November 2020: half-arsed linkpost

Assorted links on debuggers, sign errors, board games, academic writing, newsletter seasons and many other things

Hi everyone,

A giant migraine completely ate the end of November and I couldn’t do anything at all for the last three days. I like to always get this out on the first day of the month, to add a bit of sharpness to the habit, but this was definitely a time to make an exception. So it’s late, and half-arsed. I was already planning to do an easy linkpost this month anyway, but I’ve dialled down the level of arsedness a bit further.

So, here goes. November was a bad month for writing and physics, but a good month for reading and getting into interesting conversations on Twitter, so at least I have some material.

‘How To Not Make Stupid Errors All The Time’

The original post is mostly a big list of potential mistakes. This is very useful in itself, but I’m more interested in how to avoid them. The main advice the post seems to have is ‘be more careful’, but I want specifics. For example, one trick I was taught in school was to cross out terms that cancel using a different style of crossing-out for each pair. This makes them visually distinct:

Stephen Malina responded to say he’d asked an extremely similar question on LessWrong last year: ‘What are your strategies for avoiding micro-mistakes?’ This has some great replies by Gwern and others, though generally not quite in the direction I was thinking about - it’s mostly either suggestions for avoiding mistakes in programming, where more tooling is available, or suggestions on how to check your work afterwards. I’d like suggestions for avoiding sign errors the first time!

‘Convolution is just fancy multiplication’

This article from Better Explained was doing the rounds this month. It’s a great explanation of convolution, but I still had the ‘how to not make mistakes’ thread in my mind and was reading more for ideas on how to visually display things to get some of the complexity out of your head. A couple of examples screenshotted from the post:

and:

Also, I haven’t read it in much detail, but the Hacker News discussion on this post actually looks pretty decent.

Coding in the debugger

The ‘how to not make mistakes’ discussion spawned some interesting programming-related subthreads around about this point:

There’s a discussion of how programming generally has better tooling starting here:

Also a branch on ‘coding in the debugger’, so that the data is in front of you and you can keep less in your head:

This one gets into some stuff about Smalltalk and Common Lisp if you keep following the thread.

Horizontal vs vertical writing

David MacIver recommended this video by Larry McEnerney on academic writing:

I watched it a week or so later and it really is good. He has a very distinctive style that makes it very fun to watch (I’ve still got his voice shouting ‘you’re wrong!’ in my head weeks later) and dense with useful ideas. I did a follow-up thread on one of those, his distinction between ‘horizontal’ and ‘vertical’ writing:

The other great thing about it is how refreshingly honest he is about peoples’ motivations. (Favourite quote: "You've learned to write in a system where you're writing to readers who are paid to care about you. That will stop." ) Here’s David Chapman’s thread on this side of it:

I’d also recommend his video on ‘Writing Beyond The Academy’. Same ideas, different application. E.g. if you’re writing for the New York Times, the purpose is to entertain your audience.

Miscellaneous

Robin Sloan’s Advice to newsletter-ers suggests dividing newsletters into seasons, ‘just like a TV show’:

Here’s what you get from the nomenclature, the metaphor, of the “season”:

  • a sense of progress: of going and getting somewhere.

  • an opportunity for breaks: to pause and reflect, reconfigure.

  • an opportunity, furthermore, to make big changes: in terms of subject, structure, style.

When do you break between seasons? Anytime! When life gets weird. When you’re feeling burnt out. When you sense a new obsession taking shape. When you want to bring in a guest writer. When it’s raining outside. Really: anytime.

I like this a lot… maybe I’ll retroactively label this Season 2 so that I’m primed to take a break whenever I next get fed up.

I’ve started reading Kenneth Liberman’s More Studies in Ethnomethodology. So far I’ve particularly enjoyed the board game chapter, ‘The Reflexivity of Rules in Games’. This came out of an exercise he’d run with his students, where they had to get into small groups, buy a board game none of them had played before and video themselves figuring out how to play it.

The point of this is to see what people actually do when they are confronted with a formal system, such as the rulebook for a new game. Key section:

In all of the students' games, two features are ubiquitous. First none of the teams ever read all the rules. They do not even read most of the rules. They open up the box, remove the game parts and game pieces, lay out the board, and are already at the work of organizing the game before a single word of the instructions is read...

Players do not read all of the rules for a very good reason - rules are not intelligible. In themselves, divorced from the context of play that affords them their sense and reference, rules cannot be comprehended...

... Accordingly, the second ubiquitous feature of playing games-with-rules is that the players do not understand the portion of the rules that they do read.

This inspired me to reread the paper Beyond the Rules of the Game: Why are Rooie Rules Nice? (pdf link), by Linda Hughes, which was recommended by Frank Lantz on twitter some time ago. This is about children rules-lawyering a playground game called Foursquare, and goes really well with the board game chapter - it’s about the same processes of negotiation of rules. I’d definitely recommend reading them as a pair.

I also read Hubert Dreyfus and Sean Dorrance Kelly’s All Things Shining, inspired by this thread by @meekaale:

I enjoyed reading this, but I’m not sure exactly what I want to say about it. Maybe something will come up if I let it percolate a bit more.

Next month

No specific plans at all, apart from eating too much. Migraines are often a sign that I’m pushing too hard or getting stressed - I don’t particularly feel either of those this time, so maybe it was just a random brain explosion, but I’m going to listen to it anyway. Plus it’s basically Christmas now and I want to be lazy.

See you next year!