mindcrime

Solar panels? You can go down to Harbor Freight right now and pick up a setup that will run a laptop or tablet for a while, for a couple of hundred dollars. One wouldn't have to spend a ton more to get a setup that can run a decent PC with a GPU for at least a little while.

I mean, you're probably not going to run it all day, especially not in the part of year with fewer daylight hours, but anything's better than nothing. Plus the same solar setup can be used to power other useful things.

"They only do as a social contract, as long as power structures agree to enforce them."

is absolutely true. I've said many time "The Constitution is 'only a piece of paper' at the end of the day." Ultimately it's down to us, "we the people" to enforce this stuff.

Unfortunately the culture in the USA these days is sick. And not just a little bit sick... We are truly staring into the abyss right now.

Trump is dumber than box of brick bats.

You have sinned.

Your penance:

Three Hail Claude's, one Our GPT and an act of loop engineering.

Two minutes of silent prompting.

Disclaimer: I am not an economist, or a sociologist, etc. So this will be, at most, some random guy playing around with models with some help from an AI. And that's even if I find time to actually put into it.

I don't know. I was at my doctor's office a few weeks ago, and the topic of AI came up during our conversation[1]. I expressed the opinion that her profession should be fairly safe because it's so highly regulated and how doctors have the AMA to block new entrants. She said that her and her colleagues had literally just been talking about how they are worried about their jobs, and pointed out that with profit-obsessed private equity firms taking over more and more medical practices, she thought they very much would be trying to replace human doctors with AI. And she alluded to the AMA being a much less effective lobbying body than what I, as an outsider to the field, had perhaps believed.

I dunno. That's just an n=1 anecdote, but if nothing else it shows that (at least some) doctors are not just sitting back and resting easy, assured of a future of high paying career options. And this is a lady with a degree from Yale, who is a Doctor of Internal Medicine, and working for a very large and prestigious medical practice. Not some half-educated nitwit at a random back-country medical practice.

[1]: I mentioned that I'd asked ChatGPT about some test results, and told her what it told me. She replied "Yeah, I agree with that." Then chuckled and said something like "God, did I just agree with ChatGPT?" That's what started the whole sidebar about AI.

I mean, it can be that way. But it does't have to be. The Harness can be as much the brain as the LLM. You have nearly unlimited flexibility in what you do in the harness, and you can add as much of the overall "intelligence" of the system as you want in the harness.

To illustrate, you can:

1. Use multiple "brains" / LLM's, and do routing between them using a specialized model (a Jev like model, or something of your own design)

2. Use planning tools like pyperplan, jupyddl, pddl4j, etc.

3. Implement your own "memory" system using... pretty much anything you want. And I would argue that a lot of the apparent "intelligence" of a system involves how well memory works

4. Call classical machine learning models to solve specialized problems

5. Use reinforcement learning to improve performance in specific areas

6. Implement a blackboard for multiple knowledge sources to collaborate.

7. Use inference engines using forward/backward chaining, RETE, etc.

8. Use your own hand-rolled abductive inference strategy

9. Implement your own Bayesian reasoning techniques

10. Use custom causal inference code

... and so on.

Don't get me wrong. I'm bullish on what LLM's can do, but I posit that the overall expressed intelligence of a system can involve a lot more than just the LLM. In fact, I'd actually make the slightly stronger claim that, depending on what you're trying to do, an LLM might not even be required at all.

I do have it configured to use the Github MCP server and have given it access to 3 or 4 of my repos. Along with that, I put an AGENTS.md file in each repo and gave ChatGPT in general a custom instruction that says "at the beginning of each session, or when prompted by the user, consult the appropriate AGENTS.md file and follow the instructions found there."

I then keep various files in the various repos that catalog our discussions, decisions, speculations about future research directions, yadda, yadda, yadda. And I "incorporate by reference" by mentioning those files in the AGENTS.md. Setting all that up took some work, but it makes the AI much more reliable than trusting it to rely on its "memory".

1. I've been doing a lot of experimenting with neuromorphic and analog computing. ChatGPT has been my research assistant / lab partner throughout and has helped with doing experiment write-ups, analyzing scope waveforms (it's really impressive how good it is at parsing screnshots from the scope, using the graticule / divisions, and other on-screen indicators to figure things out), finding reference material, and providing general electronics advice. I've learned a TON of new stuff about using my oscilloscope, using some features I'd never used before.[1]

2. Along with that, I'd long wanted to get my lab all set up to use LXI for remotely controlling all (well, "most") of my test equipment. Once I did the work of physically connecting all of the instruments to the LAN, ChatGPT wrote most of the first cut of a batch of Python programs using PyVISA to control the various instruments (oscilloscope, function generator, power supply, multimeter, programmable load) and to download data from them.

3. I'd been "meaning to learn KiCAD one day" for about 15 years. About a month ago I said something to that effect to ChatGPT and it handily replied "I can help you with that." About 6 or 7 hours later I was sending Gerbers to JLCPCB with my first custom PCB design, and 4 days later I had my first boards in my hand. Soldered one up and it worked first time! Yay. Now granted, the circuit was VERY simple (a 555 timer blinking an LED) but the point was to learn the tool and the workflow, not the actual end product. About two weeks later I started on my second board, planning to do it with much less AI assistance (I worry a lot about cognitive abdication / brain-rot / etc). But right out of the gate, I ran into the point that KiCAD did not have a symbol for an obsolete component I was using (the 2N6027 PUT). So I turned to ChatGPT again and it helped me create a custom symbol. After that little diversion, I went back to the schematic and board layout. This time around, it only took about 85 "turns" with the AI to get to "Gerbers sent to JLCPCB" whereas the first time it took about 120 "turns". So I'm feeling pretty good that I am actually learning and retaining some of this knowledge and not "just following instructions from the AI". I do hope to get to the point soon of doing a board with NO assistance from the AI whatsoever.

Those are the ones that stand to me. There's other stuff of course, but those are ones that have really made a difference for me and/or really impressed me.

[1]: One thing I did that helped a lot, was creating a "Project" and uploading all of the user manuals and programming guides for my various instruments. So the GPT always has access to the actual documentation and doesn't have to rely on its "memory".

(d) Cellular telecommunications receivers (1) Within 180 days after October 28, 1992, the Commission shall prescribe and make effective regulations denying equipment authorization (under part 15 of title 47, Code of Federal Regulations, or any other part of that title) for any scanning receiver that is capable of— (A) receiving transmissions in the frequencies allocated to the domestic cellular radio telecommunications service, (B) readily being altered by the user to receive transmissions in such frequencies, or (C) being equipped with decoders that convert digital cellular transmissions to analog voice audio. (2) Beginning 1 year after the effective date of the regulations adopted pursuant to paragraph (1), no receiver having the capabilities described in subparagraph (A), (B), or (C) of paragraph (1), as such capabilities are defined in such regulations, shall be manufactured in the United States or imported for use in the United States.

That said, the law is complicated and there could be some other pertinent statute that I'm not aware of.

EDIT: OK, this gets even weirder. I haven't followed every line of this through in detail, so take this with the amount of salt you consider appropriate. But according to ChatGPT, the FCC, in this order:

https://docs.fcc.gov/public/attachments/FCC-25-85A1.pdf?utm_...

effectively repealed this restriction. What makes it weird is that the statute itself is apparently still on the books, but the FCC seems to have exercised their rule-making authority to say "this no longer applies". Here's what appears to be the relevant sections:

We have identified 36 rule provisions, totaling 12,008 words and covering more than 25 pages in the Code of Federal Regulations, that no longer serve the public interest because they regulate obsolete technology,4 are no longer used in practice by the FCC, industry, or the public,5 or are otherwise outdated or unnecessary.6 Applying the “good cause” standard discussed above, we conclude that prior notice and comment are unnecessary before repealing the rules identified in the Appendix.

and contained within one of those footnotes:

15.121 (contains provisions to protect consumers’ privacy from technology that is now obsolete);

[1]: https://www.govinfo.gov/content/pkg/USCODE-2004-title47/pdf/...

1. Since people are still going to find a way to buy items the FCC doesn't want them to have, what's that new outlet going to be? I saw somebody mention TikTok. OK... what else? Facebook groups? Craigslist? Private old-skool phpbb forums? Usenet? Mastodon? AliExpress? Something that hasn't been invented yet?

1. Decrypting those communications if they are encrypted

and

2. Selling equipment designed to receive on those bands.

Of course there's not much purpose in sitting around listening to the digital chirping noises of encrypted traffic, but as I understand it, if that's your hobby, that part is actually legal.

(Also a licensed US ham, not not an expert in anything in particular and definitely not a lawyer)

I'm not sure that's a healthy approach though. For a few reasons:

1. The 1st Amendment

2. Not all conservatives are MAGA's

3. If you suppress a group, they're almost certainly going to realize they're being suppressed and that's going to engender even more anger, violent rhetoric, and radicalization.

That said, I don't claim to know what "The Answer" here is. Not entirely anyway. Not sure anybody does. :-(

Will I have to pinch my nose to vote Democrat? Absolutely. Will I feel icky doing it? Absolutely. But will I do it anyway because it's the best choice given the current situation? Absolutely.

One reasonable supposition about that would be "we are measuring the wrong thing". Or maybe to put it differently: "we don't really understand how to properly measure this thing called 'productivity'".

The same could be said about AI today.

Yes, and we may still not know how to measure the thing we're trying to measure.

Some 95% of enterprise generative-AI pilots have produced no measurable effect on the bottom line.

A couple of things: "pilots" and "measurable".

A better idea: Blame AI overload The time and trouble of producing a business report, proposal, plan, slide deck, or budget used to limit how large, how complex, and how frequent such documents were. Now, thanks to AI, people can churn out incredibly complex business communications, ideas, and proposals in a few minutes.

I think there's something to this. That is, with AI the work has changed qualitatively not just quantitatively. There may also be a bit of something analogous to the Heisenberg Uncertainty Principle at play, where we simply can't have perfect knowledge of all parts of the system. When we attempt to measure something, something else changes.

Suppose AI enables you to write three times as many emails as before (say, 30 a day instead of 10). Your email-writing productivity has tripled, making you more valuable to the company. The problem is that every additional email you send lands in someone else’s inbox. Your colleagues, who once got 10 emails from you daily, now get 30. Multiply that across an organization: if 10 people triple their email output, the team gets 300 emails a day instead of 100. If 100 people do the same, that figure blows up to 3,000, three times the reading burden. The AI that makes email writing easy makes email reading hard for everyone else.

I'd guess that something like this is at least part of the issue with why companies are not observing the productivity boost they are anticipating.

(This comment was 100% written by a human, with no AI assistance or support other than the spell-checker built into Firefox)

  My wife asked me why I was carrying a gun in the kitchen
  I said "Rogue AI"
  My wife laughed
  I laughed
  The toaster laughed
  I shot the toaster

There's also a plethora of people with CS degrees who can't even write FizzBuzz on their own, even given a warmed-up IDE and a partially filled out skeleton of the program. And no, this is not idle speculation, I've interviewed quite a few of these people. So I wouldn't personally put much stock in "has a CS degree" as a marker of qualification for any particular job. That applies, of course, regardless of nationality.

Having worked with a lot of foreign tech workers - I’d take the Americans any day.

Having worked with many Americans and many non-Americans over the last 35+ years, I'd say there's no general difference in ability that correlates to nationality. I've worked with dumb Americans and dumb non-Americans, and vice-versa.

That's good to hear. I think what I was mostly reacting to above was this:

(including polished or translated)

which to me reads as more of that "bright line" stance I was objecting to. But if that's not the case, then that's cool.

Then it's a useless term and we should all stop using it.

Needless to say, reality hasn't quite sync'd up with what I expected. That said, I'm pretty happy with the progress that's been made on AI from a technical point of view. But some of the decisions made by humans about the details of training the AI's we have, and how they have been employed.... well, let's just say that leaves a lot to be desired. :-(

I mean, it's y'all site, you can do what you want. But FWIW, I think making that too much of an absolute "bright line" is a net negative for HN.

There is also a printer at work that I could use in a pinch.