportunity cost to working outside of high COL areas. The concentration of wealth and capital in high COL cities has a superlinear feedback effect on opportunity and lifestyle. There are numerous Michelin rated restaurants near me, a concierge and retinue of helpful staff in my building, world famous entertainment venues within a 20 minute train ride, numerous gyms, lots of childcare, excellent schools, etc. My commute to work is also only 20 minutes.

But those things don't interest everyone. More practically, it is also easier to quickly change jobs here, either out of necessity or for a quick 20 - 50% increase in compensation. Not only is the higher COL a justification for higher salary, but the employee power that comes with a bidding war puts a positive pressure on external compensation packages. The last time I went looking, I received about 10 offers. I don't even currently work at one of the most competitive companies according to levels.fyi.

I don't want to push this on other people because money isn't everything and it's perfectly valid to choose a lower COL area. But I do want to lay out the hard numbers from my experience so as to give a better picture for the situation.

Though we won't be able to make good on the bet, I would confidently wager a supermajority of people who fit that criteria are not particularly interesting to you, or any given individual for that matter.

In my experience, most people who spend a long time thinking about their controversial beliefs aren't especially insightful or interesting to those who disagree with them. For low hanging fruit we can just look at politics. But even beyond that, the universe of controversial ideas is so vast that it's unlikely a person's given muse will be compelling or insightful to other people.

Once upon a time, similar thinkpieces extolled the virtues of asking candidates to introspect on their greatest weakness. How...original.

What avant garde, thought provoking questions will people in a decade ask as a reaction to present interviewing trends?

Actually I think that question is a candidate for its own answer, now that I think about it...this is like borrowing cocktail party discussion for interviewing.

I'll also go further and say that, while there's a kernel of validity to your analogy, it's not the right analogy with which to deliver your overarching point. I don't think the publishing method for one of the most significant scientific advancements of the previous century is a particularly good lens for analyzing this blog post.

The critical content of this post is far below the threshold usually associated with an idea sufficiently well formed to be publishable. Einstein had a minimum viable theory before he solicited feedback; and when he did solicit that feedback, it was through what we'd consider orthodox channels.

$5 per request is not a negligible amount of money. In practice it doesn't cost anywhere near that amount to call a MechanicalTurk API which will solve ReCaptcha for you. But it's still significant for any nontrivial number of requests, such as in the use case of scraping.

You should adjust your priors here. You're focused on the narrow case where a win condition is achieved by spending n dollars to solve a single instance of ReCaptcha. People who use ReCaptcha are (in my professional experience) overwhelmingly more focused on requiring ReCaptcha to be solved for every individual request of a given type.

I have been in the position you speak of, where I had a revolving set of IP addresses, requesting servers and user agents, and $5 per request would have immediately shut my operation down. As it was, the actual ~$0.15 per request to solve ReCaptcha was sufficiently significant that I couldn't curate enough data for what I needed, despite having all the other resources you mention.

Discipline isn't the only obstacle for autodidacts. If nothing else, at a university you have professors, adjuncts and TAs with office hours who can help you learn something in a fraction of the time it would take to learn by studying a textbook or watching lectures online alone.

For example, vanishingly few people manage to teach themselves an undergraduate math curriculum without going through university.

A PhD is financially viable for anyone who could make ends meet while working at Best Buy. Most people working on a PhD have no family or mortgage.

But I'll humor your implied point: LCTM's failings have nothing to do with the core thesis I'm rebutting, which is that the only value in financial trading is provided by shady backroom dealings.

I don't know where you worked, but please stop perpetuating the myth that everything in finance is shady business in smoky rooms. Contrary to what you're saying, a lot of the alpha generated at the best firms comes from novel approaches to data analysis, not the uniqueness of the data itself.

There is real ingenuity in research which translates into consistent alpha. I'm not going to argue it's literally the maximally valuable way to generate returns in finance, but you're dismissing it entirely. Not everything in trading is relationship building and trying to curate data no one else has.

There is room to combine otherwise public datasets together to find novel insights, and this is frequently done.

I used to do research in this industry and I can tell you that, actually, there are a lot of opportunities for novel research based on huge financial details which haven’t been noticed.

What is your experience, that you write off my own experience as well as entire industry, as being illegitimate? Based on another comment you made in this thread it looks like you’ve also worked in the industry, so did you seriously never come across legitimate research efforts or are you just not mentioning those?

A lot of fundamental hedge funds turned to this in the early 2010s as awareness of big data became a thing, thinking they could close the performance gap with the quant funds. It didn’t work. The quant funds that purchase this data use it as only one dimension of analysis to confirm a hypothesis which has already been empirically tested across many other inputs.

I have a specific example I can talk about, because my old firm abandoned the data: I found a reliable method for predicting exactly how many Model X and Model S vehicles Tesla sold well before earnings each quarter of 2017, including complete configuration data for each vehicle. Even with that KPI in hand, I couldn’t successfully forecast where the stock would go after each earnings call.