Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[silverspawn]

One thing I don't entirely understand about crypto is why you can't achieve a universal order by accepting whatever was published first. (because the novelty of the blockchain comes down entirely to having this universal order).

I mean, okay, if I have 2.1 ETH and I send 2 ETH to X and 2 ETH to Y at exactly the same time, maybe it's literally physically unclear which one was published first. But then just don't count either. If there are millions of nodes simultaneously updating, then even waiting for something like 10 seconds should be enough to make sure everyone got the first transaction. Why can you not implement a distributed ledger that way -- reject any transaction with a duplicate 10 seconds later and accept the rest based on order of publishing? Why wasn't it done before there was blockchain?

[silverspawn]

Machine Learning innovations are sometimes like "So we figured out that if you do A and B and C and D and E and F and G, then you got this cool improvement on problem X". And then two months later, another paper comes out showing "Actually just using B and F gives better results, and A,C,D,E,G are worse than useless".

[silverspawn]

maybe today we'll have some luck with poker...

[silverspawn]

well 2nd and 3d, that's certainly good and there was certainly luck

feel like I've made way more mistakes than the past couple of games, but the choices were also so much harder

[silverspawn]

I'm writing this sequence of posts. i want them to be well written, as well as I can manage. But what does that mean?

here are some axis that seem useful:

- Clarity: should be easily understandable and unambiguous
- Entertainment: should be fun to read
- Elegance: should sound effortless and confident
- Humility: should not sound arrogant

[silverspawn]

However, I think these axes are problematic. It seems like a trap I've fallen into with some recent serious posts is to optimize heavily for elegance and lack of ambiguity but not taking the viewer by the hand. Axis one should probably be split up into Simplicity (the easy part) and Distinctness(?) (lack of ambiguity).

Also I suppose there's grammatical and factual correctness, but those are not going to be thing I overlook

[silverspawn]

So take 2

#1 - Distinctness (should avoid ambiguity; everything should mean exactly one thing)
#2 - Simplicity (should not just be unambiguous and thus understandable in principle but EASY to understand)
#3 - Brevity (should not be longer than necessary)
#4 - Entertainment (should be fun to read)
#5 - Elegance (should sound effortless)
#6 - Humility (should not sound arrogant)
#7 - Correctness (self-explanatory)

[silverspawn]

Right now, I'm going to guess that #3 is quite overrated. People say they want it, but I think what they really want is to not be bored. There is a tension between #2 and #3, and I think #2 should be preferred every time.

#5 is very important but I think I've mostly figured out how to do it when I really try. Does require effort, though

#4 is ridiculously important. More so than anyone would admit. You want to say you read an article because of the content, not because it was fun to read. But honestly, if it's fun to read, will you finish? almost certainly. if it's not fun, will you finish? possibly, but uncertain. Unfortunately idk how good I can be at this even with effort

#6 is a non-issue for most people but something I specifically have struggled with. I write something, it seems fine to me, but in fact it sounds extremely arrogant. I publish it, everyone hates it. I read it again weeks later and gasp in horror at how I sound. Happened e.g. with my first post of Dominion: Reverence, the expansion I created. It also got 0 upvotes, so people do notice this at least if it's too extreme. I later edited the OP and made it sound better, and what do you know, people appreciated it more.

Sometimes though rarely I also read things from other people that score awful on this metric.

I think i've mostly gotten rid of this, but I'm not sure. The weird thing is that I genuinely never did it on purpose and I suspect that other people who have this issue also don't.

#2 is something I've neglected a lot; realized recently that I've focused too much on #1,#3,#5 and thought that made great writing

[silverspawn]

If I look at this list and think about SlateStarCodex now AstralCodexTen, I feel like it knocks it out of the park on every single category except #3. Which probably has something to do with why everyone loves it.

[silverspawn]

There is also a dimension of sounding very fancy with lots of big words, and this is something that can make you successful cough JORDAN PETERSON cough but not appreciated among the audience I'm talking to

[silverspawn]

The benefit of writing something as fiction is then exactly that you dial up #2 and #4, and also #6 is usually a non-issue because you as the author are not present at all. It's absolutely awful for #3 but again #3 does not really seem to matter

[faust]

If I look at this list and think about SlateStarCodex now AstralCodexTen, I feel like it knocks it out of the park on every single category except #3. Which probably has something to do with why everyone loves it.

I don't love it; you broke rule #1 

[faust]

Also I disagree most heavily with the weight you put to #3. It is very important but this also depends on the target audience. If you give me a very lengthy post and I am not initially invested in the topic or the author, I'll probably decide right then and there that it's not worth my time (as has happened with several things you shared in this thread). I also wonder why you think there is a tension between #2 and #3; if anything #3 promotes #2 (which is another reason for why it's important). There is a tension between #3 and #1.

A final reason for why #3 is valuable is because it forces a critical reevaluation of everything you wrote. You have the text; you go over everything and ask "does this really need to be here?" By cutting the surplus, you only end up keeping the best parts of your text, which improves the average quality.

[silverspawn]

I honestly almost wrote "everyone except faust loves it" because I remembered your reaction to the mistake/conflict theory post

I'm not sure I believe you about valuing #3. Do you really not have the experience of just reading on for no reason other than that you're entertained?  Even if you don't, I still think most people do. (Do you think you're a typical reader?)

WaitButWhy is the clearest example here. Every WaitButWhy post ever is *way* longer than it needs to be. They are famous for being long, and you could absolutely say the same with half as many words. But the blog is extremely popular; the author got invited to do a ted talk and got contacted by Elon Musk purely based on the blog.

How are #2 and #3 in conflict? For example, I say something that I think makes the point for X. I can now either leave it at that, expecting people to believe X, or I can explicitly spell out the conclusion.

Or just, you say something, and you can now either add an example or not add an example.

[silverspawn]

an example of a post that I wasn't super interested in and had no plans to finish, yet just kept on reading, was the made-up story about people making weird facial expressions for videos... someone shared that in this thread a while ago

[silverspawn]

I should also mention that i never expect anyone to read anything, definitely not links I drop in this thread, and every time someone does I'm surprised.

[faust]

I'm not sure I believe you about valuing #3. Do you really not have the experience of just reading on for no reason other than that you're entertained?  Even if you don't, I still think most people do. (Do you think you're a typical reader?)

I mean, sure I do, but I was talking about the decision to start reading in the first place.

[silverspawn]

Interesting. I don't think I even check the length of a post before starting to read.

[faust]

Oh, and thinking about your criteria, I feel like a relevant one that's missing is structure. What that means exactly depends on the type of post, but usually you want the reader to have an idea of where things are going and have something similar to thesis - argument - conclusion going on. Subheadings are a simple but effective structuring tool.

[silverspawn]

Yes, full agreement there.

[silverspawn]

AstralCodexTen:


So: Polymarket got fined $1.4 million, and was ordered to make its real-money markets inaccessible to US-based traders (the rest of the world is still fine). It’s very poor news to hear that a villanous political nonentity blocked this vital prediction nexus, and I guess we Americans have no other options besides accepting that we’re vastly poorer now.

Meanwhile, Polymarket put out a rainbows-and-butterflies press release saying that:

    We are excited to continue championing our mission and building out our global footprint, information and educational initiatives, and U.S. product

I assume this means they’re excited to continue building their prediction market somewhere else, and will include a US version with play money, just like lots of other companies have done.


This is the kind of thing I'm talking about. It goes from highly entertaining to depressing very quickly if you compare yourself to it.

[silverspawn]

pizza.

[silverspawn]

currently playing poker on AmericasCardroom, which seems to be the least BS-y poker site out there. Doesn't spam you with how it cares about you, requires the least amount of personal verification, has a flat rakeback deal, has consistent rake to begin with, and a bunch of other things

but idk if that's actually a good idea. I'm afraid I may be selecting for more competent opponents. Sites that trigger my corporate disgust reflex, so to speak, may have weaker players on average. Not that the people on ACR don't make terrible plays all the time; they do. But e.g. the number of people who play just about every hand is low.

Let's give PartyPoker a shot. Sounds like a dumb side based on the name and I remember someone saying players there are weak.

[silverspawn]

Ah, yes. The installer took two tries to work, the window wasn't movable for some reason, and the site demands I put special characters into my password. This is great.

[silverspawn]

another big philosophical question is bits vs atoms

what is more real; bits or atoms? what is the cause of consciousness; bits or atoms?

the popular answer is "bits". As in, flow of information. but, like, this answer seems extremely problematic. Since the beginning of time, no-one has ever seen a bit or something made out of bits. Bits are a made-up concept used to compactly approximate the behavior of systems

take a processor. Sure it's more convenient to talk about it modifying bits. But what are those bits? Well, we've deliberately built it to look at super tiny regions of a surface, decide if they're charged or not (electrically or magnetically or however it works, I'm not a hardware expert), and treat those as a binary thing. Point being, we've *built* things to be describable in terms of bits. But bits aren't real. What's actually going on is just more physics. You could describe a processor doing stuff purely in terms of underlying particles, and this would be *more* accurate, not less.

And it's not like these behaviors have even emerged naturally; again you can only approximately describe a processor in terms of bits because *we've built it that way*. You can only approximately describe a human in terms of bits because *evolution built it that way*.

So why think that Bits are more real and in any way? Why assume that this real thing, consciousness, is caused by this made-up concept, Bits?

Pretty sure the reason is that people want to link consciousness to intelligence, and intelligence is classically described in terms of bits. If you want to say that humans are conscious but rocks aren't, it has to have something to do with what humans are doing, and that difference might as well be described by information flow. At that point, speaking about atoms doesn't help because then you'd have to say "well atoms are conscious, but only if they move in this very peculiar way...". That's not any less implausible than "Bits cause consciousness"

So you can either run with the "humans are conscious, rocks aren't" intuition, and go ahead and link consciousness to this made-up human concept of "Bits" and "information". Or you can reject the intuition, locate consciousness at the level of atoms, the thing that actually exists, and admit that rocks are probably conscious as well.

If put like this, it seems kind of obvious that the latter is more plausible? People tend to treat "rocks are conscious" as this extremely strange thing that requires overwhelming evidence, but I don't see why it would. It's not like we have any experimental data that suggests otherwise. Seems just like another of the recurring failures of vastly overestimating intuition in contexts where we have no reason to assume intuition is accurate.

I think if I had just thought in terms of "bits or atoms", I would probably have landed on panpsychism (all matter is conscious) years ago.