HN user
Matt Lakeman's "Notes on $PLACE" have become my favorite random email update to get. They're all excellent. If anyone knows more about him (does he write anywhere else?), would love to learn more about his background and travel experiences.
I have literally walked in to the one in Redwood City and ordered from someone sitting at a folding desk.
I like WFH for a lot of reasons, but the "I'm more productive" part of my claim is entirely related to the torture chamber that is the open floor plan office. If I had a private office, I'd no longer be able to honestly claim that I simply cannot be as effective in the office as at home.
If there are no substitutes, a high price is extremely justified.
The markup is because I can go to apple.com, pick the RAM and storage I want, and not really have to figure out anything about flavors or distros, knowing that it comes with the best chip on the market. This is leaving out that it just works with my phone and tablet.
It would very obviously be useful for work if you can actually get high res, effectively unlimited monitor space. Maybe not for everyone, but people already spend $3500+ on monitor setups somewhat regularly (and employers definitely do this). Apple themselves sell a single monitor that costs $2300 when fully spec'd out (5k, but the point is that they know what people spend on monitors). I can't figure out why that wasn't the highlight of the demo, since that's just very clearly the easiest way to sell a $3500 device with this specific set of features.
The recording video of a kid's birthday was one of the most ridiculous thing's I've ever seen. I'd maybe record my kid with something like this every once in a while, but I certainly wouldn't be wearing ski goggles while he blows out candles.
Maybe, but that's basically asking him to develop the opposite of the personality type that makes him abnormally good at forecasting.
FiveThirtyEight had about a 1/3 chance of Trump winning, far higher than anyone else was modeling. And he was one of the few pundits openly talking about how it wasn't some insane longshot. I have no idea where this meme came from, but it's annoyingly persistent.
It's pretty obvious when you look at who is quoted in these articles that someone is paying for at least one PR push related to this. PG wrote about how this works in one of his older essays, in fact: http://www.paulgraham.com/submarine.html
Does he mention this somewhere? Last time I spoke to him, he was working on Stripe Press, his interest in fraud and spam prevention long predates his work at Stripe.
Going from 0 to 1 internet providers is likely to have an extremely high conversion rate assuming affordable pricing and quality. While it's anecdotal, I have personal experience with this. My family is from rural eastern WA and spends the summer in the mountains up there. Until 2020, the only internet service option was HughesNet, which was extremely expensive and very low quality, as it's a notoriously rugged region that's difficult to serve (it's still ineligible for Starlink). In 2020, a local was able to get a 5G tower installed (way cheaper and way higher quality). When I had him set us up, he told me he'd already installed it for 60% of the cabins on the chain of lakes our house is on. He expected it to be 100% by 2022.
All that said, my experience does remind me that many of the people up there turn it off for the winter, when they aren't there. My assumption is that there'd be some amount of desire to do that, which would also reduce returns.
Including the installation fee and $30/mo subsidy (I am assuming this means the price he receives is $30 higher than the one customers pay), my quick math shows it would take a bit over 71 months (almost 6 years) to hit $2.6M in total revenue. However, that assumes literally every customer chooses the $55/month plan, if everyone chose the $79/mo plan, it would take almost 51 months, or a bit over 4 years (obviously the number will be somewhere in between that).
Also, this math assumes no growth whatsoever in homes served or other revenue lines. I assume adding another home will be far cheaper than building out the core network, and the article itself notes other lines of business. To be honest, this doesn't seem like a terrible investment to me. There are certainly better ones in a pure ROI point of view, but for government investments? More of these please!
The Diff is easily one of the best value’s I get despite being >$200/year. I’m not sure how I’d rank it relative to Stratechery, I personally enjoy The Diff more but Stratechery is more relevant to my work and is also excellent. Byrne churns out an all-timer like this every other month or so, and his average posts still consistently include the best sentences I read of the day. It’s one of the only newsletters that, if I get behind on reading it, I make sure I catch up on every missed issue.
Patio11 is one of the most responsive public internet users out there. On twitter, he’s not the absolute most responsive, but he has a standing invitation for anyone to contact him about nearly anything on his website, and he takes it very seriously[0]. I know multiple people at Stripe who have been asked to look into things he’s found out about, and he follows up. I won’t claim that he’s perfect about this, but I struggle to come up with anyone at any company of remotely similar scale that is more likely to do this.
Disclaimer: I’ve somewhat recently gotten to know patio11 a bit, but he’d previously responded to my tweets and replied to my emails, long before our paths actually crossed.
I agree that salary is not what dictates my level of satisfaction with my work. However, plenty of people make the (perfectly reasonable) tradeoff of work satisfaction in exchange for money. I'm not dogmatic about this - I am aware that plenty of people are deeply satisfied with both their work and salary, and that others are deeply satisfied with their work and are perfectly willing to trade salary for that. That is great! I am very happy for those people. But it's _also_ a perfectly reasonably decision to prioritize salary to enable your preferred non-work life, and it's entirely possible to still do excellent work if you're in that bucket.
This doesn't even touch on people who _have_ to prioritize salary.
The author lives in Austin. He used to live in Manhattan.
This is why AirPods are popular. While this drives audiophiles and geeks who obsess over performance of their technology (not knocking them, I’m one of them for many devices) up the wall, the reality is most people are not audiophiles and can barely tell the difference. What they want are reasonably comfortable headphones that just work, and Bluetooth is so famously bad at this that AirPods stand out.
moving ad spend to TV, flyers, anything but online because the ROI is so poor
So… deciding to move to channels where ROI is unknowable?
I was just thinking about the last paragraph. I honestly didn’t know they were illegal in CA, since I had so many friends that owned them growing up (inland southern CA). They bought food at the pet store, brought them in for show and tell, etc. They didn’t hide it at all.
At my FAANG job, we happily hired CogSci PhD’s, and I also worked with multiple political science PhDs who’d left or turned down postdocs. I’m unsure about any of the Econ PhDs that I worked with, but I strongly suspect you’re right as it did seem like they were drowning in job options relative to others (conditional on getting a FAANG job in the first place, I suppose).
It seemed like this was about someone transitioning to industry data science, not CS. From the people I have worked with that came to DS by way of academia, the post sounds extremely accurate. It may be that you can get a faculty job in CS without much postdoc experience, but it’s not true for psychology, political science, cognitive science, urban planning, economics, etc (all backgrounds of PhDs I have worked with in big tech data science).
I agree - I have regretted every time that I agreed to a take home assignment, including when it’s been successful. The reason this worked was the HM simply sent me the assignment and asked me to take a look (and clarified that there’d be no hard feelings if I decided not to move forward). I took a look, and it turned out they weren’t lying about it being well scoped!
This car looks small enough that it doesn’t have a backseat, so it’s just the passenger seat available.
In the US, you might end up having to send a whole bunch of 1099s, which I am sure would be a pain.
Two companies have offered me Amazon gift cards, one of them told me that it’s far easier for tax purposes (my understanding is that this is why user research often offers them, rather than cash). I assume they could give prepaid debit cards too, for people who prefer not to shop at Amazon.
I just wrapped up a take home project. When I was told about it, I told the HM that I withdraw from processes that include a take home. They told me that they understood, but that this one really did only take an hour or two, and that they’d send it over to me anyway. All they asked was that I take a look, and if I completed it, they’d send me a $150 Amazon gift card. I took a look, and it really did seem straightforward (basically testing what I’d expect a live, 30-45 minute DS interview to cover), so I went ahead and did it. I ended up spending 2.5 hours on it, but only because I got sucked into making very pretty charts.
I’m pretty anti-take home, because they’re usually scoped as “it should only take you 4-5 hours, here’s a project that would take an in house DS a week at minimum”, but if they were actually scoped like this one (and even mildly comped), I really wouldn’t mind much.
Thanks for sharing the second part! I’ve worked places as a DS where we were ramping up A/B tests, and it’s honestly one of the worst experiences a DS can have - everybody is asking why you need such large samples, then when you get null results, everyone wants an explanation for why (“this idea didn’t do what you wanted it to” isn’t an acceptable answer), etc.
You’re entirely right that running a bunch of tests is a really effective way to advocate for better resources to run the tests. While I’ve never worked anywhere that ran out of ideas to game the testing incentives, it’s definitely true that people who initially fall into the trap of gaming the tests come around with some experience and in general the org builds better intuitions and culture around how to use tests.
It may be that an org that _has_ a culture of testing just has them constantly running in the background but they’re minor in terms of time spent running them, but that building a culture of testing initially involves the whole org focusing on it to a seemingly ridiculous extent, as it’s the only way to generate enough momentum to get the infrastructure and institutional knowledge right.
I’m definitely going to buy the new edition and excitedly re-read it, but it would have been deeply enjoyable if the second edition _only_ updated itself to dark mode. Could have been one of the rare actually funny April Fools Day announcements.
Randall Munroe references this in an answer to a What If? question (my very quick Google of it doesn't find it on the site, so it might be book only).
This crossed my mind as well.