2026-02-11 00:50:10
Did you ever ask yourself: hey, if I added up the electoral college votes of all the US states I've visited, would it be enough to win the presidency? Ask no more: use my Electoral College Travel Map to find out for sure, and/or plan future travels to get over the line.
2026-02-09 19:20:27
I try not to Advocate Stuff too much, but here's a thing I do that I think you should too: when you have a good experience with a big company:
1) search for the primary customer service contact on https://www.elliott.org/company-contacts/
2) send that person an email appreciating the thing
Personally this mostly happens to me with good call center support, but you could really do it with anything.
I think this is valuable because:
a) customer service people get tons of complaints, it's probably genuinely meaningful for them to occasionally have someone email their manager to compliment them instead of complain about them
b) it just feels like cosmically the right thing to do, in a world where (mostly) people only email to complain and never to compliment
c) in some small way, I truly believe that this kind of thing can shape company incentives. I really want companies to invest in e.g. high quality phone-based human-operated customer service, and that's not cheap, and there will inevitably be factions within the company arguing for choosing a cheaper and worse option instead.
In at least some cases, I believe you can meaningfully strengthen the hand of the internal factions who advocate for spending money on better service if you can help them show up to meetings with a slide saying HERE IS AN EMAIL FROM A LITERAL HAPPY CUSTOMER SAYING HOW MUCH THEY APPRECIATE THE EXACT THING I HAVE ADVOCATED SPENDING MONEY ON.
I cannot prove this is true but I think it's how stuff works in the world, and if you want to see more of something you should enable the people who are trying to make it happen. And hey, if it's not-true, the first points still hold anyway.
2026-02-07 00:11:20
When I got a new job via a friend's tweet, I originally thought the org was going to give her a referral bonus (equal to about a week's salary). It turned out they wouldn't, so I sent her that money instead.
(Confession: I was dumb about this and didn't think about the pre-tax/post-tax issue, so effectively ended up sending her twice as much money as I should have – remember your taxes!).
This was a lot of money to me, and I hadn't yet earned the higher salary, so the decision was expensive, but it felt morally right. I want to reward people for sending me good things, and in the long run the job should have been worth a lot more to me than the referral cost.
I want there to be more of a norm of social acceptability for paying post-hoc bounties for good things people send your way: new jobs, lovers, places to live. I'm not sure how to make that norm happen.
I think in most situations it wouldn't make sense to have a formal contract around this, or to declare a fixed amount in advance: we probably need the flexibility to differentiate the bounty value of an offhand reference that leads accidentally to a job, from a warm introduction but no followup, from someone who holds your hand through the whole application process.
I think a lot of damage has been done by rules of thumb like "when you go to a wedding, give a gift that equals the cost of your plate," rather than some complicated formula of your income and their incomes and how close you are and how unnecessarily expensive the weddings was. Similarly, I think a fixed % bounty for introductions would be bad.
But I don't think it would be crazy to have a norm that if someone sets you up with a great job/partner/home, you're expected to send them something substantial at some point in the future as a thank you.
I do think there's something complicated here where under-compensating someone can feel much worse than not-compensating them at all. I have twice now been in situations where I set people up with career opportunities that have netted them millions of dollars in expectation, and where their response was "cool, I owe you dinner!" And, you know, I don't object to the dinner in theory but in both cases the way they said it made it clear to me they thought the dinner was fair recompense for the connection, and I was like.... no?, I'd rather get nothing, because previously I was a generous guy who did someone a nice favour, and now I'm just a chump who makes hilariously bad trades. (Similarly, I'm happy to do things and receive nothing but gratitude from people who are actually nice and actually share gratitude).
I also wonder if there's some way to signal to strangers that you're an adherent of the Bounty Community. For example, many of us get emails from near strangers asking to Pick Our Brains for job opportunities. I would feel more positively about these emails if the person had a line in their email signature affirming that they were a Bounty Keeper, who would actually give me something if my help was helpful to them, rather than (say) wasting my time with questions they could have answered online, inartfully transitioning to asking me for job leads, and then ghosting me after that.
2026-02-04 19:28:50
I just love coming up with dating app ideas; my previous batch here.
Classic swipe-based dating app, BUT you pay per swipe and the cost goes up by the number of un-responded messages the person already has in their inbox. So you're incentivized to swipe and message people who AREN'T already sitting on 1000 unread matches.
(This could be a literal pay-per-swipe, i.e. each swipe has a cost and when you see someone's profile you also see "$2.31 to swipe right on this person". But I think non-economists would find this offputting, so it wouldn't fly; you can achieve the same by making people pay $30 per month to get a certain number of Points, and then matching costs 1/2/3 points per match depending on how full their inbox is).
Ok, here's how it goes: suppose you're secretly into something mildly embarrassing, like reading ATVBT. Maybe you're not even embarrassed about it!, but still you don't need every stranger on a dating app knowing about your, ahem, predilection. You want to meet other people who share this niche interest, but currently the only way to do that is to have it loud and proud on your profile, and/or date a bunch of random people and wait to see if any of them are into this thing. Inefficient!
What would be great is if dating apps let you set a secret "flag" with the niche thing you're into, and then preferentially match you with other people who have the same flag. It wouldn't be deterministic, so you always have plausible deniability, it's just that you know that when the app tells you "we think you'll like THIS person", there's a reasonable chance that this is a secret signal: here is one of the few other sickos who read this stupid newsletter.
You can then do an exciting flirtatious dance around each other as you slowly figure out whether this is the case: it's funny we're meeting on a dating app, you say; I was just reading a blogpost with weird ideas for dating apps, as you scan their face for that sweet, guilty look of recognition.
If it turns out the person sucks in other ways, you can always feign innocence. "A blogpost? Nobody reads those any more." And then just swipe to the next.
If you run a dating app and would like to pay me to work on it, please email me at [email protected]. I am extremely serious, I missed a fork in the road ten years back to work on dating apps and have regretted it ever since.
2026-02-02 19:46:05
Many years ago, I had a friend who was extremely kind, generous, gregarious, interesting... and seemed to annoy a lot of other people we knew. I remember being stumped why she wasn't more popular: I couldn't really point at anything she was doing wrong, and so many things she was doing right.
Eventually I had the thought maybe she's trying too hard? – there was always a sense that she WANTED everyone to have a good time, too, that she needed you to enjoy yourself as much as she was.
I think that was the right direction, but one step short. Like a lot of my favourite people, and a lot of the people I know who are not as universally beloved as they should be, I think she was stressing people out.
There's a lot of different ways to stress people out. Certainly one of them is "being needy", which is unfortunate, because a lot of people are e.g. genuinely feeling anxiety, and asking for reassurance for that anxiety, and in the process self-fulfilling their anxieties.
Meanwhile, people who have status and power and popularity in a given social circle will inevitably find it easier to be relaxed and at-ease in that circle. What can I say? To those who have much, more will be given.
I am under no illusions that merely realising "you might be stressing people out" is enough to be able to tackle it: as mentioned, some amount of your ease-of-being is a hard-to-fake signal of how you're feeling. You cannot un-stress yourself by stressing about it.
But I do think that try not to stress people out is a specific instruction that you can try out in social situations, with different implications than other possibilities you could try. I have written before that parties are like babies, and that if you're stressed while holding them they will get stressed too. But I actually think every social situation is like this (approximately), and if you think in conversations my job is to be at ease rather than my job is to be interesting, you will likely get more of both? [Edit: or my job is to have fun, or some other instruction that suits your personality!]
How do you be at ease? Some things to try are:
2026-01-30 19:46:54
Last year I tried a product called MathAcademy that I really wanted to like. I was recently reminded of a review I tweeted about it, and I wanted to have a canonical version of that review on my blog as well. But please be warned that this is almost-certainly not-interesting unless you're very interested in math education.
MathAcademy: if you read this and are willing to let me use your product to learn in what you consider a slightly sub-optimal way, please let me know, I'd be delighted! I truly think you're onto a great thing and are needlessly limiting people's ability to learn from it.
Anyone else reading this who might have an in at MathAcademy and can help me do the same, ditto!
MathAcademy was one of the weirdest product experiences of my life. In short: I think they have an amazing product, but they're very opinionated about how you use it, and as a result I can't use it at all, which makes it worse for me than a worse product I could actually use.
Basically: the MathAcademy team have figured out some really smart things about how learning works, such as interweaving small units of different topics in order to help you REALLY learn (and remember) new material – they contrast this with traditional homework assignments where you do the exact same problem 100 times, to the point where you're doing it by rote that day, but then can't remember a thing a week later.
I think this is extremely true and insightful, and if I'd had this product in high school when I *had* to study a fixed curriculum I'm sure I would have learned it much faster and better.
But as a result of this philosophy, their product is super prescriptive about which topics you can study and in what order, and unskippably forces you to study the units they say in the order they've decided is best for your long-term learning (assuming a fixed curriculum).
I sort-of understand the part about forcing you to take certain units in a certain order, because of e.g. the aforementioned educational benefits of interweaving material. Sometimes there's a tradeoff between enjoyment and learning (see, e.g., DuoLingo), and I get that on some level I'm "hiring" MathAcademy to force me to keep doing the hard parts as well as the fun ones.
What I don't understand is why MathAcademy is fixated on the "fixed curriculum" part. I hated calc in college, and loved linear algebra, so I was was super excited to re-learn LA and signed up for the relevant course. When I got to the course.... I had a few days of a really fun experience, putting in an hour a day and having a great time, until suddenly my entire "next units" became unskippable trig/calc/etc, as far into the future as the site would let me see.
After several rounds of emails with the (very nice and responsive!) team there, begging them to let me study the LA I signed up for, they confirmed that this was just fundamentally their philosophy and that I had to do it their way.
The most frustrating part for me is that – among many cool things they've done! – they've built these beautiful DAGs of every unit and every dependency within each course. Therefore, more than any other educational provider that I know of, they actually do know exactly which units within a course require which other units as pre-req. So, if they wanted to, MathAcademy could grey out any areas of the Linear Algebra course that require a given unit of calc/trig, and just let me study the rest. Who knows, maybe that would even motivate me to go back and study more trig or calc!
But since I'm literally not allowed to progress at all until I do a seemingly-endless stream of trig/calc pre-reqs – and with literally 0 of the material I'm actually interested in to keep me interested – I'm very sadly giving up.
It all seems bananas. They insisted by email that they were only giving me the pre-reqs that were absolutely necessary for the course I was taking, but again: who says I want to study the whole course as they construed it? I wondered aloud to them if their product was designed for high-schoolers who HAVE to take a fixed curriculum, and if my use-case was outside their scope, but they said that actually most of their customers are adults like me learning for fun or work-interest. In that case, why does it matter whether someone wants to study the whole of "MathAcademy Linear Algebra" or just some subset?
What can I say? I suppose if I were sufficiently motivated I would just go back and learn everything they told me to before studying the thing I want to, but it's hard enough to find time/energy for independent study as an adult and this was just too far for me. Presumably there are people out there who are even more dedicated than I am, but surely just wanting to take linear algebra as an adult puts you in the 99th percentile for this, and enough to be worth supporting?
(As AI has improved significantly since I took this course, I have seriously considered just cheating: going back to MathAcademy and getting an AI to do the units I don't want to do so I can do the units I want to. But I don't have the energy to cheat a system into letting me learn maths, I'm sorry).
I do want to stress: I think a strange thing happens sometimes when someone makes a product, and someone else says "I wish you'd made a different product!", and the producer just thinks "ok, that's nice for you, but I made the thing I wanted to make." I'm not under the illusion that MathAcademy owes me anything, they have every right to make this amazing product and then lock it away from me as they have done. I am sad about all the above, because they want people to learn maths and I want to learn maths from them, and the current situation seems needlessly wasteful to me. But I'm sad about it not mad about it, they do not owe me anything.
I would still recommend MathAcademy to an interested high-schooler who has no choice but to learn all of these topics anyway: with the (strong) caveat that I never got far enough with MathAcademy to learn any new material, rather than re-learn things I once knew, it definitely feels like they've figured out smart things about teaching and learning.
But if you're an adult in similar shoes to mine, I would sadly advise staying away. I've never seen such a clear example of an organization letting the "perfect" (by their own lights, but not by mine!) be the enemy of the good, or of people who've figured out so much about the pedagogy and yet missed such a huge amount about human psychology: instead of learning slightly sub-optimally, I'm now learning nothing at all.