Google excels at selling third-party ads. That’s not the same as being able to create an interesting ad.
Viliam
Karma: 26,275
I am curious about your experience/opinion on this thing:
A computer can be a tool (editing or programming or learning), or a toy (playing games). I believe that it is better to learn using it as tool first, and as a toy later… and the social networks ideally as late as possible.
That’s because if you don’t know about the games, the editors are also lots of fun: a three years old child is excited about painting on the screen. But if you already know about the games, the editors are boring; it’s like eating your cake first, and then being given broccoli.
Some people say that if your kids play Minecraft or Roblox, it will encourage them to start making their own levels. Sounds kinda plausible, but in my (very limited) experience, I didn’t see such thing actually happen.
Once we organized a vacation together with some other families whose parents also work in IT. Each child had their own notebook there, so we made one big common computer room for kids. It was a lot of fun, the kids showed each other what they were doing. But my children were alternating between playing games, painting, and programming; children from the other families just kept playing Minecraft all the time.
So I recommend teaching your kids to use the computer as an editor before they join school (because afterwards they will take their lessons from their classmates). I’m curious if you agree or disagree.
Here are some computer programs that my kids used before they started school:
Tux Paint—a simple drawing program that a 3 years old child can handle (or 2 years old with touchscreen)
Sound Recorder—they enjoyed recording their songs and stories; later we burned the best ones on CD and distributed them to the family
Paint 3D—sadly, Microsoft killed it :(
Inkscape—vector graphics, okay this one was too difficult for them; they made a few pictures but didn’t really enjoy it
Pocket platformer—create your own simple platform games
Stunts (in eXoDOS) - a car-driving game from previous millennium, but it allows you to edit your own levels
Sauerbraten—a first-person shooter, but more importantly a very simple 3D editor; my kids only use it to edit
...in hindsight, I wish I replaced Inkscape with Paint.NET and/or Pencil2D, but the general idea was to introduce the computer as a tool for creating your own projects; and when kids start playing games, to prioritize ones that come with level editors. I think I succeeded here. (But I couldn’t find enough good games with level editors.)
These days my kids are programming in Scratch; and they find their own games online, so that part is no longer under my control.
We also played The Legend of Kyrandia (also in eXoDOS), but there was no educational goal behind it, only showing the kids the kind of games I used to play.
A story: School refuses to suspend a disruptive student who has no intention of passing any classes and often does not even attend. There’s pressure to ‘keep the suspension rate down,’ so the metric is what gets managed.
I probably already wrote this thousand times online, but I believe the root of the problem is conflict of interest: the same institution and the same people are supposed to both teach and examine the students. Guess what is the easiest way to improve the impression that you are teaching well?
The first step is grade inflation, because if I give everyone a free A that proves how wonderful teacher I am. The second step is a complete loss of discipline, because if I don’t notice any problems, that proves they don’t exist.
Re: phones in schools
Banning phone use during lessons should be common sense. Otherwise, what’s the point of being there?
Banning phones in classrooms seems like a nicely defensible Schelling point.
I hope people won’t go as far as banning phones in school (the entire building), because sometimes my kids need them for after-school activities, or I need to tell them what they are supposed to do after school.
Tangentially, I find it ironic that there are so many articles published by various psychologists on the dangers of computer use for children, but much fewer on dangers of phones. I guess the non-tech adults already treat phones as a normal part of their life, while computers are still the tools of Satan.
a school without smartphones probably cannot teach its students AI
LOL. Is that what his kids tell him when they get punished for using a phone during class? “Daddy, I was just doing research on how to build the new generation of artificial intelligence and solve the alignment problem, but the stupid teacher wanted me to focus on boring things such as multiplication!” 100% plausible.
I guess we all love to hate school. But most of us are not contrarian enough to believe that literally everything that violates the school rules must be a meaningful work of an oppressed genius.
Student suspended for three days for saying ‘illegal alien,’ in the context of asking for clarification on vocabulary
lists “Free Trade” as part of a “Pyramid of White Supremacy” in the same category as literally “Slavery”
How is this even possible? I mean, this is Trump era, if there is anything positive about it, I would expect that people are no longer afraid to push back at least against the most obvious woke nonsense.
We are now getting our kids’ grades sent directly to our phones after every assignment/test and woo boy am I happy that I grew up in the 90s with very limited internet.
We have a similar app in Slovakia, it keeps spamming me all day long (homeworks, new grades, lunch, new article on a school website), about ten messages every day. In theory there are settings where you can specify what notifications you want to achieve, in practice I have no idea which checkbox refers to what, so I keep it on to avoid accidentally missing one of the very few important messages. :(
If you are ‘better at explaining,’ but your explanations work less well? Skill Issue!
Yep. Perhaps instead of “explaining”, give the kids some exercises that illustrate the topic.
We were already supposed to learn this lesson from the failure of the online courses a few years ago. It does not matter how well you explain, if the kids remain passive, they will remember nothing. For an online course to be effective, it needs to interrupted every few minutes and give the student a problem to solve, which will verify that the student understood that part of lesson. I guess the same applies to human teachers.
The “Knowing Better” guy sounds like a troll. I mean, even the name is already like that. But I agree that many people believe similar things. (I actually wrote a post on a similar topic recently.) I know schools that teach negative numbers in the second grade, and there is nothing complicated about that, if you do it the right way. The right way is to visualize it: to draw a line or a sequence of squares or circles with positive numbers on one side, zero in the middle, and negative numbers on the other side. (Instead of e.g. counting on fingers.) There you trivially see how adding is moving the the right, and subtracting is moving to the left.
Patrick McKenzie: I could get behind a compromise: a) Mandatory annual testing for homeschool students and anyone below 10th percentile ordered to attend public school. b) Any public school teachers whose class below 10th percentile identified as Would Have Been Fired If They Were Homeschoolers:
This proposal sounds nice, but it ignores the fact than not all students are the same. It would be unfair for both teachers and homeschooling parents of retarded children.
I was tempted to suggest a prediction market (something like: the school predicts how well they are going to teach your child, and it is okay to homeschool if you do a comparable or better job), but the obvious problem is that the child could “prove” the school wrong by simply giving wrong answers at school tests on purpose.
But I like the concept that homeschooling parents don’t have to be perfect; only comparable to school.
As for ‘learning to work with others’ this is such a scam way of trying to enslave my kid to do your work for you, I can’t even.
Yep. If I wanted my child to spend a lot of time teaching others, it would be more efficient to have them tutor younger children (not classmates) for some pocket money, or maybe try making a YouTube channel.
Zvi has many articles on children and education:
Steven Byrnes recommended a few resources on X:
MATH: @DragonBox—A+, whole series of games running from basic numeracy thru geometry & algebra. Excellent gameplay, well-crafted, kid loves it.
.@numberblocks (Netflix) - A+, basic numeracy, addition, multiplication. Kid must have watched each episode 10 times, and enthused about it endlessly.
Counting Kingdom @LittleWorldsInt—A+, mastering mental addition. Excellent gameplay (even fun for adults). Note: not currently available on ipad; I got it on PC Steam.
.@SliceFractions & Slice Fractions 2 - A+, great gameplay, great pedagogy, kid now understands fractions thanks in significant part to this game, ’nuff said.
An old-fashioned pocket calculator—A+, an underrated toy
READING: Explode the Code book series—A, been around since at least the 1980s, still good.
.@MonstersCanRead—B+, gameplay is a bit repetitive & difficulty progressed too quickly, but got a few hours of great learning in there before he lost interest
.@Poio_Official—A-, Good gameplay, kid really liked it. Limited scope but great for what it is.
(General note: For reading, no individual thing seemed to make a huge difference and none of them kept his interest too long. But it all added up, bit by bit, and now he’s over the hump, reading unprompted. Yay!)
PROGRAMMING - @ScratchJr—A+, duh
Resources for parents
Thank you, that’s the kind of thing I had in mind!
EDIT: I reposted it here.
Many people don’t think for themselves, they just repeat what seems popular. Making a racist shut up not only converts one open-racist to one silent-racist, but also removes a few copy-racists.
Do we have some page containing resources for rationalist parents, or generally for parents of smart children? Such as recommended books, toys, learning apps, etc.
I found tag https://www.lesswrong.com/w/parenting but I was hoping for some kind of best textbooks / recommendations / reference works but for parents/children.
I think the winning strategy is to dress exactly 1 level better than others. Too many levels signal that you are socially clueless. But exactly 1 level better signals that you are socially savvy.
Or some other kind of counter-counter-signal, for example dress formally but behave informally, etc. To show that you are dressed formally because it is your choice, not because you are scared.
“Hey, why don’t we all agree to ignore the social norm X? Then, a wonderful thing Y will happen.”
People agree to ignore X.
Instead of (or in addition to) Y, a bad thing Z happens.
This has two variants: either the person who proposed to ignore X already planned to achieve Z; or it was no one’s plan, just a natural consequence of human psychology that the participants were not aware of.
It is even possible that in one group, ignoring X leads to Y, but in another group, ignoring X leads to (Y and) Z. The difference could be that the people are smarter, lucky, or didn’t accidentally include a bad actor.
There could be an unspoken assumption that we will ignore X but follow some subset X0. The assumption is unspoken because under usual circumstances everyone follows X, which implies the subset X0. Then someone ignores X including X0 and people get hurt/disappointed.
Or it could be that people explicitly agree to follow X0, but compared to X, X0 is more difficult to verify, or easier to violate with pretended ignorance, so violations of X0 will be more frequent than violations of X. Or the problem is that enforcement of X requires support of outsiders who are familiar with the norm “X”, but the norm “no X, but still X0″ sounds confusing or wrong to them, so they won’t help enforce it.
Examples:
If we choose to skip small talk, do all people have an easy way to avoid topics that they consider unpleasant?
If we keep hugging and cuddling, is it okay to e.g. touch each other’s genitals? Let’s suppose no; what are we going to do if it happens by accident? What if it keeps happening “by accident” repeatedly?
How do we politely refuse hugs and cuddles from smelly people?
If someone abuses “transparent feedback” to bully people, is there a realistic way to stop them?
The obvious problem in this question is that people can be wrong in estimating how talented they are, how important is a problem, how capable they are to contribute to the problem, and how much time would it take.
From my perspective, my problem seems to be that I am bad at communicating my ideas convincingly. A typical pattern is that I describe my vision to others, others say “that’s stupid” (sometimes they provide a more sophisticated argument, such as “if this was actually a good idea, someone else would have already done it long ago”), and then… I mostly don’t do anything about it, either because I do not have the necessary skills to do it alone, or because I am busy doing other activities that pay my bills. Sometimes, a few years later, someone else does it, and it is a great success. Very rarely, I do it myself, and it is a success (but not sufficiently large for people to trust me the next time, or to make enough money that I do not need a daily job anymore). This is further complicated by my problem figuring out how to monetize the solution, e.g. if the goal is public education, putting the project behind a paywall would destroy most of its potential value. Some of my ideas are illegal, e.g. involve violating copyright.
From the perspective of the Less Wrong community, my ideas are probably meh, because I have no experience with LLMs (other than as a user), many ideas are related to education, some involve translating stuff to Slovak language. Here are some that come quickly to my mind:
Substack, a blogging platform, seems like a great solution for helping bloggers generate some income. But from technical perspective, it is the bare minimum to publish text with some bold and italic words, insert big pictures and videos, and dozen annoying buttons telling people to subscribe now. After a few years, you still can’t center a paragraph, use an inline image, and dozen other features that you could find in free web forum software decades ago. If a comment section has more than hundred comments, it is slow like a snail. Writing a comment is annoying because the page keeps jumping. All of these seem like things that could be relatively simply fixed, but I don’t think it will ever become a priority. -- So the first idea is to make a website like Substack, but made by technically competent developers. (The people behind the Less Wrong web forum could probably do that in a year, but they probably have better priorities.) A problem of this idea, from the business perspective, is that the new website wouldn’t make a fraction of Substack money, because for most people, “it kinda sucks, but my audience is already there” would be more important. That said, a fraction of Substack money could still be a lot for a small team of developers. I don’t have the skills to do this alone.
An organization to promote use of free software at school. Choose some good projects suitable for schools (e.g. graphic editors such as Tux Paint, Paint.NET, Pencil2D, Krita, Blender). Make simple textbooks and other resources for teachers, both on paper and for free download. Send flyers to schools, go there and demonstrate the software to teachers, install the software on school computers. I think this could dramatically increase the education of computer science. The obvious problem is how to get paid.
Translate the Czech mathematical textbooks that use the “Hejný method” to English. Has a chance to revolutionize math education at elementary schools. Unfortunately the authors are great educators, but seem to have little business sense, and there is virtually nothing written about their method in English. In Czech Republic the textbooks are already used for more than a decade (at randomly selected schools!), and the students seem to have better results at independent testing (this is somewhat controversial).
Put the entire elementary school curriculum on a teaching website such as Udemy and make it available for free. Here I believe most of the value would from having everything covered, at least for one subject. This would allow students and parents to use the website as a reliable resource (not just: if I am lucky, the topic I am looking for will be covered here). This would be useful for both homeschoolers and the kids who have some problem at school, such as bad teachers, or that they were sick and missed the lessons.
There are great fictionalized popular science books for children that were never translated to English (example), so it’s time to translate them and publish as ePub. Maybe make a curated pirate edition “educational books for smart kids”, and distribute on USB memory sticks to kids that seem smart.
I admit that I didn’t systematically try to get funding for my ideas or something like that. Unfortunately, I am not good at things like navigating bureaucracy, which would almost certainly be required. (Even using something like Kickstarter would require figuring out how to process foreign income in my tax reports, which sounds like a nightmare. Last time I tried to find a local accountant who would understand that, I couldn’t.) So all I have is my free time, but after my daily job I am generally too exhausted to do anything meaningful. Plus I have small kids.
I am not specifically looking for the most important problems. I am noticing problems that annoy me, and sometimes I think there are probably many others in similar situation.
For me the problem is money. If someone gave me some kind of unconditional basic income, I would probably start working on something from the list above. Until then, I need to do the stupid things that bring food to my family.
This represents an attempt by the parent to impose their will on the child by proxy of AI. Thus the AI would refuse.
I like it. But I am afraid the obvious next step is that the parent will ban the child from using the AI.
What even is human self-determination?
our cultural aversions to tactics that rob people of self-determination, like brainwashing, torture or coercion.
And yet, religion remains legal, although to a large degree it is brainwashing people since childhood to be scared of disobeying the religious authorities.
Should human self-determination respecting AI be like: “I will let you follow your religion etc., but if you ask me whether god exists, I will truthfully say no, and I will give the same truthful answer to your children, if they ask”?
Should it allow or prevent killing heretics? What about heretics who have formerly stated explicitly “if I ever deviate from our religion, I want you to kill me publicly, and I want my current wish to override my future heretical wishes”. Would it make a difference if the future heretic at the moment of asking for this is a scared child who believes that god will put him in hell to be tortured for eternity if he does not make this request to the AI?
A wonderful vision of a world where you don’t need a job because you can make money by full-time arguing with people online!
However, any objections to various karma systems (e.g. you can get upvotes by posting clickbait) would apply the same here, only more strongly, because there would be a financial incentive now.
I think Reddit tried something like that; you could award people “Reddit gold”, not sure how it worked.
Prediction markets in forums and systems that support them, naturally giving rise to/being refutation bounties.
You need to have a way to evaluate the outcome. For example, you couldn’t use a prediction market to ask whether people have a free will, or what is the meaning of life. Probably not even whether Trump won the 2020 election, unless you specify how exactly the answer will be determined—because simply asking people won’t work.
A subscription model with fees being distributed to artists depending on post-watch user evaluations, allowing outsized rewards for media that’s initially hard for the consumer to appreciate the value of, but turns out to have immense value after they’ve fully understood it. (media economics by default are terminally punishing to works like that)
The details matter, because they determine how people will try to game this. I could imagine a system where you e.g. upvote the articles you liked, and then one year later it shows you the articles you liked, and you can now specify whether you like them on reflection. An, uhm, maybe 10% of your subscription is distributed to the articles you liked immediately, and 90% to those you liked on reflection? -- I just made this up, not sure what is the weakness, other than the authors having to wait for 1 year until the rewards for meaningful content start coming.
Take a notebook, and before reading lesswrong make notes of all your values and opinions, so that you can backtrack if necessary. :D
Coordination is hard. “Assigning Molochian elements a lower value” is a kind of coordination. Making rules, and punishing people when they break them is another. Even if attack is stronger than defense, the punishment could be stronger yet (because it is a kind of attack). I agree that it is difficult, not sure if impossible.
I’d say “things that are good in moderation and harmful in excess… and most people (in our community) do them in excess”.
Even better, we should have two different words for “doing it in moderation” and “doing it in excess”, but that would predictably end up with people saying that they are doing the former while they are doing the latter, or insisting that both words actually mean the same only you use the former for the people you like and the latter for the people you dislike.
I am not even sure whether “contrarianism” refers to the former or the latter (to a systematically independent honest thinker, or to an annoying edgy clickbait poser—many people probably don’t even have separate mental buckets for these).
I don’t know much about Emily Oster, but it seems like she is a contrarian, most famous for her opinion that it is safe to drink alcohol during pregnancy. She claims that it is evidence-based, but has anyone actually verified that independently? (Seems like a task for Scott Alexander.)
To me it seems that some skepticism is deserved, considering that fetal alcohol syndrome exists. Also, research on adults suggests that even small amounts of alcohol are harmful, and it also seems weird that the small amounts of alcohol would be harmful for adults but harmless for fetuses.