I see what you mean, but the problem is that the LLM provider is trying to provide all the value from the book to the user without the user needing to look at the book at all. I agree if the LLM fails to do so then there is a market for the book. But the LLM provider is trying to minimize that as much as possible. And if the LLM succeeds at providing all the value of the book to the user, without providing any value to the book creator, then in the future there is no incentive to create the book at all, at which point the LLM has no value to provide, etc etc etc.
Maybe I am wrong about this but I think a lot of recent research has shown that trial and error is a great way to learn almost everything. Even just making an educated guess, even if it is completely wrong, before learning something makes it much more likely that you remember and understand the thing that you learn. It’s a painful and time-consuming way to learn. But very effective.
Maybe Linux commands is a little different but I kinda doubt it. Errors and feedback are the way to learn, as long as you can endure the pain of getting to the correct result.
Needs qualification. Research shows trial and error learning is very durable, but it’s not the most time efficient (in fact it’s relatively poor, usually, on that front). The two concepts are a bit different. Yes, trial and error engages more of the brain and provides a degree of difficulty that can sometimes be helpful in making the concepts sticky, but well designed teaching coupled with meaningful and appropriately difficult retrieval and practice is better on most axes. When possible… good teaching often needs refinement. And you’d be surprised how many educators know very little about the neuroscience of learning!
At the university level in the US, few faculty get any kind of training before they are expected to start teaching. And the teaching requirement is more or less “do no harm.” If you’re at a research university, which includes many publicly funded universities, then your career trajectory is based almost exclusively on your research output. I could go on, but it suffices to say that it’s not surprising that the teaching could be better.
That said, most institutions have teacher training resources for faculty. I was fortunate to be able to work intensely with a mentor for a summer, and it improved my teaching dramatically. Still, teaching is hard. Students sometimes know—but often don’t know—what is best for their learning. It’s easy to conflate student satisfaction with teaching effectiveness. The former is definitely an important ingredient, but there’s a lot more to it, and a really effective teacher knows when to employ tools (eg quizzes) that students really do not like.
I am frequently amused by the thought that here we have a bunch of people who have paid tons of money, set aside a significant fraction of their time, and nominally want to learn a subject that they signed up for; and yet, they still won’t sit down and actually do the reading unless they are going to be quizzes on it.
> the thought that here we have a bunch of people who have paid tons of money, set aside a significant fraction of their time, and nominally want to learn a subject that they signed up for; and yet, they still won’t sit down and actually do the reading unless they are going to be quizzes on it.
How often have they put down the money, as opposed to their parents?
How often do they actually care about learning the subject, as opposed to be able to credibly represent (e.g. to employers) that they have learned the subject?
How often is the nominally set-aside time actually an inconvenience? (Generally, they would either be at leisure or at the kind of unskilled work their parents would be disappointed by, right?) My recollection of university is that there was hardly any actual obligation to spend the time on anything specific aside from exams and midterms, as long as you were figuring out some way or other to do well enough on those.
I suppose I should have said “nominally want to learn” etc, but I think you are right: most students simply want the credential. I maintain that this is still a strange attitude, since at some point, some employer is going to ask you to do some skilled work in exchange for money. If you can’t do the work, you are not worth the money, credentials be damned. On the other hand, I routinely see unqualified people making a hash out of things and nobody really seems to care. Maybe the trick is not to be noticably bad at your job. Still, this all strikes me as a bad way to live when learning and doing good work is both interesting and enjoyable.
Trial and error is necessary and beneficial, but not after the student becomes frustrated or anxious/bewildered by the complexity. The research shows that striking a balance between teacher intervention and trial and error is the optimal approach. If a teacher notices that a student is way off course but they keep persisting in one branch of the trial-and-error search space, it’ll be best if they intervene and put the student on the right branch. The student can still use the knowledge of what wasn’t working to find the solution on the right branch, but just persisting would be ineffective.
Gaining true understanding/insight is necessarily trial and error. Teachers cannot teach insight. But they can present the optimal path to gain insight.
Trial and error was the root of what became my IT career. I became curious about what each executable did from DOS and with that did my first tweaking of autoexec.bat and config.sys to maximise memory.
Years later I was the only one who could investigate network (and some other) problems in Windows via the command line while I was the junior of the team. Ended up being the driver of several new ways of working for the department and company.
Ditto. I found that people whose attitude was “let’s just try it” tended to be a lot more capable and effective. Nevertheless the prevailing wisdom when I was in IT was that if you had a problem that didn’t have an obvious solution, you had to purchase the solution.
Honestly with how good OpenCode is, this really just makes GitHub copilot the best subscription for the average user. It’s the cheapest. It’s free for students. You get access to all of OpenAI models AND Anthropic models AND Gemini models and you still have a pretty dang good CLI/TUI (OC, not Copilot CLI). And the limits are pretty reasonable. I’ve never hit the limits in a month though admittedly I am not a “five agents at once” kind of vibe coder.
This is fun! I thought I did pretty good with 123.
I found a path from Books to Rainbow with 123 generated topics!
Books → Johannes Gutenberg (people)
Johannes Gutenberg → Printing press (broader)
Printing press → Sedition (evil)
Sedition → Peaceful protest (good)
Peaceful protest → Social movement (similar)
Social movement → Black Lives Matter (places)
Black Lives Matter → Alicia Garza (people)
Alicia Garza → Reproductive rights (similar)
Reproductive rights → Feminism (broader)
Feminism → Women's studies (similar)
Women's studies → Gender studies (similar)
Gender studies → Queer Studies (deeper)
Queer Studies → LGBTQIA Studies (similar)
LGBTQIA Studies → LGBTQIA History (good)
LGBTQIA History → LGBTQ movements (broader)
LGBTQ movements → Galden LGBTQ Center (places)
Galden LGBTQ Center → LGBTQ community organizations (similar)
LGBTQ community organizations → LGBTQ advocacy groups (similar)
LGBTQ advocacy groups → GLAAD (places)
GLAAD → Gay rights movement (broader)
Gay rights movement → LGBTQ rights movement (deeper)
LGBTQ rights movement → Pride flag (deeper)
Pride flag → Rainbow flag (similar).
Another cool tool that’s being developed for rust is verus. It’s not the same as Kani and is more of a fork of the rust compiler but it lets you do some cool verification proofs combined with the z3 SMT solver. It’s really a cool system for verified programs.
I had a look, and it seems cool. But it doesn't seem to do the thing I love about Kani: work with only very partial annotations, only proving the annotations and being a very light lift on code not relevant for the annotations.
I can't disclose that, but what I can say is no one at my company writes Lean yet. I'm basically experimenting with formalizing in Lean stuff I normally do in other languages, and getting results exciting enough I hope to trigger adoption internally. But this is bigger than any single company!
This seems really interesting to me as I don’t often work in domains that require me to know a lot of facts, but I still feel like SRS could be useful. I just don’t quite know how to use it. Could you give me an example of what you mean here? What kind of decisions do you find meaningful to periodically reflect on?
Which really makes you wonder how well the system is really working. Of course I don’t know this but I feel like if you asked everyone, the majority of people would say that 95, or 90, or 85 is too old to be in congress. But somehow they keep getting reelected…
In my mind reading is more similar to thinking than watching. I have no basis for this but it just feels more mentally active. Of course it could just be my biases but I feel it is much easier to passively watch or listen to something rather than to read. But also I would say from my own experience writing and speaking promote “neurological health” even more so maybe the method of consumption is not as important as long as there is sufficient synthesis and thought on the other end.
reply