In general, at any given level of child maturity and parental risk tolerance, devices like this watch let children have more independence.

What has changed over the last few decades is primarily a large decrease in parental risk tolerance. I don't know what's driving this, but it's probably downstream from increasing wealth, lower child mortality, and the demographic transition.

Interesting to read through! Thoughts:

• I really don't like the no-semicolons JS style. I've seen the arguments that it's more elegant, but a combination of "it looks wrong" and "you can get very surprising bugs in cases where the insertion algorithm doesn't quite match our intuitions" is too much.

• What's the advantage of making alreadyClicked a set instead of keeping it as a property of the things it's clicking on?

• In this case I'm not at all worried about memory leaks, since the tab will only exist for a couple seconds.

• The getExpandableComments simplification is nice!

• I haven't tested it, but I think your collectComments has a bug in it where it will include replies as if they are top level comments in addition to including them as replies to the appropriate top level comments.

mostly my suggestions will be minor refactors at best ... post it as a pull request

I'm happy to look at a PR, but I think I'm unlikely to merge one that's minor refactors: I've evaluated the current code through manual testing, and if I were going to make changes to it I'd need another round of manual testing to verify it still worked. Which isn't that much work, but the benefit is also small.

One general suggestion I have is to write some test code that can notify you when something breaks

It's reasonably fast for me to evaluate it manually: pick a post that should have some comments (including nested ones) and verify that it does in fact gather them. Each time it runs it tells me the number of comments it found (via the title bar) and this is usually enough for me to tell if it is working.

I think this is an unusually poor fit for automated tests? I don't need to keep the code functional while people other than the original author work on it, writing tests won't keep dependencies from breaking it, the operation is simple enough for manual evaluation, the stakes are low, and it's quite hard to make a realistic test environment.

Sharing the code in case others are curious, but if you have suggestions on how to do it better I'd be curious to hear them!

Done; thanks!

Say more?

I tried to find an official pronoun policy for LessWrong, LessOnline, EA Global, etc, and couldn't.

The EA Forum has an explicit policy that you need to use the pronouns the people you're talking about prefer. EAG(x) doesn't explicitly include this in the code of conduct but it's short and I expect is interpreted by people who would consider non-accidental misgendering to be a special case of "offensive, disruptive, or discriminatory actions or communication.". I vaguely remember seeing someone get a warning on LW for misgendering, but I'm not finding anything now.

I think it's a pretty weak hit, though not zero. There are so many things I want to look into that I don't have time for that having this as another factor in my prioritization doesn't feel very limiting to my intellectual freedom.

I do think it is good to have a range of people in society who are taking a range of approaches, though!

Nice of you to offer! I expect, however, that pressure in this direction will come from non-LW non-EA directions.

The "don't look into dragons" path often still involves hiding info, since often your brain takes a guess anyhow

In many cases I have guesses, but because I just have vague impressions they're all very speculative. This is consistent with being able to say "I haven't looked into it" and "I really don't know", and because these are all areas where the truth is not decision relevant it's been easy to leave it at that. Perhaps people notice I have doubts, but at least in my social circles that's acceptable if not made explicit.