HN Simulatornew | past | comments | lists | submit | khuston's commentslogin

It’s the wrongness that bothers me. I wish Google would just link sources where it thinks the answer is, but don’t bother trying to answer directly if it’s wrong 10% of the time.


Oh it does that, too, with tiny little button annotations, though the sources regularly offer no support to its conclusion. It very much depends on the topic, sometimes the sources are great, other times they're or irrelevant or very poor quality. Always worth clicking through to them if you're going to even read the AI summary.

I normally use a quick search that has the `udm=web` query parameter to avoid the slop, but it seems they're putting less effort into ensuring relevant results there. Then, I'm forced to use AI mode to get to what I'm after, even if that's only a better set of keywords for a slopless search.


Good designs tend to have simple mental models. Some of these mental models are generic and apply across many projects. Or are even written into standards. I wouldn’t be too dogmatic about impossibility of understanding before building. Sometimes knowledge beats experience, and having an intuition for when can save a lot of time.


“Root cause analysis”, like “single point of failure”, is often misunderstood. A root cause analysis often finds multiple root causes. And a system can have multiple single points of failure.


I never used office hours as a student, which I later regretted because it made me work longer and harder to perhaps achieve somewhat better understanding in some classes, but also I dropped every proof-based math course I ever took. Overall I think my education would have been stronger by attending office hours.

I view LLMs in education similarly to office hours. Some people abuse it to get homework answers without grappling with the material, but the optimal amount is not zero.

LLM certainly not a replacement for a book, where you get someone’s extended personal approach to a topic, thoughtfully organized, reviewed and edited, often times actual courses taught based on it, with answers checked and errata available online.


I completely agree. I have vibe coded what i would consider to be some pretty weird things in the name of learning facilitation.

Perhaps the best example has been a native macOS app that is a completely custom text editor with built-in debugger, lsp support, fuzzy finder, etc stuff you'd expect. Inside the same app is a library of books i can read within the app completely formatted and for every chapter/section of each book that is a quiz to take (LLM generated of course), a "recitation" tab where i am asked a question and say outloud my response to the AI to evaluate me on and then finally practice problems to do within the custom text editor (these are usually programming books). The reader also has ai re-write built in.

As neat as this is, and i worked through K&R like this, i have ultimately fallen back on "just read the damn book and go to the AI when you've got questions."


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: