David R. MacIver @DRMacIver Twitter profile

Pinned Tweet

David R. MacIver

@DRMacIver

8 months

New post for you. It's about explaining things. And geese.

2

15

Last Seen Profiles

@kysucix

@gdv_de

@mynap5

@Brian0865276060

@TheWelshDragon9

@rozanne_turner6

@pentendayo

@Dell800659341

@joe_KIWAME

@koyaljainMD

@jandakembangstw

@oficialanais

@GiovanniSpano5

@JuliaKate32

@JoMaddenSports

@Tilburgeveneme1

@smileyy_val

@CassandraM4973

@spring_tf2

@FootFemininMag

@boydhilton

@ZafarCheemapti

@stw_pdg

@John1331271141

@_On1k

@ResumeWoman

@MoncrieffMusic

@MT_a_MT

@kissyvolkov

@JoshWhaley047

@sitappa_shooter

@BiancaCarratt

@abstrakiss

@context_girl

@0108_livly

@jandakembangstw

David R. MacIver

@DRMacIver

2 years

The modern condition is mostly trying to do things on your own that people have historically achieved with a large support network and wondering why you're tired all the time.

36

5K

15K

David R. MacIver

@DRMacIver

3 years

Sci Hub is such a perfect example of the difference between the legal and the moral. Significantly illegal, constantly on the verge of being shut down / blocked / excluded from platforms as a result, and a huge net positive for humanity with almost zero downsides.

37

3K

13K

David R. MacIver

@DRMacIver

4 years

"Minds are basically computers" is wrong if you think of computers as abstract turing machines but spot on if you think of computers as a horrible assemblage of kludges bridging incompatible legacy code which only work because critical bugs are masking other critical bugs.

42

1K

5K

David R. MacIver

@DRMacIver

7 months

Huh. This is a genuine actual output from GPT-4's DALL-E plugin. This is the first time I've seen an image generator quite this pixel-perfect just straight up copy something from its training set.

81

154

4K

David R. MacIver

@DRMacIver

10 months

Humanities people should really learn at least one STEM subject so that they can actually properly grasp that abstraction depth is a thing that really exists and not everything is about near-universal human experiences.

39

192

2K

David R. MacIver

@DRMacIver

3 years

I forget this isn't widely known, so here's a social tool. The "correct" way to ask someone if they know X when trying to decide how or whether to explain something is to ask something along the lines of "How familiar are you with X?"

20

256

2K

David R. MacIver

@DRMacIver

5 years

"Pandas is what happened when someone concluded that the problem with Python was that it wasn't enough like R, so built an R clone on top of numpy, the library that was written when someone concluded that the problem with Python was that it wasn't enough like matlab."

33

388

2K

David R. MacIver

@DRMacIver

1 year

A lot of visible problems that you can't seem to solve are secretly solutions you don't want to admit to adopting to problems you don't want to admit to having.

25

220

1K

David R. MacIver

@DRMacIver

3 years

It's not exactly that the truth is usually in the middle, but the truth is usually super fucking complicated and the edges tend to be simple.

24

171

1K

David R. MacIver

@DRMacIver

7 months

Full cat for those who want it

27

126

1K

David R. MacIver

@DRMacIver

5 months

New rule: You're not allowed to write about the prisoner's dilemma unless you can also explain two other distinct two-player games with different payoff matrices, provide real life examples of them, and lucidly explain why they are *not* examples of the prisoner's dilemma.

60

52

982

David R. MacIver

@DRMacIver

1 year

Every time someone posts that "inherently indescribable nature of the universe" meme I want to give them a lecture on how description works.

26

55

775

David R. MacIver

@DRMacIver

1 year

Here is how to explain a board game. In order: 1. This is a game about [theme]. 2. It is a [type of game]. 3. You are trying to [objective]. 4. The game ends [when condition occurs]. 5. Prior to that you play by [how turns are structured] 6. Here are the details.

37

50

776

David R. MacIver

@DRMacIver

7 years

The stages of talk planning: 1. 45 minutes is fine 2. Oh shit how will I fill 45 minutes? 3. Oh shit how will I fit this into 45 minutes?

13

189

709

David R. MacIver

@DRMacIver

5 years

Apparently the measles vaccine reduces childhood mortality from *all* infectious diseases, because measles is an absolute bastard of disease and impairs your immune system for several years after having it.

Scientists Crack A 50-Year-Old Mystery About The Measles Vaccine

When the U.S. introduced the measles vaccine, childhood deaths from all infections plummeted. Scientists think they might know why: Benefits of the measles vaccine go way beyond the measles.

www.npr.org

11

285

676

David R. MacIver

@DRMacIver

1 year

I used to have a recurring dream that Google Maps had a functioning teleport button, but it was hidden under increasingly hard to find developer debug options. Eventually I stopped having this dream, so I assume they must have removed it in recent Android versions.

2

53

701

David R. MacIver

@DRMacIver

7 years

PSA: If a function accepts a string then it's a parser. Parsers are hard to get right and dangerous to get wrong. Write fewer of them.

15

283

646

David R. MacIver

@DRMacIver

1 year

Thought experiment: A cunning AI can persuade a vigilant user to let it out of the box under some circumstances. Reality: People will create a subreddit about how the AI looks so sad and adorable in its box and work together to figure out how to break it out.

13

56

591

David R. MacIver

@DRMacIver

3 years

Prompted by their Twitter account being suspended. 😢

8

16

505

David R. MacIver

@DRMacIver

7 months

I kinda love this one.

6

19

573

David R. MacIver

@DRMacIver

2 years

A long time ago I had a conversation with a friend which went: Me: "There are a lot of X, so presumably one of them is good". Him: "If there are a lot of X, that's probably evidence that none of them are good." He's been proven right time and time again.

26

40

529

David R. MacIver

@DRMacIver

7 months

Prompt: Draw me a picture: There is a dog wearing a hat sitting at a table drinking coffee. The room around him is on fire. The dog cheerfully says "This is fine" Other images produced in response to this prompt were much more varied.

12

14

550

David R. MacIver

@DRMacIver

1 year

> driving > Google maps says it's found a faster route > I ask if route is actual road or insane horse track > It doesn't understand > I pull out a diagram explaining what is road and what is insane horse track > It laughs. It's a good route sir > It's an insane horse track

19

94

1K

David R. MacIver

@DRMacIver

1 year

I wish people discussing remote work vs in person work were more able to say "tradeoffs exist" I'm basically on team "in person work is, all else being equal, clearly better than remote" and also you'll probably never catch me taking a full time in person role again.

37

21

515

David R. MacIver

@DRMacIver

7 months

Further interesting experimental data.

7

18

472

David R. MacIver

@DRMacIver

3 years

Ugh, fine. 1 like = 1 thing ideally every software developer would be good at (maximum 100). (Many of these things will not be particularly specific to software developers)

7

87

436

David R. MacIver

@DRMacIver

2 years

Have you noticed how the long-term nuclear waste warning message ("this is not a place of honour"), designed to last 10,000 years, didn't last 40? It's been memed so hard that future cultures will just go "yeah the ancients put this everywhere. It doesn't really signify anything"

25

36

414

David R. MacIver

@DRMacIver

3 years

What's your best* theory for where *gestures broadly* all went wrong? You're not allowed to attribute it to broad theories (e.g. don't just say "capitalism"). Be specific. * Interpret as you wish. Both "most plausible" and "most interesting" are good.

470

18

387

David R. MacIver

@DRMacIver

4 years

It's easy to fall into a trap of going "oh no everything is interconnected and I have to fix everything if I want to fix everything" when instead you could be thinking "Yay there are so many positive feedback loops here that fixing anything makes everything else better".

6

99

416

David R. MacIver

@DRMacIver

3 years

One of the funny things about the US is the way that its culture permeates the entire internet in a way that causes people to think they understand it, and then they discover details about what it's actually like in the US and are just completely shocked.

7

78

384

David R. MacIver

@DRMacIver

2 years

There's a tool for reasoning about ethics I use and find quite helpful: The cheeseburger threshold. It's very simple: You ask "Is this act ethically better or worse than eating a cheeseburger?" and if it is ethically better, you don't hold it to higher standards.

23

57

404

David R. MacIver

@DRMacIver

2 years

What books do you feel gave you some sort of intellectual superpower by reading? e.g. they had some lasting extremely useful effect on how you think about the world.

164

30

393

David R. MacIver

@DRMacIver

3 years

There's an interesting sort of privilege associated with never having to make hard decisions that affect other people, and it's pretty pervasive in internet discourse about almost anything that matters.

11

35

385

David R. MacIver

@DRMacIver

7 months

Galaxy brain

2

0

403

David R. MacIver

@DRMacIver

3 years

Here is, I think, the best summary I've got of why people struggle with mathematics: You think you don't understand it because there is too much remember. But actually the reason there is too much to remember is that you don't understand it.

10

40

367

David R. MacIver

@DRMacIver

10 months

Because the actual answer is "I'm sorry, if you don't know what a SAT solver is, you're not deep enough into the abstraction layers of the problem to have a conversation about this, and all I can communicate to you are vibes".

7

389

David R. MacIver

@DRMacIver

7 months

You can't expect someone to honestly tell you what they're feeling unless you can commit to not punishing them for it when you think they're feeling the wrong thing.

2

39

391

David R. MacIver

@DRMacIver

1 year

I wonder if internet archive data from before the last couple of years is going to be the AI training dataset equivalent of low background steel.

12

48

383

David R. MacIver

@DRMacIver

2 years

I unironically think one of the big societal problems we have is that many people who would love to be cogs in a well-maintained machine that makes proper use of their skills towards some neutral-to-mildly-positive end struggle to find those roles.

8

21

360

David R. MacIver

@DRMacIver

5 years

If you are unfamiliar with a discipline then reliably: 1. It's more complex and messier than you think. 2. It requires more and different skills than you think. 3. The problems in it you think are hard are mostly easy. 4. The problems in it you think are easy are mostly hard.

6

123

354

David R. MacIver

@DRMacIver

2 years

Dear every company I use who is not literally an app company: I do not want to install your app in order to access essential services from you that do not in any meaningful sense require an app. Please stop.

10

44

340

David R. MacIver

@DRMacIver

1 year

Apparently the trick to unlocking access to strong emotions is "let yourself breathe as heavily through an open mouth as feels natural, neither preventing it nor forcing it" and for fuck sake it shouldn't have taken me nearly 40 years to figure this out.

13

16

348

David R. MacIver

@DRMacIver

9 months

I feel like a lot of you are very "hammers are bad because when I used to have a hammer I would repeatedly hit myself in the face with it" and, while I agree that you probably shouldn't use hammers that way, I nevertheless suspect you've misidentified the source of the problem.

10

49

353

David R. MacIver

@DRMacIver

4 years

Me, leaving Google: It turns out that being at the mercy of an arbitrary and massively dysfunctional system that I cannot change but must work within is really bad for my mental health. Me two years later: *starts a PhD* Me two years after that: Fuck

5

19

335

David R. MacIver

@DRMacIver

2 years

If you have an item that: 1. You interact with multiple times a day. 2. You find unpleasant to interact with. 3. Isn't actually very expensive. Did you know that you can replace it and your life will get substantially less annoying? It apparently took me a year to realise this.

13

17

325

David R. MacIver

@DRMacIver

3 years

General principle: In a healthy environment, almost everything reasonable you try works better than you can expect it to. In an unhealthy environment, almost nothing you do works even close to as well as you'd expect it to.

3

67

327

David R. MacIver

@DRMacIver

5 years

"mute this thread" is a fascinating artifact because it's a widespread belief that a feature exists that just... Doesn't. You cannot mute threads. I can't think of any other examples of people doing this off the top of my head. Can you?

9

198

311

David R. MacIver

@DRMacIver

3 years

The flags in classrooms / pledge of allegiance thing is a classic.

12

89

300

David R. MacIver

@DRMacIver

3 years

For all the talk of conspiracy theories, there's a disappointing lack of literature on conspiracy practice. Suppose I wanted to form a shadowy cabal to take control of the world's governments, where would I start??

33

29

310

David R. MacIver

@DRMacIver

1 year

I cannot conceive of the sort of mind that thinks smart completion of quotes and brackets is a good feature. It wastes significant time when it gets it wrong, saves almost no time when it gets it right, and is wrong far more often than it's right.

30

12

318

David R. MacIver

@DRMacIver

1 year

Fuck sake. I accidentally had a philosophical insight and now I believe in the unity of the virtues. This is the worst thing that has ever happened to me.

15

14

308

David R. MacIver

@DRMacIver

5 years

I maintain that LaTeX is an abomination against God and humanity, but I confess that when looking at papers written in other fields I do sometimes find myself wondering if the formatting of the end result justifies the blood sacrifices we must make to our dark gods after all.

23

40

290

David R. MacIver

@DRMacIver

1 year

I remain extremely confused that people are under the impression that Apple is good at UX.

38

20

307

David R. MacIver

@DRMacIver

1 year

One of my big moral disagreements with ~everyone is that it seems like normally people consider incompetence as a mitigating factor in how much you blame someone for their actions, and most of the time I consider it an aggravating one.

13

20

301

David R. MacIver

@DRMacIver

2 years

The thing I find most confusing about other people's behaviour is that it seems like most people don't have the reflexive thing where they go "Huh. That's weird", think about it for 5 minutes, and then it makes sense and they update their world model.

15

22

299

David R. MacIver

@DRMacIver

1 year

Did an art

17

16

305

David R. MacIver

@DRMacIver

3 years

Almost everybody is operating at a tiny fraction of the potential they could have reached if they'd had access to dedicated personal tuition and coaching in their education and career, and I find this fact surprisingly distressing.

13

27

294

David R. MacIver

@DRMacIver

3 years

One of my more frustrating beliefs is that treating your employees well is really profitable, and it's actually not that markets are aligned against human flourishing, it's just that running good companies is hard and people who accrue power mostly don't care to try.

12

26

283

David R. MacIver

@DRMacIver

3 years

Goodhart's law (as coined by Marilyn Strathern) is "When a measure becomes a target, it ceases to be a good measure." A thing I don't see emphasised enough is that this causes the measure to be used more, not less, because it gets entrenched by people with power who can game it.

7

73

284

David R. MacIver

@DRMacIver

7 months

Dammit Twitter ate that that was meant to be a quote tweet.

Dr. Dad, PhD 🔄🔼◀️🔽▶️

@GarrettPetersen

7 months

Game: Describe a meme from memory to an image generating AI. Quote tweet this with the resulting images.

35

16

158

1

295

David R. MacIver

@DRMacIver

1 year

Wait hang on I just realised something. Are you... are you all just thinking in your head? Like... you're just trying to form thoughts in your bare brain before putting them out into the world?? You know that doesn't work, right? Brains aren't for that.

14

35

291

David R. MacIver

@DRMacIver

3 years

Software developers seem to reliably have terrible mental health. Assume for the moment this is a real effect and monocausal, what's your best guess for what causes it?

186

12

256

David R. MacIver

@DRMacIver

5 months

I've been reading "Daily Rituals: How Artists Work" on and off over the last few months, expecting to learn a lot of very worthy advice about creativity but it seems like the secret for 95% of them was just to drink unreasonable amounts of coffee and alcohol.

10

9

287

David R. MacIver

@DRMacIver

3 years

Anecdotally it seems like people who understand probability think very differently from people who don't understand probability, even when not explicitly using probability. Is this a real thing?

57

17

255

David R. MacIver

@DRMacIver

10 months

But if you're used to things that are humanities shaped, I think it actually is mostly true that everything is explainable in this way. People often *don't* explain things clearly in the humanities, but that's a separate problem.

9

3

269

David R. MacIver

@DRMacIver

3 years

Vague theory: The idea of needing alone time to recharge is probably at the root of a lot of our failures to rest well, and actually most people need company for proper leisure and we just lack good ways of being with people in a restful manner.

21

17

263

David R. MacIver

@DRMacIver

3 years

The thing blocking your progression is probably that you need to learn when to stop doing the thing you're very good at and do something else instead.

9

22

256

David R. MacIver

@DRMacIver

3 years

One like = one* interesting book pairing * with an option to give up any time after 100 tweets

10

17

255

David R. MacIver

@DRMacIver

2 years

New cause area: Chaotic Altruism. Acts of good which will produce maximally strange and interesting impact on the world.

9

30

256

David R. MacIver

@DRMacIver

7 months

2

4

268

David R. MacIver

@DRMacIver

3 years

I occasionally feel like an obsession with tools is probably a sign of depression, because if you had something you actually cared about you'd be focused on that, while "I want to do the thing I'm required to do better" is a focus you can maintain without really having to care.

19

10

243

David R. MacIver

@DRMacIver

1 year

I think most people's strategy for trying to feel safe is to believe that bad things won't happen. This doesn't work, because ultimately it rests on a lie. Bad things will happen. Safety comes from knowing that when the bad things happen you'll still be basically fine.

9

20

249

David R. MacIver

@DRMacIver

3 years

I've previously suggested that talent is mostly interest, but actually I think for a lot of things talent is mostly just not having stupid emotional blocks in place that prevent you from being good at the thing.

9

29

223

David R. MacIver

@DRMacIver

3 years

Rule for life: Don't be friends with people who clearly hold people like you in contempt, even if they make an exception for you.

9

11

232

David R. MacIver

@DRMacIver

10 months

Was recently having a conversation on discord about something I'm working on and was asked to explain it in plain English because "explaining something simply often helps you understand it" and I did explain it, but there was no way it would help.

3

1

243

David R. MacIver

@DRMacIver

3 years

The popularity of some things is often a sign of how utterly starved people are for anything like that thing rather than that the thing itself is particularly good. e.g. People recommend Polya's "How to solve it", which has the great virtue of being not quite worse than nothing.

12

6

231

David R. MacIver

@DRMacIver

2 years

Are there any billionaires out there who just have like... a dedicated R&D department at their beck and call, for answering whatever questions they can think of?

23

5

235

David R. MacIver

@DRMacIver

3 years

"Pick something you find interesting and become an expert in it for the sheer joy of it" is a remarkably life improving strategy that more people should adopt.

9

20

221

David R. MacIver

@DRMacIver

2 years

I was worried I was about to be unfair to Kant so I looked up the original "is it OK to lie to a murderer" thing and gosh this is so much worse than I expected it to be.

19

14

229

David R. MacIver

@DRMacIver

2 years

People don't actually work 40 hour work weeks and I regularly have to explain this to friends who are breaking themselves trying, so I figured I should write this down somewhere.

People don't work as much as you think

There is a persistent idea that people work 40 hours a week. They don't, and if you try to do this it will break you. Here are some more reasonable guidelines.

drmaciver.substack.com

18

71

230

David R. MacIver

@DRMacIver

3 years

The reason this is the correct form is: 1. It provides *much* more interesting and useful answers than "Do you know X?" 2. It is almost impossible for it to cause offence, because it has no prejudgement about how likely it is that they know about X at all.

3

5

226

David R. MacIver

@DRMacIver

4 years

What are your favourite books about software development? Base requirements: 1. You have to have actually read the book. 2. Yes, all of it. 3. You have to have actually improved at software development by having done so. Nice to haves: 1. Short 2. Not language specific

160

37

225

David R. MacIver

@DRMacIver

3 years

"Have you met an animal? I don't think you've met an animal." is a surprisingly common reaction for me to have when reading philosophy.

6

23

216

David R. MacIver

@DRMacIver

2 years

Advanced social skill: You know how someone tells you their name and you immediately forget it? There's an easy trick for stopping this problem. The easy trick is that after talking to them for a bit you say "I'm really sorry, I've forgotten your name, can you remind me?"

10

8

223

David R. MacIver

@DRMacIver

1 year

What is the boring unsexy foundational work that you're avoiding doing despite knowing it would make the cool exciting stuff you want to work on dramatically better and easier?

60

20

231

David R. MacIver

@DRMacIver

3 years

If your opinions have no consequence because you never have to deploy them in making a choice, you can maintain a moral purity that is not available to actual decision makers.

4

24

210

David R. MacIver

@DRMacIver

8 years

Irregular verbs: I "come up with a clever workaround" You "made some intriguing design choices" They "create unmaintainable legacy code"

4

159

197

David R. MacIver

@DRMacIver

4 years

Controversial take: You can't actually have controversial takes about programming because programmers disagree about everything and it's always boring, so there are no real controversies.

8

22

213

David R. MacIver

@DRMacIver

2 years

It's wild how the prevailing attitude to negative emotions is "A warning alarm is alerting me to some problem. The alarm is annoying, how do I disable it?

7

32

219

David R. MacIver

@DRMacIver

3 years

My two main moods: 1. Why am I still learning this basic shit now when everyone else has it already figured out? 2. Why am I still having to explain this basic shit to everyone, surely they should have it already figured out?

6

20

207

David R. MacIver

@DRMacIver

2 years

Something I'm often puzzled by is how little technological export there is from games to other areas of software development.

18

8

209

David R. MacIver

@DRMacIver

3 years

There tend to be a lot of things that are very low probability of success, but also to within rounding error zero cost of failure, and very very high pay off if they go well. You should obviously doing lots of these. You're probably not doing enough of them.

11

22

209

David R. MacIver

@DRMacIver

7 months

Its attempt at an American Chopper meme diverges significantly from the original but is still pretty good.

2

4

215

David R. MacIver

@DRMacIver

2 years

@the_wilderless [unhappiest guy you know]: I feel so much more grounded and connected after this latest spiritual awakening. I bet after another three years of meditation training I'll be ready to hold down a job.

7

13

202

David R. MacIver

@DRMacIver

10 months

It doesn't even really have to be STEM, I just think STEM is probably the most accessible source of abstraction depth, because technologies (both practical and conceptual) often arise building on other technologies built on other technologies built on etc.

6

2

208

David R. MacIver

@DRMacIver

3 years

Don't know what to do with your life? Why not go into software development! Now you still won't know what to do with your life, but you'll have more money.

11

198

David R. MacIver

@DRMacIver

2 years

Inside you there are two wolves. They're having a great time. Wolves are pack animals, they love hanging out with their buddies.

2

18

201

David R. MacIver

@DRMacIver

3 years

Adulthood is coming to terms with the deep existential terror that you're allowed to just do things.

9

22

195

David R. MacIver

@DRMacIver

1 year

I feel like most "children are XYZ and at some point we adults lose that and we must try to get back to it" takes I run into are from people who either have never interacted with a child in any serious capacity as an adult, or completely ignored reality in favour of ideology.

14

10

210

David R. MacIver

@DRMacIver

3 years

There's something utterly magical about the point in learning a skill where you go from "I am bad at this" to "These are the things I need to work on to get good at this"

4

12

190