alecbenzer

I've never heard an email sent via Gmail referred to as a "Gmail" before.

Given the author admits:

> Note that those outside the company, such as investors or regulators, seem like they should effectively count as a level under some circumstances, but not under others.

I think there's obviously some room to fudge with what the exact boundaries of the "organization" are. So yeah, maybe a sufficiently-empowered department head functions more like a CEO/boss.

Sure, but what's early? I don't think it's unheard of to get to O(1 basis point) at 100s of employees.

I've seen an offer at a ~50 person company for 1-2 basis points.

Having 400 employees doesn't say much about how equity is distributed, though. E.g., (made up numbers, I know nothing about Plaid's comp specifically) if an employee gets 1 basis point over 4 years, that's 0.01% * $5.3 billion / 4 = $132,500, which combined with a startup salary might land you at around Google L4.

Yeah from what I've heard living outside of SF without a car is pretty hard. Does pretty much everyone drive? Is biking common?

Yeah I guess this is fine as long as you account for it. Starting with $800k, $40k expenses, 2% inflation and 5% return, retiring at 35 you end deplete your money at 67, but starting with $1M takes you to 83.

Taking out more than $40k/year erodes your principal though, I thought that was pertymcpert's point?

Like, if you've got $800k and can reliably earn 5%, as long as you're withdrawing only $40k you're good. But if you withdraw $41k one year, your principal is now only $799k. Next year if you need $42k, the return from your $799k is only $39,950, so your principal goes down to $796,950, etc.

From a quick spreadsheet, if inflation is 2% annually and you can get 5% on your principal, you deplete your $800k in ~32 years.

Doesn't seem too hard to fix with some supplemental income here and there?

Per https://en.wikipedia.org/wiki/Exponential_growth, geometric growth is exponential growth.

TBH having worked with some ex-Bridgewater folks I'm pretty skeptical how much Bridgewater actually lives up to it.

Obviously you need certain core pieces of infra that handle protos generically (like serializing and deserializing) but

a) total SLOC for that logic is much, much less than SLOC for code that works with _specific_ protocol buffers

b) "consumers" of protocol buffers as a tool/technology mostly don't worry about what's going on under the hood of the generic serialization, etc. code

So it will often make sense to to make that core generic logic even significantly more complicated if it means making the stuff that everyone has to write over and over again even a little bit easier.

> many of them around the type system, as this article points out, but also a lot around the language APIs.

I think there's a lot not to love about gRPC/protobufs (used them at Google and again now at a startup) but I don't feel like this did a good job highlighting those issues.

This was like 2 or 3 sentences worth of good points sprinkled in between a bunch of stuff that I'm pretty sure isn't an actual problem for anyone. If anything, the things pointed out are issues for people trying to write code that manipulates protos generically, which is not what most people spend their time writing and is probably exactly the wrong thing to optimize.

The main good point: Google's problems are probably not your problems, don't just blindly adopt Google tech for no reason.

Also: calling people amateurs without really substantiating is a huge smell IMO. The average Google engineer isn't a genius or particularly amazing by any stretch, but especially for something as core/foundational as protobuf, the answer is much more likely something like "these decisions made more sense for Google internally, especially when weighing against the cost of significantly re-architecting how proto works". The ad-hominem at the beginning reeks of someone who had an email chain that went like:

"You guys are doing proto wrong, don't you realize protos should obviously be like XYZ?"

"Well actually we'd like to do X but it would've been too hard, I'm not actually sure Y is a net-win, ..."

(omg what amateurs...)

Not sure I understand the point you're trying to make...

Money is a factor but it's not the thing that should draw you to a startup -- expected money is better at Big Tech. A startup is the right choice only once you start to consider non-money factors.

> I feel a lot of the folks here are focusing on the wrong parts here. The $200M loss was a cheap way of getting user attention, but hey it worked!

It "worked" in that that it got people's attention but distracted from the post's main message.

> The real reason you should consider working at a startup or starting your own: It's not the money

I think this is correct, but makes the post's title seem really click-baity. I think expected financial upside is exactly where Big Tech unequivocally wins over startups (esp. joining and not founding), so it seems weird to frame the post around money.

I hope that eventually as more and more people realize how much more they could make at big tech (IME a lot of people just straight up don't know, esp. when a lot of the comp comes as equity grants), startups will have to start paying more/offering bigger slices of equity. But seems we're not there yet.

I also wonder how much of this is driven by software engineers who couldn't (or, probably more realistically, don't think they could) get a job at Big Tech?

I share the intuition that price discrimination feels wrong, but I have trouble really articulating why.

Why is "the seller would have given it to me for less" any better than "the buyer would have given me more for it"?

> the maintainer's having written another popular project.

Potentially relevant: https://vorpus.org/blog/why-im-not-collaborating-with-kennet...

Yeah from my limited view when I was there as an L4, Google felt very much run by a weird network of politics and some bureaucracy. Mid-high level VPs seemed like they controlled things a lot more than real higher-ups.

I think this can be... interesting? Trying to give more power and autonomy to individual teams is nice (there were some efforts that tried explicitly to map things to some kind of free market model). Sometimes higher-ups would be like "hey we're all gonna migrate to technology X" but individual teams were like "lol nah the new thing sucks" and keep using it and nothing really ever happened. But I think that attitude also kinda opened the door for a large political/influence market. And there's not _real_ market pressures on individual teams that I think tend to keep the actual free market a little closer to ground truth.

This all reminds me of a critique I'd heard of flat organizations, which is that all groups of people are going to have power structures, and by not formalizing them, all you're doing is allowing them to opaquely evolve in strange ways on their own.

I really wish that someone could study internal Google politics and actually understand it and map it all out.

I think a hard engineering problem in general is knowing when to adopt what level of flexibility. Erring too far in either direction gets bad.

I think the EE vs. CS framing is somewhat inflammatory, but the general lesson of not over-complicating designs is a good one.

Yeah, I don't have any informed opinions on Tim's role in DRM or related things, but the idolization surrounding him always left a bad taste in my mouth.

Like I said, I just think it's a poor name for that mindset. If what you mean is "Count on your equity being worth $0", say that, and not "Value your equity at $0".

Maybe this has to do with the contexts I've heard it in, but my gut is that it too often deters people from thinking more critically about their equity and whether they're being treated fairly. If someone new to startups read just the comment I originally replied to and not any of this follow-up, would they have understood what you're trying to say?

I think you're making a fair but unrelated point.

Time isn't liquid. If instead of giving a startup time, you gave them literal cash, then it would make sense that you get preferred (instead of common) stock, to protect against the founder just selling the company immediately for the value of your cash (as was mlyle's point). If you give them your time instead, there's no risk of them immediately selling off "your time", and so it makes sense that you get common stock instead of preferred stock (at least for this one consideration).

The other general point I think you're making though, is that if we look at the value of the time being invested by a startup employee, their ROI is much worse than a literal investor in the company.

This is true, and somewhat unfortunate I think, but I think happens because VCs control larger amounts of capital and thus have more leverage. Both in terms of absolute amounts, but also how quickly they pay it out.

You taking a $400k paycut for 4 years and a VC investing $1.6M are numerically the same, but the VC gives up the $1.6M immediately, whereas you give it up over time, and can at any time decide to stop investing. When you first start, maybe your ROI is worse than a VC's, but three years in, maybe the ROI of the time your investing _then_ will have improved relative to a VC that tried to invest three years in.

The time still isn't liquid, though. Yes, if you worked for someone for a year, you've given them $X (where $X is the paycut you took), but the founder can't just turn around and sell that $X to someone else. To the extent that they can, it's (to some extent) because the founder was able to take your time and turn it into something valuable, which is like, what founders/CEOs are supposed to do. At that point these is some sense in which the founder has also "earned" the liquidity.

With a direct cash infusion there's none of that. Any idiot can immediately sell $10M for $10M.

Does any startup give market rate salaries?

This is a really good point that I hadn't considered. Employees are "investing" money in a startup too in the form of opportunity cost, but a founder can't turn around and "sell" the investment when it's just time; when it's actual cash they can.

I wonder if there are stories about this happening in certain contexts that predated liquidity preferences?

> if the employee worked for substantially below market wages all of that time

For an engineer that could get a job at FAANG or similar, I think this is pretty much universally true for all startups.

having an opinion == omniscience?