Cofounder of Beeminder
dreeves
Karma: 777
Ooh, nice! I had a very compressed version of that as tip #4 in my list of tips from Inkhaven and actually paid out a $15 honey money prize for it (long story).
Actually that’s a small subset of your advice. Probably the more important part is what your title refers to. The thing where you tell someone, “I can’t figure out how to say XYZ” and they reply “what you just said works”. Scott Aaronson said that was his secret to adding value as a contributing writer at Inkhaven. He’d read the draft, say “I don’t get it”, the author would explain, and Scott would say “great, write that!”.
Good call. I should’ve included more of that! Off the top of my head, I loved it because Lighthaven as a venue is cozy and delightful and because everyone involved was absolutely lovely. The participants were impressive and of course getting to hang out with other Contributing Writers (eg, Scott Aaronson, Scott Alexander, Dynomight—filling out my top 3 favorite bloggers of all time) was pretty amazing. The organizers worked their butts off to make everything as conducive to writing as humanly possible. Plus a ton of fun activities as well—hiking trips, open mic nights, you name it. And so many nice touches, like having an assortment of typewriters and fancy pens to try writing with. Or the Winner’s Lounge with ice cream and video games that you’re only allowed into once you’ve published for the day. And I’m only scratching the surface here. Very highly recommended!
Oh, yes, great point! It didn’t even occur to me to mention that you absolutely can’t say “this dessert has fewer sugar”.
I have an old post on the scare quotes question: https://messymatters.com/scarequotes/
In short, I have the following cases where I claim you should not use quotes:
(0) of course no quotes for emphasis, (1) don’t use quotes to indicate that you’re not going to explain a word, (2) don’t try to distance yourself from a phrase by putting it in quotes, (3) use italics instead of quotes for introducing a term, unless it’s also a mention as opposed to a use of the term, in which case either is ok.
And that leaves the following as the remaining acceptable uses:
Quoting someone or otherwise referring to the literal string of characters
To indicate you don’t mean something literally — like adding a parenthetical “not really”
Layers of further irony: (1) I actually started this post around 6pm, honestly believing it would take a couple hours and I could take the rest of the evening off for a change. Of course I ended up hitting publish right before midnight. (2) I fell into a quintessential Wikipedia rabbit hole along the way, uncovering a case of citogenesis (see footnote).
Ha! I’m embarrassed to have missed this reference without neuroprosthetic assistance.
https://www.gnu.org/fun/jokes/ed-msg.en.html
Maybe I should give ed a try??
Ooh, thank you! I can’t get it to work though. I have it in my bookmarks but when I click it, nothing happens anywhere that I can see—not in the textarea or elsewhere. I’ve tried it on lots of websites and, nothing.
PS: I tried one more time and suddenly it’s working in GitHub Issues comments at least!
Does it work for you for LessWrong posts?
Good point! I’m noticing that in VS Code, the autocomplete is getting scary smart. You’ll start doing some tedious edit and the AI is immediately like “so… continue for the rest of this block like this?” and you can just hit tab. For a while I would hit escape in annoyance when it did that. Why would I trust the AI not to introduce subtle errors? But (a) that doesn’t seem to happen, and (b) it does a good job of highlighting the parts that will actually change so you can vet it pretty easily. It’s pretty freaking magical (modulo the part where it’s possibly a harbinger of dooooooom).
One more update. I’m thinking more about the “permanent DST” idea and what the logical extreme of that would be. I think it would mean that whatever time dawn is on summer solstice, that’s what we call “8am”, year round. No sleeping past dawn ever! (Unless you want to; we’re just talking about the standard “when things happen during the day” range, aka business hours.)
Then the downside is that it’s dark till like lunch time in the winter. Maybe that’s ok!
I made a tool to experiment with such questions: dreeves.github.io/daylight
I made this annoying meme after thinking about how most of the anti-DST arguments involve wishful thinking about how we could get the best of all worlds if “everyone would just”. Of course everyone will never just.
I think the following are valid ways to argue against DST:
Estimate the value people get from the additional summer evening daylight. Like what would people collectively pay to hit snooze on the sun setting for an hour during the summer. Then compare that with the many costs, like sleep disruption and software bugs. And of course the lost morning daylight.
Propose a realistic scheme by which business hours could be shifted without changing the clocks.
I think the nay-sayers have a visceral feeling that changing Time is just deeply, fundamentally stupid. Maybe even wrong in a very literal way. Time is what it is. Changing our clocks does nothing but sow confusion. Can everyone not just do what they want to do when they want to do it, regardless of what numbers are displayed on clocks? I think people really can’t. We are social creatures and those numbers are critical infrastructure for our ability to coordinate.
I’m getting into lots of fun arguments since posting this. I think a lot of people are hung up on how they personally aren’t beholden to 9-to-5 so it’s all cost and no benefit to them and they suspect that it’s probably a minority of people who are really beholden to 9-to-5.
I think even if it’s a huge majority of people who are perfectly flexible in when they wake up and how they use the day’s daylight, my argument works. It’s about Schelling points.
Also, to be clearer about my model, in winter we wake up at dawn and use all the daylight—little that there is—efficiently. Then the days lengthen and we keep walking up at the same clock time. So we’re sleeping through a lot of daylight, which is inefficient. We want to shift to starting the day earlier in the summer. Changing TIME ITSELF is the only realistic way to do so.
Note that this only applies to a certain range of latitudes. Closer to the equator the length of the day doesn’t change enough for this to be an issue. And close enough to the poles you either have so much or so little daylight that there’s little room to optimize.
In between those latitudes there’s a pretty huge upside to shifting the standard hours when group activities happen to align with when the sun is shining.
And I’m not saying there aren’t big downsides. Mostly I want the debate to acknowledge the tradeoffs!
It’s been two years and I’m wondering if “epilogenics” ever got any traction. I got in an interesting debate about it in an ACX subscribers-only comment thread and the worry was—due to typographic similarity (or even the existence of this call-to-action?) -- that “epilogenics” was too easy to be spun, for political or culture-war purposes, as “rebranding eugenics”. If so, maybe a bland acronym like “human genetic augmentation (HGA)” would be better? (Also, it may be obvious but articulating the case for having a hypernym at all may be helpful.)
Ah, thank you! Sounds like Obsidian users will find this more convenient than eat-the-richtext. Maybe we could start a list of other editors or tools that solve this problem...
UPDATE: See the full LessWrong post on Eat The Richtext. Original note follows for posterity.
A couple days ago I wanted to paste a paragraph from Sarah Constantin’s latest post on AGI into Discord and of course the italicizing disappeared which drives me bananas and I thought there must exist tools for solving that problem and there are but they’re all abominations so I said to ChatGPT (4o),
can you build a simple html/javascript app with two text areas. the top text area is for rich text (rtf) and the bottom for plaintext markdown. whenever any text in either text area changes, the app updates the other text area. if the top one changes, it converts it to markdown and updates the bottom one. if the bottom one changes, it converts it to rich text and updates the top one.
aaaand it actually did it and I pasted it into Replit and… it didn’t work but I told it what errors I was seeing and continued going back and forth with it and ended up with the following tool without touching a single line of code: eat-the-richtext.dreev.es
PS: Ok, I ended up going back and forth with it a lot (12h45m now in total, according to TagTime) to get to the polished state it’s in now with tooltips and draggable divider and version number and other bells and whistles. But as of version 1.3.4 it’s 100% ChatGPT’s code with me guiding it in strictly natural language.
Here’s a decade-old gem from Scott Alexander, “If Climate Change Happened To Have Been Politicized In The Opposite Direction”
I think this is a persuasive case that commitment devices aren’t good for you. I’m very interested in how common this is, and if there’s a way you could reframe commit devices to avoid this psychological reaction to them. One idea is to focus on incentive alignment that avoids the far end of the spectrum. With Beeminder in particular, you could set a low pledge cap and then focus on the positive reinforcement of keeping your graph pretty by keeping the datapoints on the right side of the red line.
I guess in practice it’d be the tiniest shred of plausible deniability. If your prior is that alice@example.com almost surely didn’t enter the contest (p=1%) but her hash is in the table (which happens by chance with p=1/1000) then you Bayesian-update to a 91% chance that she did in fact enter the contest. If you think she had even a 10% chance on priors then her hash being in the table makes you 99% sure it’s her.
To make sure I understand this concern:
It may be better to use a larger hash space to avoid an internal (in the data set) collisions, but then you lower the number of external collisions.
Are you thinking someone may want plausible deniability? “Yes, my email hashes to this entry with a terrible Brier score but that could’ve been anyone!”
This should be fine. In past years, Scott has had an interface where you could enter your email address and get your score. So the ability to find out other people’s scores by knowing their email address is apparently not an issue. And it makes sense to me that one’s score in this contest isn’t particularly sensitive private information.
Source: Comment from Scott on the ACX post announcing the results
Ah, nice. I can’t argue with this (de gustibus non est disputandem). Although I almost feel like you’re making my point for me. For those for whom the ineffable joy of Emacs wizardry isn’t a factor, learning such wizardry is an investment that’s unlikely to pay off. But that’s an empirical question. The joy might even be effable after all, making the whole question empirical: Will you maximize your utility by embracing or eschewing powerful text editors?
I don’t actually have a very strong prediction. I just wanted to make the points that repetitive mindless editing is less costly than it seems and the wizardry is more costly than it seems (modulo the intrinsic joy, as you say). In fact, I just thought of an analogy: the mindless repetitive editing is like doodling during a lecture; creating a macro or other wizardry to avoid the mindless editing is like texting during a lecture. The former leaves your brain engaged with the topic at hand and the latter engages it elsewhere.
You can say that in this analogy it would need to be particularly life-affirming texting. Again, I can’t argue with that. I’m just highlighting the cost. If the cost is worth paying, that’s fine.