murgindrag

Author of Open edX left a long time ago due to corruption at edX / MIT. See commit log for who built the platform. See press releases for who got the credit. MIT promised all open courses, all open platform, all open everything.

The $800M will be used to line the pockets of privileged MIT professors. It will be as effective at closing equity gaps as supply-side (trickle-down) economics. 60% will go to overhead, which will fund faculty clubs and yachts. From there, a ton will go into generous salaries and benefits packages. And so on down the line.

I am willing to bet that this will be equivalent to giving maybe $10M to an HBCU, in terms of benefits to the poor.

In abstract, no. In practice, in this case, yes. This was an odd way of channeling money into private pockets of well-connected people at MIT.

Anyone know of a good way to reach a good investigative reporter?

Your math is wrong:

* 0.25% of 100B is 250M, not 10M.

* 2.5% of 100B is 2.5B.

OP's math, if you believe the assumptions, was right. Of course, the value of money goes down over time, since there's less time to spend it. I would rather have 10M at age 25 than 10B at age 80. One of the downsides of startups is you need to go through a few before you get to a successful exit, and as you point out, unicorn exits take a decade. 250 startups to wealth would mean you'd be wealthy long after you were dead. I'm not sure it's 250 startups, but it's also usually not 1.

OP was wrong about Facebook having a hard time beating 1M per year. I know plenty of people at FAANG above that. It's competitive but not impossible.

As for me, I'm making a modest salary (less than you make, most likely) at a job where I'm having the time of my life. I do impactful, meaningful, interesting things with all of my life. It's hard to beat that. Money is a means to an end, and if you're already there, it's not rational to optimize for wealth.

The only downside is possibility of loss of job, but if my job had the equivalent of tenure (e.g. I could keep doing this for the rest of my life), I wouldn't worry about money at all. But money has risks too.

Having worked at both small and big companies, I can say you can learn a lot of different things at both. Diversity of experience is valuable.

Speaking of YC, I'd love to have a Y Combinator for older, more established folks with track records....

A lot of the structure doesn't support older founders. E.g.: "We’ll invest $125k in return for 7% of your company using a 'post-money' Simple Agreement for Future Equity (the 'YC Safe'). We think that $125k is currently the right amount for founders to be able to run their company and pay expenses for around 5-6 months, and sometimes even longer."

$125k will support a team straight out of college. It won't support a team at a stage of their life when they have e.g. families or kids.

Population is the big one. 7 billion people today means we have many times more opportunities for viruses to come up, combine, and mutate than we once did. I'm not sure which is the bigger phenomenon.

It's not impossible either.

Cheap brute force hack: Have a chip burn power to heat the chip up to a constant temperature. This was done in one of the early precision band-gap references.

More expensive brute force hack: Stick a Peltier cooler on the thing, and servo to a well-controlled temperature.

Communication equipment is critical to work for emergency use, and coincidentally, emergency use is when you're most likely to see high temperatures.

tl;dr: Military equipment is required to operate like this, and has insanely high-performance RF equipment.

This is one of the many reasons I'd never use Google in any business setting. I had GSuite / free-edition set up for my family. Now, new members don't have access to basic features like Google Voice without shelling out $6/month. A lot my family signed up, so I'd be looking at thousands of dollars per year.

Google is happy to drop you, mostly for an obnoxious up-sell. You're a statistic, and if they drive your business under, that's a statistic too.

I'd like to know the disability rate, which seems much higher than the death rate.

My general feeling is that we will see a structural adjustment:

- We will see a total of about 50-100% inflation. This is not necessarily a bad thing, given the amount of debt in the system. People's mortgage, educational debts, and credit card debts will be lightened.

- The dollar will fall in value relative to world currencies. This makes sense since the US no longer has a monopoly on much of anything. There is no business like the US automotive/manufacturing post-WWII, or tech in the eighties and nineties. The US is a superpower mostly out of velocity.

I don't think this necessarily results in a depression, unless we mess up. This could be healthy or unhealthy, depending on the cards we play.

I don't see a world in 30 years where a SWE in the US makes $350K and a SWE in China/India makes $10k, or where unskilled labor in the US makes $40k, and unskilled labor in India/Africa makes $1k.

The key question is how we get the the new world. This could be through an economic collapse in the US, or through pretty healthy changes. I just don't know.

E.g. 50-100% inflation is enough to trigger a spiral to hyperinflation, depending on how we react.

There are a ton of scams on Amazon. Fake products. Products which never ship. Sellers who exploit Amazon policy loopholes. Etc.

In this case, the scam was convoluted enough that I won't get into it, but no, I didn't pay hundreds of dollars in shipping fees.

On the other hand, I'm sitting at home without having ever received the product I ordered.

As a consumer, I've been !@#$%'ed a few times when I've bought from Amazon and didn't notice it was a 3rd party seller with their own shipping method. One time, I was !@#$% to the tune of a few hundred bucks.

Amazon can do more QC if it goes through them.

I'd like a tutorial mode, where:

1) Only the commands you need are available.

2) They are explained.

I find myself using the commands I know well. This would be a nice tool for learning command I don't know well.

I'm not sure how you count over a billion people in Africa as 5%.

That only works out about right if you're still on the 3/5ths compromise, but accidentally apply it twice.

It means very little. I can show several charts with the same coefficient, and they'll mean very different things.

Why?

That mindset leads to bullying police departments believing it's okay to randomly shoot people.

The whole system is designed as an indirect democracy. I elected people who hire people who work on my behalf.

I can one-up that. Entire Google Workspace domain.

Google sent emails direct to spam, and then wiped the whole thing. All data. All accounts. Everything.

I can't stand fishbowl jobs. Cubicle farms make me depressed. I don't like being on display when I work.

Humans are animals. We stretch. We burp. We fart. We scratch ourselves. We eat. We drink. We dance. We might want to randomly punch something. We nap. We naturally do all sorts of things which are "unprofessional." When we get tired, we need a break to recoup and come back supercharged.

None of those interfere with getting stuff done.

I hate going to work in any job where I'm on display since I need to suppress all of that for 8 hours a day. More with most jobs.

I don't mind going in nearly as much if I have an office with a door which closes.

When I was an executive, I'd sometimes hold longer business meetings with people I knew well hiking, walking, on a kayak, or similar. Those were hyper-productive meetings compared to sitting in office chairs across a coffee table. It only worked for meetings which didn't need a whiteboard or computer, but that's a lot of meetings.

I won't say whether WFH or in-person is the future, but I suspect many traditional corporate practices won't survive.

My general take is that it's not worth it. Professional traders will beat you to it, and you're looking at maybe a 1% drop.

* If it goes up, you lose money.

* If it goes down, and you have $100k invested, you make $1k.

$1k is 3 hours consulting time for a SWE. It seems easier to find consulting gigs and consult for 3 hours than to play the market.

Unless I have specialized knowledge on a particular industry beyond what pros do, I'm index funds, all the way. Those annual gains are helpful. Micromanagement pays less than a typical salary, or investing in up-skilling.

Tex has unprecedented backwards-compatibility. Microsoft does pretty well too. Adobe is a mixed bag. Calling it "almost unprecedented" is simply dishonest.

My 30-year-old Photoshop files used PostScript fonts. Even multiple master fonts, which don't have the same broad support as mainstream PS fonts.

Good luck opening them with Photoshop.

Truth be told, most of my 30-year-old files are in PageMaker.

InDesign has some support for them for a while, but dropped support. Adobe's official line is "Open the PageMaker files in Indesign CS6 or an earlier version first, and then save them to native Indesign Format (.indd ). Once the files are saved in InDesign format, Indesign can open the files."

Of course, Adobe won't sell you old versions of InDesign to do that with, so YMMV.

I dropped Adobe for Affinity not so much for price, but because I can buy the software and use it forever. Even when I retire. I still have an Adobe subscription (paid for by someone else), but I don't use it.

And this seems like yet another good reason to drop Adobe.

I want things I author to be archival, not ephemeral.

* Dropping Flash? Good idea. Dropping Flash with no compatibility/emulation path? Bad idea.

* Switching to license servers from owning software? Bad idea.

And so on. I like the things I do, and I like many of thing things I did 35 years ago too. I'd be sad to see them dead. At the end of the day, I'm willing to take a productivity hit to ensure that if I do create something really good, it's eternal.

Even in retirement, someday, I won't expect to be able to afford the Adobe tax. Do I want my creative work disappearing? No.

I foresee the Canon EF-M competing favorably with the Nikon One mount.

One major limitation of E-mount is diameter. E-mount is 46.1mm to Canon's 54mm and Nikon's 55mm. I think a lot of the other limitations can be fixed, but that one is a bit tough for Sony. That means they will run into issues with vert fast/wide aperture lenses.

Nikon goes down to f/0.95 on their 58mm.

Nikon seems like the clear winner here.

Hey, I accept pieces of paper (and now bits in a computer) for my valuable work, and bought a valuable home, car, and computer with those papers and bits.

Money works purely on belief. If we all suddenly disbelieved in the dollar, the value would be zero.

Curiously, most cycles of hyperinflation come more from loss of faith than from money printing. When people lose faith in a currency, they sell it off as a value store.

Elon Musk's net worth is $28B. His faith in a crypocurrency as a value store counts for a lot.

Kinda. The lunch table is shrinking too. Camera sales peaked in 2010, with 121M units sold. By 2019, they were at 15M units sold, and 2020 was a disastrous year, with COVID19, and it's not clear if there will be a rebound.

My camera is from shortly after 2010, incidentally, and I don't see a need to upgrade. At the same time, mounts are chaos. I have many thousands of dollars of lenses, and I'm not ready to make that kind of investment again, especially with the kind of uncertainty we have in the market.

Nikon and Canon likely got in too late, but the flip side is their mounts are better designs than E-mount. Coming in years later, they learned from Sony's mistakes. If they don't die, Sony might be squeezed out, or might make yet another mount.

Sony's support for A-mount all these years was less than acceptable.

What I would buy into is a mount-agnostic camera which adapted to A-mount, F-mount, K-mount, and EF-mount lenses with everything working (autofocus, video, etc.), and conversely, lenses which adopted to multiple mounts. Sigma almost did that, but mount conversions cost about as much as new lenses. There ought to be a swapable part.

I had exactly one class in college where cheating was rampant. I don't know why, but most of that class cheated on exams; the prof reused exams, and frats had bibles. I didn't see much cheating in my other classes; it was exceptionally rare.

That class was graded on a curve.

I got a 'B', and I would say I was in the 90th percentile for understanding the material. It would have been neigh-impossible to get an 'A' without cheating.

I like zero bug policies in my code. If it's a browser bug, it's a bug not in my code.

My experience is that you pay a premium early on, but it pays dividends down the line. At some point, little bugs interact.

A big correlary of "zero bug" is "clean, simple architecture." If fixing bugs increases complexity, that's usually a symptom of something deeper....

Yes. It's pretty common for vaccines.

One mechanism: If you have a vaccine designed to deliver a spike protein (without a virus) surrounded by a carrier. The body learns to destroy the carrier. On future doses, the body recognizes the carrier, and the vaccine is destroyed before the body can react.

Of course, a vaccine with a different carrier won't be affected by this mechanisms, but there are many other mechanisms as well. I'm not an expert, and I don't know all of them. I'm not sure science knows all of them either, for that matter.

There have even been instances of vaccines making diseases worse too (actually, viruses not too dissimilar to COVID19). And there are a bunch of other possibilities.

I'd buy the basic premise of giving each person one dose for giving everyone imperfect immunity if not for these sorts of complications. I'd rather just wait a few more months than risk a small (but definitely non-zero) chance of something really, really bad.

Path for an eight year old (who started at 5):

1) Hour of Code and Blockly, as well as some Blockly Games, age 5-7

2) Khan Academy JavaScript and processing.js at age 7.

He's pretty awesome at JavaScript now. The programming courses on Khan Academy are great. I will mention your child needs to be ready for it. At that age, a few months one way or the other is huge, as is having the mechanics of typing and mousing.

Next step from there is the book "Code: The Hidden Language of Computer Hardware and Software (Developer Best Practices) " and Nandgame, as well as Micro:bit.

I recommend against Scratch. It's a rabbithole that goes in the wrong direction. You get stuck in a local minimum, where it's really hard to move onto real programming, and within Scratch, there are a lot of local minima (move to random position).

It does provide protection, but what we don't know is:

1) For how long

2) How it impacts future vaccination efforts

(As well as a slew of other things)

If it was just #1, I'd be with you. If we give a big chunk of the population a vaccine which lasts 6 months and makes them unimmunizable, we've changed a serious problem into a serious, serious problem.

Candidate thinks they're worth $250k. Employer pay bracket is $200k. Options:

1. Bend policy at hiring. $250k. $250k. $250k. $250k. $250k. $250k (no raises)

2. Follow salary brackets. $200k. $210k. $220k. $230k. $240k. $250k (low initial salary, catch up over time)

Having been on both sides, it's nearly impossible to make the case for bending policy further at annual reviews. Over time, employees converge to pay brackets. If you were hired with a on-off exception, you're getting no raises for a long time.

No raises feels shitty to a lot of people. It also brings risks.

True story: I was hired by a not-for-profit doing good work. Pay was maybe 1/3 of Google, and I'd be the top tech person at the org (not by ladder, but by skills). Pay was about 85% of what I needed to cover cost-of-living. They stretched brackets to get up to covering my cost-of-living exactly, so I was almost 20% over the upper cap of my pay bracket. Come next re-org, what do you know, they re-jiggle salary brackets standardize things, and my salary comes down 20%.

Fortunately, in the meantime, my cost of living had come down, so it worked out okay. I'm doing important, fulfilling work, and it's fun. So long as I'm doing that and making ends meet, I'm not tempted to jump ship to FAANG.

And no, my employer wasn't simply f-ing me; I understand the organizational structure, psychology, and dynamics well enough to understand what happened, how, and why. These were dynamics well above anything my boss, their boss, or their boss has any influence over, and I'd rather spend the social capital of going to their boss on something more important (e.g. making sure everything I do is open source).

A common career mistake people make is to assign intentions to places where the other side has no control. If you're worth $350k on the free market, and you receive a $180k offer, odds are the other side isn't low-balling you or doing anything mean or improper. They're operating from their constraints. You're operating from yours.

If you need $350k, and they can spend up to $500k, you'll be hired, and how well each side negotiates sets at where you'll get paid. If you need $350k and they can spend up to $200k, you're not getting hired, period. It's good to discover that earlier rather than later, but in practice, that usually comes up after the interview. In the no-hire $200k/$350k case, although you might not get a job now, things might change in five or ten years, and it's good to leave without bad feelings, and I'm not just talking about expressing bad feelings, but about what you're actually feeling. If you come out thinking "low-ball exploitative !@#$%," that's not very healthy; if you come out feeling "good people; no fit; difference in expectations; maybe next time," that is healthy.

I want Youtube to recommend educational videos, a la Extra History and 3B1B. That's most of what I watch on Youtube.

Youtube recommends stupid crap based on what other people like.

If I watch one stupid video not in incognito, ALL of the recommendations become stupid videos because those videos have better metrics. Youtube actively guides me away from educational content.

Even within educational videos, if I click on a click-baity channel once, that channel dominates things I've watched MANY more time. It seems to all be about click-through metrics.

I've never had Youtube recommend really upscale educational content like Zach Star before. It's quality stuff, but you gotta find it yourself.