Elliot's Microblogging

There’s something really problematic about how much people prefer memes, gifs and art over written sentences. I guess I’ve heard many times before stuff like that, statistically, most people don’t read many books if any. And I’ve heard about high schools failing to teach literacy well, poor standardized test scores, etc. But I don’t think I have it concretized well. I have issues dealing with people who are bad at reading and writing, but I’m pretty sheltered in who I actually communicate with. Although I read bad writing samples all the time online (mostly from self-selected, non-representative people who are well above average at writing), I don’t have a good enough intuitive grasp of how regular people are worse and uncomfortable with written English (I don’t think the situation is better for other languages).

I found another channel with Elon Musk debunking/skepticism. First video I watched seemed pretty good:

I looked into this today after patio11 promoted a pro-Musk article:

I thought the article was really bad and unconvincing. I then looked at several pro-Musk thunderf00t debunkings, which I also found broadly unconvincing. In some cases they appeared to be pointing out lists of minor errors (some correct and some incorrect) that didn’t actually change thunderf00t’s anti-Musk conclusion. But they basically had no introduction, no conclusion, no “what does this mean?”, no higher level analysis, just getting lost in biased details where they are trying to poke any holes they can in thunderf00t’s side. (One guy was better and basically concluded that he thought thunderf00t’s numbers were a bit off but were still good enough to correctly debunk the thing he was trying to refute, which was off by significantly more than thunderf00t’s numbers.)

1 Like

Yeah. It used basic math and made straightforward arguments. Easy to follow video.

I’ve been pretty skeptical of Musk in general for a while. His lies about the “self-driving” capacity of his Tesla vehicles are actually dangerous. If he’s willing to lie about a serious business like automobiles, it fits that he would lie about other stuff too. I can’t imagine relying on his judgment for interplanetary travel! I wonder how much better, if at all, the other space-related efforts are (like from Bezos or whoever). I haven’t investigated them.

ppl won’t say things to me like that they think some other thinker or writer is good

they just afk and read/watch others

if they said someone was good i’d have questions like: i disagree with him about major issues. suppose i’m right and he’s wrong. what is the most reasonable series of actions i could do which concludes with him changing his mind?

they would have no answer and have to concede the guy they like is irrational, likely wrong, closed to debate, and blocking criticism.

it’s not exactly complicated logic. it doesn’t take a bunch of study to get the point.
they could talk thru and consider this despite being beginners.

they don’t want to face what this means about the world and about ppl they like – and about themselves.

they simply don’t want to engage with that sort of Paths Forward thinking and take it seriously

broadly i can’t get anyway to discuss the state of the world. i talked about it some on YouTube live streaming a while back, like about Musk and Bezos. you can look at a few basic facts about them, including their choice of women, the schools they went to, and a few things they told the press, and reach some major, negative conclusions. ppl are sometimes like “yeah” about particular points but don’t want to look at the world overall in an organized way using reasoning like that.

Around 64min, Lenderman gives a rule of thumb for chess that I’d never heard of before but it makes sense.

If you think for a while about a move (e.g. 10min) and you still aren’t sure, you should try to go with the move you thought about originally. What would you have done if you had to move in 5 seconds? What was the first move that looked good to you before you analyzed the position in detail? That is what your gut/intuition favored. Your gut/intuition is right ~90% of the time, especially if you’re an experienced player, so you need to put some trust in it.

Although I’ve written some negative posts about Elon Musk recently – which I stand by – he does have some good traits. He says some stupid things but also some good things. For example he argued with the “Executive Director for the UN World Food Programme” today on Twitter.

Musk, responding to a news headline saying 2% of his wealth would solve world hunger:

If WFP can describe on this Twitter thread exactly how $6B will solve world hunger, I will sell Tesla stock right now and do it.

This reminded me of my belief that Aubrey de Grey should actually write out a detailed budget of how SENS would spend the billions of dollars he keeps asking for. I suspect AdG can’t convince Musk, Bezos and others to do a huge donation because he doesn’t actually have a good enough plan.

The Director replied saying headline was inaccurate but $6B would do tons of good and he still wants Musk to donate. Musk replies:

Please publish your current & proposed spending in detail so people can see exactly where money goes.

Sunlight is a wonderful thing.

Director replies refusing to do that without admitting he’s refusing:

.@elonmusk Instead of tweets, allow me to show you. We can meet anywhere—Earth or space—but I suggest in the field where you can see @WFP’s people, processes and yes, technology, at work. I will bring the plan, and open books.

Musk also asked this and got no response:

What happened here? UN officials 'force children to perform oral sex for food' in warzones | World | News | Express.co.uk

(That article is even worse than the headline makes it sound.)

Fame, social status and wealth doesn’t get you Paths Forward. (Nor does a news article mean anyone answers an issue.) Elon Musk can’t get meaningful responses from a UN director who is asking for billions of dollars from Musk.

2 Likes

I recognized that I was wrong about something.

A bureaucrat insisting that stacked containers are an eyesore, causing freight to pile up because trucks are stuck sitting on empty containers, thus causing a cascading failure that destroys supply lines and brings down the economy. That certainly sounds like something that was in an early draft of Atlas Shrugged but got crossed out as too preposterous for anyone to take seriously.

When I first read this, I thought the blogger likes Ayn Rand and has read Atlas Shrugged.

On further thought, he could pretty easily have written this without reading the book and/or while disliking Rand. The book has enough of a reputation and he may have read someone else saying something similar before. I think I’ve occasionally seen people outside my community say it, though I don’t remember any particular examples.

I’ve made this kind of comment before myself. But he might not have meant it in all the ways I do. Often my point is about defending Atlas Shrugged, which is a much more realistic book than critics give it credit for. But the blogger’s focus could easily have been on attacking the law; it might not have even occurred to him that he was implying basically that the stuff in the final version of Atlas Shrugged is more realistic than reality. (The government in the book makes laws that are less bad than real laws, not unrealistically bad, so the book’s portrayal of bad government is realistic not unrealistic.)

1 Like

Wordpress design is awful

Notify me of new comments via email.
Notify me of new posts via email.

You can get (email) notifications for new blog posts. That’s good. And you can get notifications for new comments on a particular post. That’s useful sometimes.

You cannot get new notifications for comments on any post, which is a useful option for low traffic blogs that you like.

You cannot get new notifications for comments that directly reply to you (or are nested under any post you wrote), either in general or in a particular thread.

There’s also no obvious way to sign up for notifications without posting a comment (I don’t know if there is a way or not).

writing unclearly baits ppl to ask you for stuff, like clarifications, which makes them seem weak and needy. i think it’s a widespread not-just-accidental tactic. partly people are bad at writing but partly there’s an incentive and manipulation there.

A little similar to some stuff I’ve said about being able to zoom your perspective in and out. It works well with idea trees: you should be able to see and think about the same issue at different levels of detail, and you can connect this to collapsing parts of trees, and to paying attention higher or lower in the tree (higher is lower level of detail). It’s important to be flexible instead of being stuck in one mode.

https://www.tiktok.com/@edwinpos1/video/7030944646222122246

There are lots of these videos of people hatefully quitting over a loudspeaker.

Many companies don’t seem to think being hated – by customers and employees both – matters. And then the public blames “capitalism”.

Being that stupid and incompetent is bad for profits. Capitalism is helping some by at least giving them an incentive (money) to do better. Some companies do better, like Costco, and are successful. Under socialism, a lot of bosses have no clear downsides for being shitty to the people under them.

Capitalism is not an automatic or full solution but it isn’t the problem here and helps some since being hated is actually bad for profit. This video is one of the many consequences. The consequences are not always immediate or obvious, especially when the government helps prevent competition (hello Comcast and health insurers), but there are consequences.

Related, Blizzard had so much good will with gamers and has been destroying it for years and they are still making money but 1) they could be making way more money if e.g. WoW subs were up instead of down 2) they are running their business into the ground and strongly, repeatedly alienating a lot of their biggest fans and promoters (who loved them enough to give third and fourth chances but many are now too fed up). This is not even close to how to make the most money over time and most of the bad things were bad for short term profit too.

Re Costco treating employees better and having a lot more employee loyalty and low turnover (and also being liked by its customers) … that came up recently when I was discussing with someone whether Bezos actually did a great job with Amazon or not. Like how much did Bezos earn his fortune? Was he actually a good leader of the company? In some ways probably yes. But tons of Amazon’s employees hate them and they have been burning a huge amount of good will from customers with the fake reviews, mislabelled Chinese knockoffs, etc. So one of my first thoughts about what I’d do as Amazon CEO was look into what Costco is doing so that its employees like it. Hire a bunch of managers from Costco and fix the problem where their warehouse workers and drivers hate you. That’s seriously important to the value of the business. (My knowledge of Costco is mostly be reputation. If I was really CEO I’d have to research them in way more detail first and also look at other companies. But the point is to find some companies that are able to get along with their employees while also being financially successful and copy that. Even if there is some flaw in the Costco example I’m confident that’s a thing which really does exist and Amazon is doing something seriously wrong that’s bad for business.)

1 Like

A while ago I read Edith Hamilton’s Mythology which I liked.

I recently read 48 Grimm’s Fairytales. I believe what I’m reading is based on early versions of the stories before the Grimm brothers added a bunch of changes. I stopped cuz they were getting kinda repetitive.

I’m now reading 1001 Arabian Nights. It’s a very long edition which I doubt I’ll finish.

Enduring stories provide an interesting perspective on what people are like and how they think. Some of the traits in the stories are similar to some of the traits I observe in people today. The stories help explain and illustrate some of people’s flaws (merits too). Maybe one day I’ll actually read some Bible or Koran or some Eastern books along those lines.

People think they’ve changed so much from these stories. And they have changed in some ways. Our world is less violent and superstitious. But people are still rash, angry, and have violent urges. They suppress that stuff more in a lot of circumstances but not all circumstances. They disapprove of it more in a lot of circumstances but not all. The main point is I think a lot of how people got more civilized is by suppressing some really bad behavior more than changing how they actually think and feel for their initial reactions. Also people seem less rash today but they still have rash initial reactions/thoughts/feelings, they just suppress those some and usually don’t do major actions immediately. Often they still do what they initially wanted, just later with some intellectual reasoning to justify it, but the reasoning was to rationalize the bias they had right away.

So far (3/4 through round 2) I don’t like these broadcasts as much as the St. Louis Chess broadcasts because they go into less depth about individual games. It’s a larger tournament but I’d prefer if they’d focus on the top boards or best games more instead of trying to cover every game some. I’m hoping they may focus more in later rounds when there are clear contenders to win the tournament, and therefore some key games.

Also, I like the St. Louis segments where the two main broadcasters go to Maurice Ashley who has been using a strong chess engine to find the right moves while off camera, and he goes over some of the right answers in interesting positions. The chess.com broadcast is making little use of computer analysis to reveal correct answers to some of the hard positions.

Also, St. Louis has better roles. Yasser Seirawan (despite being older, wiser, highly experienced) plays the role of the more naive person who wants to try things out. He asks some of the questions viewers would have and is the most relatable for viewers. Then Alejandro Ramirez or Ashley can act as his foil and give criticism of some of his experiments.

On the chess.com broadcasts, Anna Muzychuk is doing both roles: she’s the primary person analyzing good lines and giving answers about what works or not, and she’s the primary person experimenting with moves that might be bad, trying things out, asking questions, etc. She’s the more active person who talks the most and also does the most chess moves and has the most chess knowledge, while Stuart Conquest isn’t contributing as much.

It’s pretty standard on two-person broadcasts to have one person who is more social and one who is more of the expert. People use these roles on purpose even when they could play a different roll. E.g. I remember a different chess broadcast where a woman kept doing the social stuff but then when a guy left temporarily she immediately switched into acting more like a chess expert until he got back. So doing the more social side was an intentional choice/role, probably because she was better at it than the men, not because of lack of chess ability by her (I do think the men were stronger chess players, but she was plenty strong – far far stronger than the audience including me, and capable of playing an effective chess expert role).

Similarly, on Starcarft broadcasts, Tasteless intentionally plays dumber than he is and acts like more of a viewer representative and social/fun guy, while Artosis plays the expert role. The roles are intentional and help organize who says what kinds of things. Tasteless helps support Artosis’ expert reputation and tries to avoid challenging or undermining it. Tasteless will sometimes ask questions for Artosis to answer, even though he knows the answer and could answer it himself. Artosis is somewhat more of a smart expert than Tasteless, but they play it up and have some division of labor. (Possibly, over the years, they actually became more different due to playing different roles on broadcasts. Tasteless had less incentive to keep studying expert stuff, and more incentive to learn how to do his role well, e.g. getting better at making jokes, filling dead air time, or understanding and remembering to speak to the perspective of audience members who don’t know a lot about the game.)

I currently have 5 articles scheduled to post on the CF site. They will continue to be on Thursday and Sunday mornings for now.

1 Like

There’s something bad and unrealistic about this article. Why does Reisman think Fox is or should be good? They’re just shitty. Any network could improve. Why focus on Fox improving? Why not ignore Fox and seek progress elsewhere? I think Reisman overestimates Fox.

lol I was reviewing old exploratory writing to find something to edit and I found an 865 word paragraph.

That’s so long it’s funny, but then I read read the beginning to see what it was about:

structuring [knowledge] as many small parts with connections

I laughed at the irony.

(I knew the paragraph was way too long when I first wrote it. I just figured I could split it up during editing later. Sometimes I have ideas flowing and just write them down quickly without getting to an automatized paragraph break. Usually I separate out paragraphs while writing automatically but the automatic rules don’t cover every case. So it’d take some attention away from writing my ideas down to split the paragraph using conscious thought about how to handle it. In that case, it’s often better to do that conscious attention later instead of letting it distract from the topic I’m writing about.)

Sad seeing one of the world’s best players trying to play super fast chess games using software that’s worse than we had 20+ years ago.

He doesn’t have smartmove. Normally to specify a chess move you have to input two squares (where you’re moving from and to). This can be done with click and drag (as in the video) or two clicks.

Smartmove is a feature where you click on a single square and if you have exactly one legal move involving that square, the move is played based on just one click. So if you click one of your pieces and it can only move one place, it moves there. If you click an empty square or square with an enemy piece, and you have exactly one piece which can go there, then you play that move in one click.

If you do a single click on a square that works for 0 or 2+ legal moves then smartmove has no effect.

Smartmove is a significant convenience and speed increase, and he’s getting into many games where moving just slightly faster would make a significant difference that would help him win (or if his opponent had it too, it’d help both of them put on a better performance – play more moves faster. it’d help their hands keep up with their minds better.)

I think the problem is the newer chess websites force you to play with their shitty software in a web browser. And I think none of them support the wild 5 variant or various other chess variants that we also had 20+ years ago. Yet these newer chess websites are now dominant in terms of popularity and having strong players use them.

Old chess servers let you connect with telnet to play with an ASCII board, and anyone could write software with a graphical chess board which connects and plays moves over the simple protocol. The servers were designed with an open protocol to be compatible with whatever chess playing software anyone wanted to create, rather than forcing you to use a single piece of software.

I recall there was a chess interfaces named “Thief” because it purposefully copied/stole the best features from all the popular chess interfaces at the time it was made. This was, of course, a good thing. It was made by a strong player who played fast games. Both smartmove and premove were innovations that people came up with in I think the late 90’s and which ended up in multiple apps because they were useful. Thief had both though I don’t think it originated either. Premove is still around but somehow smartmove got forgotten or something. (Premove is making a move on your opponent’s turn. If it’s a legal move when it’s your turn, you play it instantly. In most positions, premoving has a risk of playing a very bad move because you can’t react to a threat your opponent just made. Premoving recaptures is routine because they aren’t a legal move unless your opponent plays the capture. Premoving is also common when you have only a few seconds left and there aren’t many pieces left. A common alternative to premove with lower risk is, on your opponent’s turn, dragging a piece to a square and keeping your mouse held down. Then as soon as you see his move you let go to move, unless he did something unexpected that makes your planned move really bad.)

EDIT: Here is someone 2 years ago asking for smartmove as a feature request at one of the current, popular chess websites: Smart Move • page 1/1 • Lichess Feedback • lichess.org (he got some dumb replies and seems to have been ignored).

1 Like

I don’t know about chess software specifically. But something I’ve noticed in general is tons of software and service development focus and effort has shifted away from empowering the user to do what they want and then (maybe) charging the user for that value. Instead, tons of effort now seems to go into nudging, dark-patterning, de-featuring, and outright coercing users into doing what you (the developer or, more often, the developer’s big-tech employer) want them to do so as to generate as much revenue as possible from some combination of direct user payments, advertisers, content owners, and venture capitalists.

A couple of areas I’ve especially noticed it in -
Music - 20 or so years ago Microsoft (and others, but I’m most familiar with Microsoft) were putting lots of development effort into library based music players. For Windows, this was Windows Media Player. The user owned the music (as files in an industry standard format). The software gradually got better and better at letting you do whatever you wanted with that music. The best feature was automated playlists with programmatic rules. When combined with individual, granular (star) ratings and tags for genre, sub-genre, mood, etc. it became possible to program an automated DJ that did exactly what you wanted, nothing more/nothing less. If it played a song you didn’t want to hear at that time, you didn’t just have to hit “skip” and hope some opaque and other-controlled algorithm insured it didn’t happen again. You could figure out why, and update the rules of the automated playlist so it didn’t happen again. If you heard a new song you liked on the radio, you could buy that song, tag it appropriately, and then it’d show up in all the correct automated playlists. You could keep running an old version of the software as long as you saw fit, and you could change player software itself (or even write your own) without re-buying all the media files, knowing all your music would play in the new software.

10-15 years ago Microsoft quit adding features to Windows Media Player. Nevertheless, it’s still what I almost exclusively use for playing music. I’ve expected any day to hear they are dropping it. Instead I recently heard they’re going to have a “new” version soon in Windows 11. I don’t have high hopes but we’ll see.

What seems to be getting most of the attention now is streaming services where as a user you don’t own the music and your control over what plays is limited. Sure, you can play an individual song or create a static playlist and you can thumb up/down songs. But (at least from what I’ve seen) you can’t exert the kind of absolute control over your music experience that a good library player like Windows Media has. Thumb up/down in particular is pretty useless for anything other than “I never want to hear this song again”. And you have to pay every month or you lose access to it all. And for everything non-static it gives the service an “in” to play you…whatever the service thinks will maximize its revenue. Which may or may not be what you actually want to hear at that time. And if you want to move to a different streaming service good luck taking your ratings (such as they are) with you - you have to start over trying to get enough metadata into that system to get it to behave the way you want. Also, maybe not all the music you like is there and if not it’s not practical to add it into your mixes. And the services can & do change their interfaces and features - as a user you have no control over if or when this happens.

The only advantages I see for users in the streaming services are not having to buy a bunch of music you like up front and not having to store & manage music files yourself. And I’m aware neither of those were issues for me by the time the streaming services came along but maybe they are significant issues for lots of people. So it’s entirely possible the streaming services work better than library players for the majority of people. But the lack of ownership and control in such services is also glaringly obvious and I don’t think it’s accidental.

Video - Netflix is the main example here although my impression of other video services is broadly similar. As a user you never owned videos at Netflix - something I found fine as (unlike music) I only rarely want to watch something multiple times, years apart.

When Netflix started it had two very cool user control features: A queue, which you could order as you wanted. And (like the music example) star ratings of what you’d previously watched (whether on Netflix or elsewhere). Once you’d rated enough videos, Netflix would get pretty good at estimating the star rating you’d give to other Netflix content you hadn’t yet rated. The algorithm was opaque, but at least it worked in so far as the predictions seemed reasonably accurate. You could then add a video to your queue (or not) and even put it in priority position based on the predicted rating. So when you had time to watch videos, you’d go to your queue and your highest priority items (Meaning: stuff you’re most likely to enjoy) would be right at the top.

Some years ago they did away with the star ratings in favor of (like the music streaming services) thumb up / down. Way less informational to their algorithm, and at the same time they dropped the predicted rating that made their algorithm particularly useful. Instead, they suggest shows based on general criteria like genres & popularity, along with some (opaque and in my experience not accurate) effect from what you’ve thumbed up or down.

But at least they kept the queue, and I could guess (not as well as the old algorithm, but better than random) how much I’d like a video and put it at the right place in the queue, and only watch stuff from my queue as opposed to what the service prompted on any given day. Lately they took even that feature away, removing the ability of a user to order their queue as they see fit in favor of an algorithmic queue ordering. So now what’s at the top is explicitly not what you as a Netflix user think you’re most likely to enjoy watching. What’s at the top is whatever the algorithm wants to nudge you to watch.

I’m not sure if, taken as a whole, the no-rating / no queue-ordering is better for Netflix’s subscriber retention than the rating system and self-ordered queue. I think it’s possible - maybe people want to be more passive about what they watch than I do. Or maybe not.

But I’m quite sure it gives their execs much more power over what people actually watch than the old system. It feels super manipulative, and my guess is that’s why they did it. I’d be surprised if it’s not heavily influenced by stuff like what agreements they have with content providers, what shows they’re trying to create “buzz” around, and other factors that I as a user do not want influencing which videos I watch. I have to keep my own list outside of Netflix to maintain the level of control over my video watching that I had under the old system, and near-zero people do that.

In both cases (music and video) I think it’d be easily possible to keep user control features intact while providing an “easier” algorithm-guided interface to users who didn’t want the control. But the developers (or their employers) seem to actively not want that. It seems like they’d explicitly rather manipulate users than give them control.

I don’t know enough to say whether this user-manipulation is economically productive or not. I just know as a user I notice it, find it super annoying, see it as a general industry trend and I’m skeptical about it being good in an economic sense.

2 Likes

Search results on a lot of websites are quite bad. I have often searched on Amazon for a book that I know exists with the title and author and it’s not among the search results. So I have to go to a non-google search engine to find the book on amazon.

Attention seeking is problematic in a deep way b/c error correction is attention. So errors (at all other goals) suffice for their meta goal.

Something that works particularly well, for getting attention, is errors at goals other people care about.