polotics

So what do you think the rational response to the nukebullies should be? Give him all he wants? How are reasonable deterrence preparations "Bravado"?

Can you explain what you meant with this sentence on that page you wrote:

"That is the whole system, not a stripped-down mode you switch into."

Did I miss something or does this article not bother to indicate how much RAM their M4-Max had?

yeah my 'promptly AI' alarm clock almost always spends 7 days, as soon as the binary changes, even descriptions changes take several days

I was on unquantized 27B, and also Q8 K XL, same "Need do..." paragraphs. We landed in near but distinct morrasses I guess.

They didn't actually steal in the sense that the information is still there on the internet.... About these shredded rare books, now we're talking.

If I may propose instead of "steal" I think we could agree to write they "Aaron-Swartz'ed" the information from the internet, what do you think, is this too harsh on Sam Altman or Carmen Ortiz ?

I have seen plenty of Qwen 3.8 27B's caveman-like "Need doing this & that" thoughts. And on cerebras now I've seen them come real fast!

Glucocorticoids in your early amygdala? That's a bummer!

Blame the game, not the player! In today's kind spirit of early-internet you're also not a dog on the internet communities, there you go...

No warranty granted or implied: A bit of sarcosine in the morning, and a bit more glycine at night I say. :-)

Repeating myself here but anyone who has not yet watched the whole 2024 series of Prof. Sapolsky "Human Behavioural Biology" lectures series... really should.

https://www.youtube.com/watch?v=mZI2GQlNdJA

RIP Lee Holloway

"The gap became the north star" is written large enough though: I could read it with my own Aieyes!

It's also the only sentence I read on that page before closing it, of course.

The one weird trick of reassigning all the income from that contents-stealing to the original posters would I think cover a lot of the realignment.

I wouldn't call this a bias. More likely an A/B tested find that (1) algorithmic YouTube video-surfacing with keyword "graph" plus AI related name-dropping was effective, and (2) the already well established journalistic trick of leading with fear to trigger loss-aversion, as "prompting is soon dead" implies some threat of skills-value loss for the reader.

driving engagement to scrape the bottom of our ad-driven engagement dis-attention economy, tragedy of the commons pure.

Are you serious about this imbalance? it's the seriously rich that have taken all of the spoils, that's where the massive imbalance is.

I'm not disagreeing that some manual work require forms of intelligence that are undervalued in our society, and don't get me started about nurses, but still: What the Jeff Bezos!

So much of this yes, colleagues who think doing AI work is asking Claude to, as you so eloquently wrote "vomit a novel", and allow themselves to disengage their brains.

The joke is at the level of developers, the constant sharpening of AI merge-requests reviewing (sometimes too much) has made the new job of AI speedriding & reviewsfighting a serious brain squeeze.

The value of conscienciousness has skyrocketed, and the value of openness is still the same as it ever was.

how do you know it is for ads? how do you know it will always be only for ads if it now is?

GDPR compliance lawsuitssss in 3...

I put it to you that seeing a chord: on paper, or in math or in chord notation, (eg. FminSus4) is something you should only master once you are able to hear that chord. Otherwise you are (that's a quote) "dancing about architecture" and hurting your ability to understand sound.

this my friend, is the point. I refuse to not hold the human fully and entirely accountable.

I put it to you that all slop is human slop.

why oh why does the author think that it is ok to have the first thing people will read be completely devoid of clarity and to-the-point substance?

"""meclaw is three things, and you only install the first one. meclaw is the substrate: a directory tree that runs — every folder an actor, every edge a route, one Rust binary underneath. meclaw-os is a small, experimental operating system for agents, grown onto that substrate at runtime. An assistant is grown into the OS the same way — a JSON file, not a deployment. Install once, grow everything else."""

Like seriously dude, what semantic value does "Install once, grow everything else." actually add? Can't you see the complete self-defeating nature of pushing out this kind of vapid prose?

Also "a directory tree that runs" what the hell, you mean like any folder structure that contains code? Hey my python package is also a "directory tree that runs"! How is that good, again: what is the value of a "directory tree that runs"? seriously?

And then "every edge a route, one Rust binary underneath" You mean the code that runs is "underneath"? Like my mousepad is also underneath my mouse, should we also put this in your package description?

Any value this effort may have had is completely destroyed, and anything else you may have produced, by association.

Haha well it's easy then please send us the MIDI, Openscore, Lylipond, MP3... anything really, of a composition of yours that will be as appealing as Twinkle Twinkle Little Star.

What else do I see: congruence between melody and lyrics, harmonic depth, dominant half-cadence in the tonic arrival. Also the versions by Mozart add a lot and explain a lot of the endurance of the tune I guess. That has chromatic passing, secondary dominants, even a parallel minor (in the 8th variation)

If I may plug here the Berklee College online courses as a satisfied customer I can say if you are ready to put in the effort they are great.

Sure.

It is a trope amongst gigging musicians that over-intellectualizing the feels is a massive no-no.

Hence the joke about Db being the saddest chord... also Phil Collins said that, he would know :-D... ...and C# being the most evil chord. Hilarious.

I persist: "before you can be good, you first have to be bad" is more helpful music theory than that whole JavaScript visuals page.

Let us try an experiment shall we: listen to any music you like and try and match the information the article provided to the complexity of what you are hearing. How much do you see a path from A(rticle) to S(ound in your ear).

This article is trivializing music down to a few not-so-important facts and mistaking some math for anything that would help `programmers` (?) understand music.

Maybe start by rewatching "This is Spinal Tap" instead, and consider that https://www.reddit.com/r/musictheory/comments/59n3pg/remembe...

it really depends how you do it, I don't think one serious non biased study has ever been done. probably an ultrasound bath is the most energy and water-efficient method

what's moral panic got to do with inflated energy prices, noise, and polluted water?

regulatory moat is the theory I guess

yep "next-embedding" predictor is more correct, and not just at the end but through the layers, and folding back dimensions into that one next token is one small final step, and next-embedding could be named "next-meaning" as well, and we're getting there...

this sentence above would made a longer article if I bothered to so blog as is being blogged here

Not trying to be provocative or anything, but ELIZA 60 years ago was already a better psychoanalist than many people I know.

`By your metrics, we achieved AGI in 1966.`

https://en.wikipedia.org/wiki/ELIZA

Nah as long as you don't give them first-names, I think you should be ok. also:

Pretty pleaser please people don't get your agents registered as direct-reports in the org-chart with HR!

Call me when Astra gets the Nobel prize... We'll have AGI when prizes have two categories, one for assisted humans, and one for pure-AI.

At the end of the first paragraph there is this sentence:

"Training support is a hypothesis consistent with these results, not something this study proves."

I understand it, but I find the wording very unnatural. To improve readability I would have written it in the active form:

"We cannot prove training explains this result, but this fits the data best."