Thread by @nota_bennett

On suggestion from @vgr I’m going to do a few of these

1 like = 1 opinion about... learning

Starting with:

If you want to get depressed get to know a class of Education students (some individuals will be great but most will destroy a little piece of your soul)

1 like = 1 opinion about... software engineers

Starting with:

The industry and practice has largely been warped around what software engineers think they want, at the expense of what’s effective & actually makes them happy

1 like = 1 opinion about... leadership

Starting with:

Most people with formal leadership roles aren’t competent at them.

1 like = 1 opinion about Los Angeles (which I am woefully under qualified to comment on but I have feelings, people)

Let’s start this one with
Every Los Angeles burrito I have ever had has been better than every San Francisco burrito I have ever had

1. (‘cause we’re zero indexed here)

A lot of dysfunction in software engineering comes from my generation of software engineers was pretty badly lied to in school— we got sold Computers as a good career to avoid talking to people.

What a funny joke for everyone to enjoy.

2. Corollary to above, this is part of why many of the best engineers are 2nd career folks or otherwise non-CS — we knew what we were getting into— and actually like it!

3. Most software engineer software have way too low of standards for how good our working conditions can be... and not nearly enough appreciation for how much skill good working conditions require.

4. Software engineers who think “pairing doesn’t work” or “I don’t like pairing” either aren’t very good at it or have only paired with people who aren’t good at it.

1. Leadership is mostly seeing clearly— especially what the meta-seeing of what your teammates are and aren’t seeing.

2. The most important skills for leadership are the skills that cause a group to identify you as a leader and these mostly aren’t skills, they’re accidents of group dynamics and social status, oops

@vgr 1. Being good at learning is mostly believing that you’re good at it and, therefore, giving yourself sufficient time to fuck around

2. Most people are kind of fucked up about learning because school by and large destroys your ability to learn while all the adults around you tell you how Important it is

3. Learning is a largely destructive process, especially past about age 25

4. Kids are better at learning e.g. languages mostly because they’re more motivated than adults (what the hell is everyone saying) and have more time

5. Classroom instruction, as a career and as a service, mostly attracts and rewards psychopaths

6. I’m extremely good at learning and extremely bad at studying and I believe that this is evidence that studying is mostly a waste of time but probably mostly what it means is that my memory is very good— both writing and reading are *fast*

7. Most humans have an innate sense of what’s valuable for them to learn and everyone would be happier if we left kids alone to follow that more

5. Most software engineering opinions are pointless and/or fashion. This is a feature not a bug— being able to detect other engineers who share my aesthetic and cultural preferences is more important than using the right tool.

6. Software engineering is mostly a jobs program for mediocre white men

7. The best software engineers are mostly mediocre white men who understand that they’re mediocre and are cool with that

8. The worst software engineers are the best programmers. Past a certain point being great at writing code is an obstacle to doing great engineering.

8. Learning lots of things a little bit is better than learning one thing a lot because it’s less useful

9. Learning takes a lot of energy and if you’re doing it a lot you should take a lot of naps to clean your brain out

10. Learning and genetic adaptation are the same process on different time scales

11. Rocks can learn, mostly about which places are and aren’t good for rocks to be

3. You could do a lot worse for a leader than someone tall, good looking, good at listening, and aware that their job is mostly to tell people they already know what todo & occasionally break ties

4. The military is the only organization that seriously studies leadership and even they aren’t particularly good at it

(Lots of people unseriously study leadership but we aren’t going to talk about MBAs here)

1. Los Angeles is *weird*

Like weird the way that San Francisco was supposed to be weird

12. Most improvements in my capacity to learn have been improvements in my belief in my ability to learn.

Like, learning ping pong improved my ability to learn physical skills in general because I proved to myself I could go from “awful” to “actually pretty good”

13. Most things worth knowing are learned accidentally and unconsciously.

14. I’m underselling my own learning ability a little bit because the other thing that makes me good at it is an obsessive drive to chew every experience I have apart to extract maximum learning meats from it

This is mostly not good for me

15. Eating and learning are also basically the same process

9. Software engineers like computers because they are basically incomprehensible but their incomprehensibility is bounded and describable

Compare to humans, who are basically comprehensible but infinite & indescribable

2. Not having seasons is objectively good and I don’t understand why anyone complains about this

5. Hierarchy is required for large organizations to have inclusive decision-making, because it’s necessary to make decision-making legible.

16. If you can avoid learning you probably should. Do something useful with your time instead.

3. I was not prepared for how many European youths there are in (my particular corner of) Venice, especially Eastern European and French teens.

17. On the other hand if you're an entity you're probably constantly learning and can't help it, learning is just having experiences and being changed by them.

18. *Un*learning is a really underappreciated skill, this is mostly what therapy is about.

19. Restating a couple of earlier points into something more Useful

Learning is interest + time. You can fake the former but it's better to cultivate and discover it.

20. One interesting way that animals can vary is, how long of association chains can they learn. Some cats can link 2 events but not 3.

21. It's really easy to over-learn and over-generalize. At 30 I'm mostly focused on being worse at learning -- learning more slowly and more deliberately.

22. Probably the most fun thing I've done recently, though, was to pick up a skill that's very different from what I do in every day life. (Photography.)

10. Software engineer appears to be the only easily(ish) accessible job that's any good any more and that scares me.

11. The other option is that successful software engineers are super segregated from successful <other things> and that scares me even more.

I think I have some friends-of-friends who are... lawyers? Maybe? I don't know what other jobs are even available anymore.

12. There's no good way to measure software engineer performance, it's impossible, just stop.

13. Really bad software engineers are usually (like most people who are bad at things) scared software engineers.

23. I like the "dive in and quit" heuristic for learning. Most of the major skills that I'm actively "working on" cycle in and out of my life on a cadence of months or years.

24. The really key thing for learning quickly is feedback. If you can get instant feedback about whether you've done something well or not you're going to do much better than if you have to wait hours or days.

25. Some of the most confused, unhappiest people I know are the ones who spent a bunch of time studying things that didn't turn out to be good for them.

26. A lot of hobbies are basically learning for fun, plus some component of either showing off what you learned for social status, or entering flow-state on command.

27. One of the many, many disservices formal schooling does to humans is it teaches us that "learning" is this distinct activity that takes place in certain zones-- and can be optimized by experts.

28. I mean who knows maybe it *can* be optimized by experts, but my experience has been that learning is damaged by teachers more often than it is facilitated by them.

29. Teaching has a similar addictive charge to it as standup comedy and you have to be real careful about your motives if you're going to do it.

30. I was a writing tutor in college, education students cannot write for shit.

31. I think it's kind of corrosive how much of our learning-energy gets channeled towards a narrowly-defined productive effort. We're designed as animals to poke around in weird holes and realize why it was useful later.

32. It's probably our capacity to deliberately teach that sets humans apart from other animals, more than our ability to learn. (Though, we're less apart from other animals than we think, always.)

33. As a puppy my dog tried to eat a toad-- and immediately spat it back out and sat down very hard.

The next year, he remembered that toads were scary but not that he shouldn't eat them.

14. As software engineers we're a lot more like brick layers than we are like lawyers but we like to pretend otherwise.

15. Alternatively we're a new information management helper to the ruling class/scribe class.

16. A lot of the tech goofuses everyone loves to hate are not, in fact, engineers, which is a fact I'm not sure is especially clear to people outside the tech bubble.

17. Almost all the software engineers I know well enough to talk politics with (including and especially the managers) are borderline communists, which probably says more about me and my crowd than it does about software engineers.

18. The engineers I respect the most are almost all really funny. Again this probably says more about me than it does the class but there's a lot that's absurd about our work and I think it's a sign of a well-organized mind to appreciate that.

6. The dominant factor for leaders, especially formal ones -- and like Hofstadter's law, this is under-appreciated even by folks who know about it -- is that you know maybe 5% of what's happening at the layer below you, and even less of the layer below that.

7. Your boss very literally does not know what is going on. They can't. There is so much more happening than they can possibly know, even before incentives to hide information come into play. It is a goddamned miracle when a VP has any sense at all of what line ICs are up to.

8. The Wire is the very best media illustration of this principle that I have ever seen. It is majestic.

9. Sometimes people will argue about command-and-control vs. empowerment leadership.

Guess what! Command-and-control does not and cannot work! There's no world where Just Telling People What To Do will produce the results you intend. You just get headless orgs instead.

10. Most organizations are basically headless, the layers on the ground only having the vaguest of alignment with each other and with their supposed "leaders."

4. Los Angeles is extremely America, probably the most American city I've lived in.