hathawsh

Several months ago, just for fun, I asked Claude (the web site, not Claude Code) to build a web page with a little animated cannon that shoots at the mouse cursor with a ballistic trajectory. It built the page in seconds, but the aim was incorrect; it always shot too low. I told it the aim was off. It still got it wrong. I prompted it several times to try to correct it, but it never got it right. In fact, the web page started to break and Claude was introducing nasty bugs.

More recently, I tried the same experiment, again with Claude. I used the exact same prompt. This time, the aim was exactly correct. Instead of spending my time trying to correct it, I was able to ask it to add features. I've spent more time writing this comment on HN than I spent optimizing this toy. https://claude.ai/public/artifacts/d7f1c13c-2423-4f03-9fc4-8...

My point is that AI-assisted coding has improved dramatically in the past few months. I don't know whether it can reason deeply about things, but it can certainly imitate a human who reasons deeply. I've never seen any technology improve at this rate.

That sounds like the recommended approach. However, there's one more thing I often do: whenever Claude Code and I complete a task that didn't go well at first, I ask CC what it learned, and then I tell it to write down what it learned for the future. It's hard to believe how much better CC has become since I started doing that. I ask it to write dozens of unit tests and it just does. Nearly perfectly. It's insane.

Good to know. Now I think I know why micropayments for news media never took off: because people who want to read news media probably don't want to waste mental cycles on keeping track of a micropayments account. They want a set-and-forget solution with a predictable cost. If micropayments can't fit those expectations, then the market probably wants something other than the thing we're calling micropayments.

Please help me understand better, because it feels like part of the problem has already been solved. Specifically, I've been told that the independent journalists that I watch on YouTube Premium receive a portion of my subscription fee. Is that not a form of micropayments? The system seems to work well enough for videos. Isn't there some way to adapt that kind of system to other media?

Python has an "email object" that you should definitely use if you're going to parse email messages in any way.

https://docs.python.org/3/library/email.message.html

I imagine other languages have similar libraries. I would say static typing in scripting languages has arrived and is here to stay. It's a huge benefit for large code bases.

Ha ha, well that's a relief. I thought the article was going to say that enabling TCP_NODELAY is causing problems in distributed systems. I am one of those people who just turn on TCP_NODELAY and never look back because it solves problems instantly and the downsides seem minimal. Fortunately, the article is on my side. Just enable TCP_NODELAY if you think it's a good idea. It apparently doesn't break anything in general.

Erasure coding is an interesting topic for me. I've run some calculations on the theoretical longevity of digital storage. If you assume that today's technology is close to what we'll be using for a long time, then cross-device erasure coding wins, statistically. However, if you factor in the current exponential rate of technological development, simply making lots of copies and hoping for price reductions over the next few years turns out to be a winning strategy, as long as you don't have vendor lock-in. In other words, I think you're making great choices.

OTOH, all that data is built on patterns that evolved from many years of evolution, so I think the LLM benefits from that evolution also.

> as well as an axis for rounded terminals (as in terminals in letters, not command-line apps).

Now I want to see a rounded terminal (as in command-line apps, not terminals in letters.) Would I type in a circle? Sounds cool.

I'll take a stab at this: LLMs currently seem to be rather good at details, but they seem to struggle greatly with the overall picture, in every subject.

- If I want Claude Code to write some specific code, it often handles the task admirably, but if I'm not sure what should be written, consulting Claude takes a lot of time and doesn't yield much insight, where as 2 minutes with a human is 100x more valuable.

- I asked ChatGPT about some political event. It mirrored the mainstream press. After I reminded it of some obvious facts that revealed a mainstream bias, it agreed with me that its initial answer was wrong.

These experiences and others serve to remind me that current LLMs are mostly just advanced search engines. They work especially well on code because there is a lot of reasonably good code (and tutorials) out there to train on. LLMs are a lot less effective on intellectual tasks that humans haven't already written and published about.

Are you sure? The third section of each review lists the “Most prescient” and “Most wrong” comments. That sounds exactly like what you're looking for. For example, on the "Kickstarter is Debt" article, here is the LLM's analysis of the most prescient comment. The analysis seems accurate and helpful to me.

https://karpathy.ai/hncapsule/2015-12-03/index.html#article-...

  phire

  > “Oculus might end up being the most successful product/company to be kickstarted… > Product wise, Pebble is the most successful so far… Right now they are up to major version 4 of their product. Long term, I don't think they will be more successful than Oculus.”

  With hindsight:

  Oculus became the backbone of Meta’s VR push, spawning the Rift/Quest series and a multi‑billion‑dollar strategic bet.
  Pebble, despite early success, was shut down and absorbed by Fitbit barely a year after this thread.

  That’s an excellent call on the relative trajectories of the two flagship Kickstarter hardware companies.

I think your intuition matches mine. When I try to apply Claude Code to a large code base, it spends a long time looking through the code and then it suggests something incorrect or unhelpful. It's rarely worth the trouble.

When I give AI a smaller or more focused project, it's magical. I've been using Claude Code to write code for ESP32 projects and it's really impressive. OTOH, it failed to tell me about a standard device driver I could be using instead of a community device driver I found. I think any human who works on ESP-IDF projects would have pointed that out.

AI's failings are always a little weird.

The article suggests that because the power and cooling are customized, it would take a ton of effort to run the new AI servers in a home environment, but I'm skeptical of that. Home-level power and cooling are not difficult these days. I think when the next generation of AI hardware comes out (in 3-5 years), there will be a large supply of used AI hardware that we'll probably be able to repurpose. Maybe we'll sell them as parts. It won't be plug-and-play at first, but companies will spring up to figure it out.

If not, what would these AI companies do with the huge supply of hardware they're going to want to get rid of? I think a secondary market is sure to appear.

Yeah, that was a little joke I slipped in, but OTOH, I've seen headphone kits that used 3D printed parts and it's not difficult to imagine someone replacing the 3D printed parts with handcrafted wood.

While I agree with you in principle (people should do their own thinking if they want gifts to be genuine), I thought I would go ahead and see how well Gemini can advise someone on choosing a gift for you.

https://gemini.google.com/share/88b694a09a89

The advice seems very good. What do you think? A donation to EFF or an open source project, a rare book, or handcrafted headphones seem like a good start for someone who can't afford anything extravagant.

I usually want to cut pies into 14 pieces. Some might want 11 or 13. (17 is just too many.) I demand that we implement a system where a circle is 2 * 3 * 4 * 5 * 7 * 3 * 11 * 13 = 360360 degrees, so that we can cut pies evenly at anywhere from 2 to 15 slices. If my baker cuts a slice at 25739 degrees, I want a refund! (I'll keep the pie, because the pie is obviously useless.)

(720720 might be OK too so we can cut 16 pieces, but honestly, if you're cutting 16 pieces, you're not going to measure. You're just going to divide pieces in half until you have 16. 360360 is the future.)

I've seen similar results in physics. I suspect LLMs are capable of redirecting the user accurately when there have been long discussions on the web about that topic. When an LLM can pattern-match on whole discussions, it becomes a next-level search engine.

Next, I hope we can somehow get LLMs to distinguish between reliable and less-reliable results.

Most of HN has probably seen this gem about "thinking meat", but in case you haven't: https://www.mit.edu/people/dpolicar/writing/prose/text/think...

Very insightful and very weird. It's impractical now, but it's a glimpse into some significant part of our future. I can imagine an app called The Last Game, which morphs itself into any game you might want to play. "Let's play 3-D chess like Star Trek: TNG. You should play as Counselor Troi."

(I also just thought of that episode about Moriarty, a Holodeck character, taking over the ship by tricking the crew. It doesn't seem quite so far-fetched anymore!)

The "Wham Baam" YouTube channels have a running joke about Hondas bumping into other cars with concerning frequency.

Consider these thoughts: https://wiki.c2.com/?SoftwareIsArt

I think you're headed in a helpful direction, but I'm looking for ways to narrow the phenomenon a little more. For example, yesterday I heard from my mom, who is not into technical things, that a lot of the Internet was down. She had heard it on the news. I didn't believe it at first because that information was surprising and clearly targeted at laypeople, but soon I learned it was true: AWS us-east-1 had major issues. So my doubt was unfounded. I'd like my doubts to be more accurate.

A few years ago, I plugged a single 100W solar panel into a battery pack that advertised it accepts 18 volts. I left it plugged in, on my deck, for hours. When I came back, there was a foul smell and some parts of the battery pack had turned black, apparently from getting charred. The battery pack no longer worked at all. I was very fortunate that only the control circuit had been destroyed and the battery cells (and my deck) had not been touched.

Lesson learned: don't skimp on Li-ion battery packs!

Also, I have a question about this article. Don't EcoFlow battery packs have a circuit that checks the incoming voltage and automatically shuts off charging if the voltage is too high? I would also expect a loud alert.

Good point. I suppose he has addressed this stuff already. I wish every city planner would watch his videos.

I really hope the Not Just Bikes YouTube channel takes a look at this. I love the analysis they do there.

In case anyone else has never heard of POSIWID: https://en.wikipedia.org/wiki/The_purpose_of_a_system_is_wha...

That's an interesting question that hasn't been tested yet. I suspect we won't be able to answer the question clearly until something bad happens and people go to court (sadly.) Also IANAL.

Here is what Illinois says:

https://idfpr.illinois.gov/content/dam/soi/en/web/idfpr/news...

I get the impression that it is now illegal in Illinois to claim that an AI chatbot can take the place of a licensed therapist or counselor. That doesn't mean people can't do what they want with AI. It only means that counseling services can't offer AI as a cheaper replacement for a real person.

Am I wrong? This sounds good to me.

My point exactly. Thanks.

Perhaps. Intelligent beings are always more skilled in some domains than others. I don't know why AGI would be an exception to that rule.