abecedarius

The Screen of Damocles: could be a market for it.

Bookies determine the odds and typically refuse to take bets from skilled bettors.

A market is open to all, with the odds influenced by all participants. In established betting markets such as for stocks, pros dedicate their careers and their organizations to improving the public estimates emerging from the market (though not for the sake of that improvement).

General prediction markets might turn out bad, but the above isn't an argument why, it's namecalling.

A suggestion born of experience: besides printing the seed for an image, add it to the image file as metadata. Otherwise, if you're me, you'll lose it.

Thanks for explaining. I'm still confused: CakeWallet (and similar) were a reason to doubt the original claim. Are these "popular" wallets rarely used, or are you considering the nodes that they trust as equivalent to your own node?

It was a form of "huh, interesting. I tried to quickly find some more evidence for this but failed."

If Claude as search engine were able to link to some backing (maybe like "we estimate around n nodes regularly joining the network, which roughly matches the order of magnitude of estimated users" ) -- that'd be great! I'd have said I was surprised but look what I found.

Instead:

- it couldn't dig up anything supporting, except that Monero sites encourage users to run their own node;

- one point it raised against was confirmed by another reply to my comment ("apps like CakeWallet, where their node is used and assumed as trustworthy"). (Claude listed the same and a couple more wallets it called "popular" with similar trust dependence.)

I agree with GP that just relaying a chatbot is rude. That's why I didn't do that.

> running your own node, as most people do.

Huh, surprising -- it's very different from most people using most software. (Of course HN is not most people.)

I tried to fill myself in by asking Claude Opus neutrally "do most users of Monero run their own node?" and was told it couldn't find good data, it's community-promoted behavior, but there were multiple reasons for skepticism.

I have no idea, I'm just noting my surprise.

When they changed it is also when they misspelled his name. Opus got it right. I was surprised Stephenson took the misspelling as an AI tell.

The UCSD system was indeed astonishingly, unusably slow. When I got to try it in high school computer lab, in the 80s, I was like "Did whoever ported it to this particular computer just totally fuck it up? WTF?!"

An Infocom adventure on a machine with 16k RAM also had frequent pauses to fetch from floppy, but it was much more tolerable.

Re verb lookups in Basic: you could use DATA statements and READ in a FOR loop for lookup. I don't know what was typical but that's what I recall from some examples.

Interesting -- you can keep mitochondria alive outside of cells? Are there papers on what they need for that?

Maybe it's better now, but what I ran into in trying twice is that if you're not into installing by "curl | sh", then trying to build from source was an awful experience. It had out of date instructions for installing a whole lot of dependencies. I'd figure out one problem only to run into another, and another. Gave up both times, a few years in between.

That doesn't sound small?

IME both ChatGPT and Claude had a sycophancy problem, but I'm surprised by the claim it's more of a Claude thing. Is that the general opinion of people who keep up with both?

(I unsubbed from OpenAI after Altman's coup. ChatGPT was annoyingly sycophantic up to then at least.)

Good question. I tried to phrase a concrete-enough prediction 3.5 years ago, for 5 years out at the time: https://news.ycombinator.com/item?id=29020401

It was surpassed around the beginning of this year, so you'll need to come up with a new one for 2027. Note that the other opinions in that older HN thread almost all expected less.

Calling out names was an argument just for not dismissing AI as a thing "everyone knows" is fake.

Above you wrote "we all know the only real Intelligence ... is" as your support for attributing venial motives to people taking AI progress seriously. OK, now I know your basis for that claim. I've read three of the guys you mention, agree they're intelligent and except for Searle have some good things to say. But it's really unconvincing as support for an AI-is-fake claim, and especially for an everyone-knows claim.

Yeah, the headline is misleading for just that reason. But rotation in 3d does have an unambiguous orientation which we call right-handed or left-handed (the "right hand rule" if you're unfamiliar with this). I don't know which orientation our galaxy's rotation has, the article only says it's opposite the majority in this sample.

They're doing natural science on a thing full of complex purposive undesigned machinery. There used to be Artificial Life conferences -- the proceedings were pretty interesting. Now the objects of study are getting past a "gosh that's cute" level but I doubt anyone here's misled by the title.

If it's just cash-heads pushing a narrative, where do Bengio and Hinton fit? Stuart Russell? Doug Hofstadter?

I mean fine, argue that they're mistaken to be concerned, if that's your belief. But dismissing it all as obvious shilling is not that argument.

Berger wrote two books on SpaceX in a very positive light generally, portraying Musk as extraordinarily driven and capable though not passing over less admirable traits. Berger is not a hater.

When we're talking about quality of evidence, saying "in reality" is skipping to your conclusion. Are smart careful people rare among known criminals? Well, yes. What would you expect to see? https://www.econlib.org/archives/2015/02/the_wittgenstei.htm...

It depends. If you were a hacker who'd read Bamford and the news from whistleblowers like Klein, talking with other hackers, that general sense was common knowledge. But if the topic came up in conversation with, like, the guy you're subletting a room from in NYC, you could get a very skeptical look.

(I wonder if these people remembered those conversations after Snowden.)

Smalltalk-76 was around 10k lines, though probably you need to leave out the microcode/VM to get that number, I forget. (I have the source I'm thinking of on another computer powered down at the moment.) -80 was definitely bigger but -76 was a lot more like it than -72 was.

Yes, I remember using it in the early 90s.

Really if I ever knew it existed, I've forgotten. But https://oratronik.de/atariage/TI_FORTH_Manual-2ndEd(LES).pdf says yes on page 14. General discussion at https://99er.net/tifaq2.html#TIFAQ011

(I used Wycove Forth in 40-column mode, though I have a dim memory of trying something like your 3x7 font...)

Thanks for the correction. From my link:

> Each source statement you enter is immediately assembled into object code and stored into memory. Some source code is retained in a nine-page text buffer. You can scroll the screen to review previously entered lines of source code by pressing the Up- and down-arrow keys.

I gave up on this system pretty quick -- with so little space for your code, it just wasn't worth so much trouble.

No argument, it was terrible. :) But still I had good times and started learning to program.

The assembler cartridge was nearly as useless as the BASIC cartridge for the Atari VCS. (Not that I ever tried the latter, but similar problems of very limited memory holding text you'd typed in, the assembled program, and whatever data it's processing at runtime -- and needing to save to cassette tape before you can start on anything else.) However, the CPU architecture actually did give you a nice clean assembly language once you had enough of a system to really code in it.

Being away from the mainstream... isn't an advantage but I'd have to call it part of my development as a programmer. I had to get into Forth for a reasonably powerful system.

After so long I have no idea, but... that floppy file system also had a notion of copy-protected files. It turned out to be straightforward to work around that copy protection, so maybe there was some similar way around the data/program thing.

(All I remember is reading the docs of the filesystem calls and thinking "hm, could you do thus-and-so" and being a bit surprised it just worked.)

I had a similar setup eventually with Wycove Forth -- can't remember even the existence of TI Forth. Anyway playing with that was my foot in the door to a summer job at FORTH, Inc. as a teen.

Lost all of that code because my 99 system got stolen a few years later.

There was a cartridge with an assembler & text editor plus a little CPU RAM. (4k? My memory is gone but it certainly wasn't much even for the 80s.) You'd still need to use cassette tape to save your work. Ah it was this: https://www.arcadeshopper.com/wp/ti-99-4a-faq-mini-memory/

For actually usable development you could buy a TI expansion box, 32K RAM expansion, and 5 1/4" floppy drive. This cost the equivalent of like two thousand bucks today. Less than an Apple ][, somewhat more than a base C-64, but a lot more than the TI-99 itself.

My parents were very indulgent in this. Once I had this setup, I bought a third-party Forth and coded my own Forth assembler vocabulary, and finally had a reasonably capable dev env, for maybe a year before leaving for college. But still had basically no way to share my work (wasn't online).

I think the "Eudoxus reals" construction is worth thinking about too. It builds directly on integers instead of building on top of the rationals. https://ncatlab.org/nlab/show/Eudoxus+real+number

I wish history recorded how Eudoxus got to this. There was a paper speculating (iirc) that the Greeks started by thinking about continued fractions as a way to deal with incommensurables, and then realized you don't need that much machinery to define them. (You get continued fractions by running Euclid's algorithm on incommensurable input.)

1. Some blogs are better than most books.

2. Some carefully-written ebooks now are self-published, and I don't care if their cover design and promotion campaign aren't as professional.