eldavido

This is called being pragmatic. pg is excellent at this.

If the rest of the world operates on credentialism, you either go with the flow and just play along, or refuse to acknowledge how the rest of the world works, and get caught up in a lot of stupid arguments.

It's important to understand, there's a cost to being unconventional. I think pg understands this better than most. Sometimes that cost is worth it. A lot of times though, it isn't.

This comment reflects a broader problem in our discourse: why can't it be both?

Why can't we acknowledge that some of the US constitution's framers were great men who also owned slaves? They weren't perfect. You can be great, and flawed at the same time.

Why can't we acknowledge that YC has had a massive positive impact on the world, while still acknowledging that it's a bit of an echo chamber and does promote a certain amount of self-congratulation?

Cost and type of capital.

Office buildings are funded with mountains of cheap debt.

Pinterest is funded mostly with equity. Maybe debt later when the growth is more stable but mostly equity.

Financially, equity is jet fuel. It's expensive and you use it to build startups, movies, and other intangible-heavy things.

Real estate debt is more like gasoline or diesel. There's a lot more of it than jet fuel, it's cheaper, and lower-quality/energy.

If you try to run a gasoline engine (office space) on jet fuel, you'd likely destroy it and would definitely waste money putting something of too high a "grade" into the engine. Likewise, you can't run a jet engine (startup) on diesel. It won't run.

Getting back to the question at hand, it's more efficient to run the offices on cheap real estate funding, and have the Pinterest operating company lease the asset, than have it own the office directly. Sometimes companies will buy offices themselves (like Apple) but it's something investors put up with, not encourage.

Not quite that simple, Amazon tried to buy Whole Foods last year and has basically failed to get traction on groceries.

Not saying you're completely incorrect but there's room for a lot of companies in retail doing a lot of different stuff.

Show me another place with 100,000 100k+ jobs.

There aren't many markets like this in the US and they're all expensive. NYC, Northern Virginia/DC. Overheated, maybe, but not by much.

It really isn't hilarious. This is how the US loses its long-term reputation as a level playing field for business, a great place to work/set up shop, etc.

This whole process has been unbelievable. The president dusts off some obscure law to ban or block tiktok. Now a forced sale is about to happen and the administration gets a say in who buys it? And they want some kind of finder's fees!?

This stuff isn't supposed to happen in the US! We aren't Zimbabwe, with a dictator who runs around with suitcases full of cash making no compunction about awarding contracts to people closest to him! This is major, high-level corruption! Are we going to start judging other M&A deals by who donates the most to a campaign?

This whole thing is just a huge, stinking, disorderly mess. We can't decide whether we're going to ban Huawei or not. Then the whole mess with ZTE. Now Tiktok. It's capricious and corrosive to the rule of law. And what's worse, not only is it ham-fisted, but this administration makes such a big deal out of "freedom", when having a president stepping in like this is anything BUT freedom -- it is the very antithesis of it, where winners and losers are decided based on personal favorites?

Came here to say this (mostly).

You mentioned a bunch of computer science stuff. There's a lot in pure software development/engineering that's rather permanent, as well. Was working on a project 7-8 months ago where we had a node codebase and a C++ one, and needed to share behavior. Rather than write it twice, I figured there had to be a way to share code directly, because I knew many node extensions have bits written in C. And of course there was -- but it required knowing about things like ABIs, reactors and event loops, linking and compiler internals, etc.

There's all kinds of timeless hard skills in software dev. Knowing how to modularize things properly, how to create the right abstractions, how to name things well, how threading really works.

I'm pretty hardcore capitalist but I also have started to think more about the ideas that go under the heading "market society" -- namely, that everything has its price, that we'll trade away all kinds of things, family time, vacation, meals with friends, etc. for some kind of price. That everything can be bought and sold, that everything has its price.

Some things have to remain outside the realm of market exchange. Marriage is a good example. You just can't put a price on it. Ditto for kids, and a lot of other highly meaningful life experiences.

I think the problem is thinking that markets and commerce are the whole story. They aren't. There's just so much more to life: serving others, community, family. None of this is anti-capitalist per se.

This sentiment expresses something I'm starting to see a lot in our discourse. I call it "third-party outrage", where you (bystander) get mad because you think some "aggressor" did something bad to some hypothetical "victim".

What cracks me up is that oftentimes, the victims don't even care or don't get nearly as mad as the bystandars. Same story with tech workers getting mad about how Amazon pays their employees. Is there any real harm being done here?

So I gotta wonder, why are the bystanders more mad than the alleged "victims" themselves? Do you realize how screwed up this is?

Yes. I'm genuinely surprised how timeless this is.

You watch it today, in 2020, and it's shocking that it was made 20 years ago. This tells me that Judge and the other writers managed to latch on to some very timeless things.

Not sure what they are, but they nailed it.

This is why services like Google Apps or managed exchange hosting exist. Most people are terrible at IT management. So bad they're far, far from realizing how bad they even are.

When you consider you're getting like 1000 of the smartest tech people in the world to manage your infrastructure for you, for $5/month per user, it's really such a no-brainer. If people are too stubborn to see that, or want to waste time trying to do it better themselves "because it's cheaper", with redundant power, OS patching, zero-downtime changes/deploys, proper capacity planning, proper redundant connectivity, provisioning the right network around it, physically securing the server room, ensuring things are properly cooled, not wet...I could go on forever, and this isn't even the main focus of the business...I'm sorry, but they deserve go to out of business.

I had a very stubborn client once who ran a hotel chain. Won't say what or where, but I wasn't surprised when their random "security through obscurity" VNC server got compromised. I wasn't finishing migrating to the new PCI-DSS compliant system we built, either, so there go 5000 credit cards "encrypted" with some sweet rot13-level bullshit in Turbo Pascal I cracked in about 30 minutes with no code access.

This hasn't been my experience, but would love more info if you could provide it.

I'm actually surprised how slowly most tech companies shed staff. In general, it seems the industry is pretty hit-driven, and if you're attached to a good product that's still making money, there's rarely much attrition, even when perhaps there should be. Conversely, a lot of great people get let go when a product isn't working (misses the market or no sales). It all seems kind of arbitrary and random. It's a lot different than, say, a restaurant with really tight margins where you're a day away from getting fired if you aren't perceived to be pulling your weight.

I feel like there are basically three ways to lose your job in tech.

In one, the company cuts an entire business unit, or a company shuts down, or something else cataclysmic happens. This is where you have dozens/hundreds of people let go. Outside of small startups, this seems pretty rare. I think selling/divesting a failing business unit is more common than outright shutdown in tech.

In the second, the employee does something really egregious and gets straight-up terminated for cause (fired). Absenteeism, theft of company property, sexual harassment, something so bad it's borderline illegal and potentially a legal risk for the employer if they don't do something about it.

The third, which is what we're seeing here, is a general reduction in bloat done under the guise of "the economy". In my experience, great people don't tend to get let go in situations like this--it's political cover to remove the bottom 10%. Companies, at least in tech, don't make a routine practice of doing this without "a reason".

Being in the bottom 10% can happen for a lot of reasons. Maybe you just lost interest in the work. Perhaps you don't get along with your manager, or something's going on in your personal life, or the role was never a good fit in the first place. I used to think some people were just "bad", and some indeed are, but it also seems like peoples' performance really does change over their careers. Provided there's good unemployment insurance, probably best for both parties to part ways. Not only does it give the employee a kick in the pants, but it also improves the morale of the rest of the team, because it doesn't feel like someone isn't "pulling their weight".

A lesson I learned too late.

You have to be willing to do the work, yes, but also figure out what you really, really like doing, and then do it -- a lot. Over, and over, and over, for years or even decades.

You say that, but I'm not so sure.

At some point where you live matters a lot. You can be rich, but if you can't get things shipped because the roads are bad, you can't get clean water, there are no local services around (haircut places, restaurants, tradespeople to fix your house), not to mention fun things to do, it looks worse.

I have this conversation with my friend Mike. I work and live in SF as a software engineer. He lives in Indiana and earns 8-10x what I do as a surgeon but I'm not sure he's actually better off. Especially considering there aren't great schools for him to send his kids to or interesting people to see in meetups.

This reminds me of a quote from my international econ class in college: "You can take our tomatoes, or our workers. It's up to you to pick which you want" - Mexico to US

Unintended side effect of this is to prevent competitive entry. No other TNC will be able to achieve enough density from day 1 to run at the wage floor without huge amounts of VC.

That alone should make people suspicious of this legislation. In the long run, this is really, really good for Uber/Lyft.

Not only does it permeate, but this particular issue touches on a much larger, and more timely discussion: the extent to which companies should be loyal to a nation, and its citizens.

In one view, the one I had for a long time, companies are free agents. They can set up where they want, come and go as they please, hire and fire at will. They add value by creating jobs and profit and society should be thankful for that.

A different view, one I hear a lot in the more intellectual conservative circles (Brexit and some Trump supporters), is that companies have a duty to their people, and society at large. They should train people (through apprenticeship, rather than strip-mining universities / the local market), try to promote stability, avoid huge disruptive layoffs, etc.

I used to think the first but have come around to the second a bit.

Everybody's talking about the cost side of this and not the benefit: customer intimacy. India is a huge market for many tech companies, and having more on-the-ground presence means they'll innovate in ways more appropriate to the Indian market.

This is kind of a contrarian viewpoint but one I happen to agree with. As a single person, you can live a great life on $100K.

No, maybe you aren't going to $100/plate restaurants every night or living in a $5000/month luxury studio, but there are plenty of great apartments in Oakland (where I live) for $1500-2000/month.

What's actually started to become jarring is the difference between SF proper and the rest of the region. SF rents seem to be almost 2x comparable places in the East Bay, plus the general price level is higher from Prop C and all the other insane junk the electorate wants. 2x the cost and the city is rapidly becoming a tent camp, I'm happy to live a little farther out and commute to work 1-2x/week when I'm not WFH.

"College for adults" captures the feel a bit better imo. (Wife's phrase.)

McKenzie (original article's author) worked probably a decade outside a major center of tech R&D before taking a job at Stripe. A lot of his advice is written from that viewpoint. It's valuable because a lot of the world operates as he describes.

I think the big lesson is that life inside industry clusters (finance/media in NY, Hollywood, Silicon Valley, oil and gas in Houston) is quite different from life outside. In industry clusters, jobs are way more specialized, standards are higher, and "geek skills" pay off way more. A few jobs ago, we engaged a consultant for decent money to tune our postgres implementation. I can't imagine that job existing, or paying that well, outside of a major center of Internet software development like SF/SV. I'm sure there are analogs in oil and gas, or any other clustered industry.

I like this. It's flexible enough that it could be used in times of true crisis, but it's also super-simple, a major under-appreciated aspect of policymaking.

One thing you definitely DON'T want is a California-style "we can spend with majority but raising taxes requires a 2/3 supermajority". That has produced some truly horrific outcomes in government and needs to be repealed. https://www.westerncity.com/article/californias-two-thirds-l...

Great question. It's not that cut and dried. You could probably ask ten Americans what either the balance of power between the federal government is, OR what it should be, and get ten different answers.

The reality is that, today, the federal government gets the lion's share of attention but the bulk of day-to-day stuff (roads, schools, police, utilities eg water) are run by state and local governments. BUT--and this is an important caveat--SOME (not a lot, but some) money for this is provided by the federal government through about 20 different mechanisms including federal block grants for education, federal highway funding, and other mechanisms.

It's actually quite complicated, but to a first approximation, most local things are operated locally, financed mostly locally, but the federal government does kick in (fund) a little on certain things, in a way that's fairly haphazard and subject to the whims of politics, than any sort of tradition-driven or constitutional way.

Incidentally, this is part of why the US's COVID response has been literally "all over the map". It's mostly in the hands of the states and you're seeing how 50 different political cultures deal with this crisis.

I think the psychology of spending actual cash, in a bank account, that could theoretically be spent on anything, is completely different from using debt to fund projects.

A lot of my views come from operating perhaps the lowest level of "government" possible: a large HOA. Our CC&Rs prohibit entering into any contract (including debt) lasting more than a year. It works for us because the entire place is run by volunteers, many of them nonspecialists, who could do a lot of damage getting us into, say, a 10-year contract with Comcast that was too expensive, not negotiated well, etc.

I think we expect our politicians to be superheroes. This might be borderline reasonable for large, well-funded federal agencies like the Federal Reserve, BLS, BEA, etc who can attract large numbers of truly great people, and pay them well. It's not reasonable to expect someone like Tim Geithner or Ben Bernake (a Princeton economics PhD) to run the finances of a small city or HOA. You're going to get part-time volunteers who know how to balance their checkbooks. Expecting these people to be perfectly rational decision-makers without formal accounting training, let alone knowing how to think carefully about how to analyze NPV, cashflows, and long-term liabilities, in the face of all kinds of local small-ball politics (e.g. trying to help their friend win a contract, or a personal vendetta against someone's project) is laughably far from realistic.

Reading articles and seeing situations like this really makes me question whether public agencies, as a rule, should be allowed to use debt to finance projects.

I'll begin by saying I have no aversion to debt whatsoever as a private individual or for an incorporated business. I own several rental properties, I'm probably more indebted than most.

The thing about debt in general though, is that it makes the bad decisions a lot worse. And I think when you design a political system, you have to think about both the best outcomes possible, as well as the worst. Bad things happen. People make mistakes. Calculation errors occur. I'd gladly take a bargain where I could trade 5% of the upside for 50% of the downside.

What I've realized is that most municipal governments are run by people of fairly average intelligence and drive. There just isn't a lot of incentive to kick the apple cart. They want to go home at 5, they're going to see the owner of the construction company whose project they dump on in the grocery store, etc. This is actually a feature, I don't really want too much innovation in how my government does things, I'd rather go with the tried and true for things like road design and water pipes, than something like the BART where they used a novel track gauge, and we're still paying for it 30 years later.

I just think avoiding municipal debt can curtail some of the worst outcomes. It forces cities to save up for maintenance and really make hard choices about what gets maintained and what doesn't, and it also prevents can-kicking to later generations. You might not get a high-speed rail project, or a major bridge, without debt, but I wonder whether we'll ever get high-speed rail in California at all, after spending billions of dollars already [1] on it without anyone having taken a single trip.

[1] https://www.govtech.com/fs/transportation/California-High-Sp...

I do this and it works, with the slight tweak that mine's on s3.

But yeah, zero running code is a really great thing.

OS API sets have evolved toward more sandboxing. Things are more abstract. Fewer files on disk, more blob-store-like things. Fewer INI files in C:\Windows, more preference stores. No registry keys strewn about. .NET strong naming rather than shoving random DLLs into memory via LoadLibraryA()

(Hi, I'm a windows dev)

I think the other major, major thing people discount is the emergence of viable sandboxed installs/uninstalls, and the accompanying software distribution via app stores.

Windows 95 never had a proper, operating-system supported package manager, and I think that's a big part of why web applications took off in the late 90s/early 2000s. There simply wasn't any guarantee that once you installed a native app, you could ever fully remove it. Not to mention all the baggage with DLL hell, and the propensity of software to write random junk all over the filesytem.

Mobile has forced a big reset of this, largely driven by the need to run on a battery. You can't get away with as much inefficiency when the device isn't plugged into the wall.

Agree 100%.

I wonder how much memory management affects this. My journey has been a bit different: traditional engineering degree, lots of large Ruby/JS/Python web applications, then a large C# WPF app, until finally at my last job, I bit the bullet and started doing C++14 (robotics).

Coming from more "designed" languages like C#, my experience of C++ was that it felt like an insane, emergent hodgepodge, but what impressed me was how far the language has come since the 90s. No more passing raw pointers around and forgetting to deallocate them, you can get surprisingly far these days with std::unique_ptr and std::shared_ptr, and they're finally even making their way into a lot of libraries.

I sense there's a bit of a movement away from JVM/CLR-style stop-the-world, mark-and-sweep generational GC, toward more sophisticated compile-time techniques like Rust's borrow checker, Swift's reference counting, or C++ smart pointers.

I mention memory management in particular both because it seems to be perceived as one of the major reasons why languages like C/C++ are "hard" in a way that C#/Java/JS aren't, and I also think it has a big effect on performance, or at least, latency. I completely agree we've backslid, and far, but the reality is, today, it's expensive and complicated to develop high-performance software in a lower-level, higher-performance language (as is common with native), so we're stuck with the Electron / web shitshow, in large part because it's just faster, and easier for non-specialists to develop. It's all driven by economic factors.