Eukryt Wrts Blg

post by eukaryote · 2019-09-28T21:42:11.201Z · LW · GW · 27 comments

...It's blogging but shorter. I'll give it a better name if I think of one.


Comments sorted by top scores.

comment by eukaryote · 2020-02-03T06:47:07.271Z · LW(p) · GW(p)

Here's something I believe: You should be trying really hard to write your LessWrong posts in such a way that normal people can read them.

By normal, I mean "people who are not immersed in LessWrong culture or jargon." This is most people. I get that you have to use jargon sometimes. (Technical AI safety people: I do not understand your math, but keep up the good fight.) Or if your post is referring to another post, or is part of a series, then it doesn't have to stand alone. (But maybe the series should stand alone?)

Obviously if you only want your post to be accessible to LWers, ignore this. But do you really want that?

  • If your post provides value to many people on LW, it will probably provide value to people off LW. And making it accessible suddenly means it can be linked and referred to in many other contexts.
  • Your post might be the first time someone new to the site sees particular terms.
  • Even if the jargon is decipherable or the piece doesn't rely on the jargon, it still looks weird, and people don't like reading things where they don't know the words. It signals "this is not for me" and can make them feel dumb for not getting it.
  • (Listen, I was once in a conversation with a real live human being who dropped references to obscure classical literature every third sentence or so. This is the most irritating thing in the universe. Do not be that person.)

On a selfish level,

  • It enables the post to spread beyond the LW memeosphere, potentially bringing you honor and glory.
  • It helps you think and communicate better to translate useful ideas into and out of the original context they appear in.

If you're not going to do this, you can at least: Link jargon to somewhere that explains it.

Thank you for coming to my TED talk.

Replies from: ChristianKl, leggi, SaidAchmiz, Chris_Leong
comment by ChristianKl · 2020-02-11T15:15:12.304Z · LW(p) · GW(p)

There are two reasons for jargon.

(1) Developing rationality@LW as it's own paradigma by reusing other concepts from LessWrong.

No field of science can stand on it's own without creating it's own terms and seeing how those terms interact with another.

(2) Defensibly against being able to be quoted in a bad way.

Charles Murray succeeded in writing "The Bell Curve" in a way, where almost nobody who criticizes the book quotes it because he took care with all the sentence to write nothing that can easily taken out of context. Given the amount of criticism the book got that's a quite impressive feat.

Unfortunately, in many controversial topics it's helpful to write as defensibly or even Straussian.

Depending on the goal of a particular post (1) or (2) sometimes matter and at other times it's worthwhile to write for a wider audience.

comment by leggi · 2020-02-05T05:44:47.020Z · LW(p) · GW(p)

Thanks for writing this.

"people who are not immersed in LessWrong culture or jargon."

This is me. A creature from another time and space. I read about a website about rationality and got excited about potentially finding a group of people who think rationally.

There's a lot of interesting stuff here on LW but could be more accessible. More formatting for ease of scanning allows readers to start picking up the important points.

There's a lot of unnecessary words used - I wonder how much editing (pruning?) is done. The habit of giving something a few days to settle then re-reading it before publishing?

New perspectives would be useful for a lot of questions/discussions that I see here.

comment by Said Achmiz (SaidAchmiz) · 2020-02-03T09:12:55.951Z · LW(p) · GW(p)

And making it accessible suddenly means it can be linked and referred to in many other contexts. … It enables the post to spread beyond the LW memeosphere, potentially bringing you honor and glory.

There are often very, very good reasons not to want this, and indeed to want the very opposite of this. In fact, I think that the default should be to not want any given post to be linked, and to spread, far and wide.

If you’re not going to do this, you can at least: Link jargon to somewhere that explains it.

I do wholeheartedly endorse this, however.

Replies from: eukaryote, leggi
comment by eukaryote · 2020-02-03T16:44:31.695Z · LW(p) · GW(p)

In fact, I think that the default should be to not want any given post to be linked, and to spread, far and wide.

Say more?

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2020-02-03T18:03:24.580Z · LW(p) · GW(p)

Several reasons.

The most important one is: the further an idea spreads, the more likely it is to be misinterpreted and distorted, and discussed elsewhere in the misinterpreted/distorted form; and the more this happens, the more likely it will be that anyone discussing the idea here has, in their mind, a corrupted form of it (both because of contamination in the minds of Less Wrong commenters from the corrupted form of the idea they read/hear in discussions elsewhere, and because of immigration of people, into Less Wrong discussions, who have first heard relevant ideas elsewhere and have them in a corrupted form). This can, if common, be seriously damaging to our ability to handle any ideas of any subtlety or complexity over even short periods of time.

Another very important reason is the chilling effects on discussions here due to pressure from society-wide norms. (Many obvious current examples, here; no need to enumerate, I think.) This means that the more widely we can expect any given post or discussion to spread, the less we are able to discuss ideas even slightly outside the Overton window. (The higher shock levels become entirely out of reach, for example.)

Finally, commonplace wide dissemination of discussions here are a strong disincentive for commenters here to use their real names (due to not wanting to be exposed so widely), to speak plainly and honestly about their views on many things, and—in the case of many commenters—to participate entirely.

Replies from: mr-hire, eukaryote
comment by Matt Goldenberg (mr-hire) · 2020-02-04T19:38:19.234Z · LW(p) · GW(p)

It feels quite suboptimal to have a public forum that's indexed on google, and at the same time be trying to deliberately keep the riffraff out by being obtuse.

If you want to not worry about what people will think, while being able to use your full name, you should use a private forum. Not understanding what Moloch means won't stop an employer from not hiring you for considering heterdox views.

On a public forum, where anyone could stumble on a link from google, I think eukaryote's thoughts are quite important.

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2020-02-04T23:13:30.629Z · LW(p) · GW(p)

I didn’t advocate being obtuse. I only said that by default, we probably do not (and/or ought not) want a post to be disseminated widely.

What is the best way of accomplishing this, is a separate matter.

Replies from: mr-hire
comment by Matt Goldenberg (mr-hire) · 2020-02-04T23:52:20.711Z · LW(p) · GW(p)
I only said that by default, we probably do not (and/or ought not) want a post to be disseminated widely.

My point was that if that's a thing you want, you probably do not want a public site like LW. The thing you want is a different thing than what LW is.

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2020-02-05T00:00:10.935Z · LW(p) · GW(p)

I don’t think I agree.

Or, to be more precise, I agree denotationally but object connotationally [LW(p) · GW(p)]: indeed, the thing I want is a different thing than what Less Wrong is, but it’s not clear to me that it’s a different thing than what Less Wrong easily could be.

To take a simple example of an axis of variation: it is entirely possible to have a public forum which is not indexed by Google.

A more complicated example: there is a difference between obtuseness and lack of deliberate, positive effort to minimize inferential distance to outsiders. I do not advocate the former… but whether to endorse the latter is a trickier question (not least because interpreting the latter is a tricky matter on its own).

comment by eukaryote · 2020-02-05T17:27:16.826Z · LW(p) · GW(p)

I think I agree with mr-hire that this doesn't seem right to me. The site is already public and will turn up when people search your name - or your blog name, in my case - or the idea you're trying to explain.

I don't especially care whether people use their real names or pseudonyms here. If people feel uncomfortable making their work more accessible under their real names, they can use a pseudonym. I suppose there's a perceived difference in professionalism or skin in the game (am I characterizing the motive correctly?), but we're all here for the ideas anyways, right?

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2020-02-05T21:50:55.645Z · LW(p) · GW(p)

The “real name” issue is only one part of one of the points I made. Even if you reject that part entirely, what do you say to the rest?

I suppose there’s a perceived difference in professionalism or skin in the game (am I characterizing the motive correctly?), but we’re all here for the ideas anyways, right?

This is not a realistic view, but, again, I am content to let it slide. By no means is it the whole or even most of the reasons for my view.

comment by leggi · 2020-02-05T05:49:02.687Z · LW(p) · GW(p)

Interesting to see the differences in thoughts about purpose of LW and what users want.

Is there a need for the differentiation between posts that are looking for a wide audience and those that want to remain contained to a small group?

Replies from: Pattern
comment by Pattern · 2020-02-06T00:54:04.050Z · LW(p) · GW(p)

Differentiation could also be used to enable a more organized effort to make material more reachable to a wider audience. (Like wikipedia versus simple wikipedia.)

comment by Chris_Leong · 2020-02-03T19:27:39.926Z · LW(p) · GW(p)

One problem is that completely avoiding jargon limits your ability to build up to more complex ideas

Replies from: quanticle, Pattern, eukaryote
comment by quanticle · 2020-02-10T19:28:24.432Z · LW(p) · GW(p)

I think there is a happy medium in between having zero jargon (and limiting yourself to the style of Simple English Wikipedia) and having so much jargon that your ideas are impenetrable to anyone without a Ph.D in the field.

I would also note that not all jargon is created equal. Sometimes a new word is necessary as shorthand to encapsulate a complex topic. However, before we create the word, we should know what the topic is, and have a short, clear definition for the topic. All too often, I see people creating words for topics where there isn't a short, clear definition. I would argue that jargon created without a clear, shared, explicit definition hurts the ability to build complex ideas even more so than not having jargon at all. It is only because of this form of jargon that we need to have the practice of tabooing [LW · GW] words.

comment by eukaryote · 2020-02-04T17:28:47.057Z · LW(p) · GW(p)

Yeah, building on more complex ideas - that you really need to read something else to understand - seems like a fine reason to use jargon.

comment by eukaryote · 2020-06-03T22:57:00.696Z · LW(p) · GW(p)

I have a proposal.

Nobody affiliated with LessWrong is allowed to use the word "signalling" for the next six months. 

If you want to write something about signalling, you have to use the word "communication" instead. You can then use other words to clarify what you mean, as long as none of them are "signalling".

I think this will lead to more clarity and a better site culture. Thanks for coming to my talk.

Replies from: Dagon, AllAmericanBreakfast, Chris_Leong
comment by Dagon · 2020-06-04T16:59:45.587Z · LW(p) · GW(p)

I like this attempt at meta-signaling! Good luck on making your signals more effective by preventing people from noticing that aspect of things.

comment by AllAmericanBreakfast · 2020-09-17T02:22:23.685Z · LW(p) · GW(p)

I think that what "signalling" does that "communication" does not is when we use it to analyze how specific actions convey meaning. For example, there's a rich literature on flirting, in which scientists try to break down how various physical postures and gestures interact with things like laughter to signal attraction or aversion. "Communication" tends to imply a conscious, explicit, primarily verbal way of getting information across. "Signalling" tends to imply a subconscious, implicit, and primarily nonverbal way of getting information across. I think what we need isn't so much a taboo on these terms as a clarification of what the difference is between them.

comment by Chris_Leong · 2020-09-16T08:40:27.565Z · LW(p) · GW(p)

I don't suppose you could provide a specific example of when you think this would improve the conversation?

comment by eukaryote · 2019-09-28T21:42:11.381Z · LW(p) · GW(p)

I don't like taking complicated variable-probability-based bets. I like "bet you a dollar" or "bet you a drink". I don't like "I'll sell you a $20 bid at 70% odds" or whatever. This is because:

A) I don't really understand the betting payoffs. I do think I have a good internal sense of probabilities, and am well-calibrated. That said, the payoffs are often confusing, and I don't have an internal sense linking "I get 35 dollars if you're right and you give me 10 dollars if I'm not" or whatever, to those probabilities. It seems like a sensible policy that if you're not sure how the structure of a bet works, you shouldn't take it. (Especially if someone else is proposing it.)

B) It's obfuscating the fact that different people value money differently. I'm poorer than most software engineers. Obviously two people are likely to be affected differently by a straightforward $5 bet, but the point of betting is kind of to tie your belief to palpable rewards, and varying amounts of money muddy the waters more.

(Some people do bets like this where you are betting on really small amounts, like 70 cents to another person's 30 cents or whatever. This seems silly to me because the whole point of betting with money is to be trading real value, and the value of the time you spend figuring this out is already not worth collecting on.)

C) Also, I'm kind of risk averse and like bets where I'm surer about the outcome and what's going on. This is especially defensible if you're less financially sound than your betting partner and it's not just enough to come out ahead statistically, you need to come out ahead in real life.

This doesn't seem entirely virtuous, but these are my reasons and I think they're reasonable. If I ever get into prediction markets or stock trading, I'll probably have to learn the skills here, but for now, I'll take simple monetary bets but not weird ones.

Replies from: rossry, rossry
comment by rossry · 2019-09-29T01:59:11.495Z · LW(p) · GW(p)

(continued, to address a different point)

B and C seem like arguments against "simple" (i.e., even-odds) bets as well as weird (e.g., "70% probability") bets, except for C's "like bets where I'm surer...about what's going on", which is addressed by A (sibling comment).

Your point about differences in wealth causing different people to have different thresholds for meaningfulness is valid, though I've found that it matters much less than you'd expect in practice. It turns out that people making upwards of $100k/yr still do not feel good about opening up their wallet you give you $3. In fact, it feels so bad that if you do it more than a few times in a row, you really feel the need to examine your own calibration, which is exactly the success condition.

I've found that the small ritual of exchanging pieces of paper just carries significantly more weight than would be implied by their relation to my total savings. (For this, it's surprisingly important to exchange actual pieces of paper; electronic payments make the whole thing less real, ruining the whole point.)

Finally, it's hard to argue with someone's utility function, but I think that some rationalists get this one badly wrong by failing to actually multiply real numbers. For example, if you make a $10 bet (as defined in my sibling comment) every day for a year at the true probabilities, the standard decision of your profit/loss on the year is <$200, or $200/365 per day, which seems like a very small annual cost to practice being better calibrated and evaluate just how well-calibrated you are.

comment by rossry · 2019-09-29T01:38:57.423Z · LW(p) · GW(p)

Hi! I've done a fair amount of betting beliefs for fun and calibration over the years; I think most of these issues are solvable.

A is a solved problem. The formulation that I (and my local social group) prefer goes like "The buyer pays $X*P% to the seller. The seller pays $X to the buyer if the event comes true."

The precise payoffs aren't the important part, so long as they correspond to quoted probabilities in the correct way (and agreed sizes in a reasonable way). So this convention makes the probability you're discussing an explicit part of the bet terms, so people can discuss probabilities instead of confusing themselves with payoffs (and gives a clear upper bound for possible losses). Then you can work out exact payoffs later, after the bet resolves.

(As a worked example, if you thought a probability was less than 70% and wanted to bet about $20 with me, if you "sold $20 at 70%" in the above convention, you'd either win $2070%=$14 or lose $20-($2070%)=$6. But it's even easier to see that you selling a liability of $20p(happens) for $2070% is good for you if you think p(happens)<70%.)

You've right that odds are a terrible convention for betting on probabilities unless you're trying to hide the actual numbers from your counterparties (which is the norm in retail sports betting).

comment by eukaryote · 2021-02-08T07:55:45.165Z · LW(p) · GW(p)

Inspired by the failures of WebMD as outlined here, because this was a problem WebMD characteristically failed to help me solve. 

In the spirit of writing up one's findings, and in the off-chance this is useful to someone, here is a research-based but totally uncited list of indications that a sudden musculoskeletal injury is a break rather than a sprain or the like:

  • If there's a visible deformity, e.g. "something is not where it should be". This is a big indication that you need to go to a doctor, whereas if you don't have this you only maybe need medical attention.
    • (if there's a lot of swelling and you can't tell if there's a deformity, if possible, you might try moving it and comparing it to the other side of your body in the same position - this might show if the injured side is clearly doing something that the healthy side isn't.)
    • My impression is that generally, a minor injury can lead to swelling or make certain motions painful but won't physically shift the underlying structure, whereas a break or dislocation - something that always needs medical attention - can do that.
    • (But it won't always. Stay vigilant.)
  • More serious injuries do typically hurt way more than minor injuries.
  • It also generally takes more force to break bone, especially large bones - jogging probably won't break your tibia, but a car crash might.

But sometimes they don't or you're still not sure. A musculoskeletal injury is more likely to be a break if:

  • The pain is worse at night
  • The area has decent flexibility but very little strength
    • (+ esp. if strength doesn't improve over the next few days - sprains don't bounce back instantly, but you'll probably see some kind of improvement.)

Also, if you get the injury in sort of a distinctive fashion you suspect happens a lot - maybe playing sports, or falling - look up something like 'injuries associated with XYZ', because there are a lot of weirdly distinctive types of tissue injuries with well-characterized symptoms, and if you do have one of those, you might be able to save yourself a bunch of time early on.

This is not medical advice! The safest action is probably always to get your weird thing checked out. But this is, uh, the list of findings I wish I had found about a month and a half ago when I was debating over whether my situation actually merited going to urgent care or not. (It very much did... which I realized upon further research about two weeks after it happened.) 

So, learn from my mistakes, friends. On the "upside", my hand is much better now, and I've learned some interesting things about anatomy in the process?

Replies from: ChristianKl
comment by ChristianKl · 2021-02-10T14:09:34.834Z · LW(p) · GW(p)

indications that a sudden musculoskeletal injury is a break rather than a sprain 

It's worth noting that the two aren't the only possibilities. Torn muscles and ligaments matter as well. Inflamation is another important possibilty concern.