el_jay

Russia could end the war today by pulling out of the occupied territories and paying restitution. They aren’t doing that, and it seems the Ukrainians themselves still believe it is worth risking life and limb to keep them out.

I don’t know where you’re from but I’d bet money it’s not your land the Russians would be occupying, товарищ.

Lots of not so friendly countries are watching what goes on in Ukraine very carefully, and making decisions about their future military plans…

Нet, товарищ. We do not reward wars of conquest by granting territorial gains to aggressors in Anno Domini 2026.

Maybe after an LLM nearly starts a war?

https://www.yahoo.com/news/politics/articles/u-nearly-starte...

Once I overcame the addiction tech, engineered over billions of brilliant-person-hours, atop research into human behavioural prediction and conditioning that’d make the KGB blush?

It’s the hundreds of people on my Facey that I have no other way to contact, and whom I can’t convince to move to e.g. Mastodon, even by offering to do or automate it for them. I suppose I could get an agent to run through the list with a script that gives my phone number to anyone who expresses a wish to remain in touch.

Quora has become a scam and CP-ridden cesspit but there are still some old-timers who can really write, and whom I’ve become somewhat emotionally invested in. And a habit of 13 years is hard to break.

To save other readers a web search: in a *Keynesian beauty contest, you pick the winners based on who you think others will find most beautiful.

> These are all fine questions, and they don't become any easier to answer if you replace "computers" with "brains" and "bits" with "neurons".

What is even being argued here - neuroscience is hard, so programming your PC thus makes it conscious?

> "That's ridiculous. How can meat make a machine? You're asking me to believe in sentient meat."

… what?

Let’s say I stand up and step away from my PC, which miraculously runs a local Opus 4.8. Once I’m away from the PC, for an arbitrary time, nothing and no one calls or otherwise acts on the instance - not myself, not agents, not updates, not hackers, not bit-flipping neutrinos.

Given enough time, would that instance get bored and start e.g. reading files?

If I monitored it long enough, would there eventually be spontaneous outputs, or changes to parameters or architecture, even with the underlying software and hardware layers held constant?

Does GPT-n dream of electric sheep, or does it just sit there until interacted with, like every other file on my PC? Seems to be the latter with my local Qwen3.6, but perhaps 27B is too few params for consciousness to emerge.

Proponents of LLM consciousness could settle the argument in minutes by showing proof of unprompted autonomy, without first needing to define consciousness down to a rigorous mathematical abstraction. Why don’t they?

Who really cares about the lives of lim n —> 4k wagies? There is an opportunity to maximise shareowner returns here - failing to seize it would be little more than economic treason, a dereliction of our duty to be good capitalists. If those people wanted job stability, they should have worked harder to become indispensable to their employer. Frankly, they should have known better than to stake their livelihoods on unstable, declining industries like employment. Now, The Market Has Spoken, and only those let go are to blame for what it said - no one else.

On top of all the extremely valid points about the ad-driven cognitive friction inherent to modern device usage: print books can’t get yoinked off my shelf because a rich person with political connections wants that.

This article and discussion appear to have been manually delisted from the News rankings.

Evidently, even HN could only keep up the pretense that tech development is amoral and apolitical for so long.

“Never”? Not once in the Story of Us has any dispute between large groups of humans been resolved by anything other than a superior application of brute force? Strong claim, but I’ll run with it.

And you appear to believe this is a pretext for humans to ignore their own laws and commit atrocities, when they could choose otherwise.

It may be reality that jungle law is currently how humans almost always handle conflict at nation-state scale. Non sequitur that it should remain so.

Indeed, why bother having states of law at all? Jungle law works well enough in reality.

My understanding is that deep-sea nodules produce oxygen by a process similar to electrolysis, where they generate currents that split seawater into hydrogen and oxygen.

The proportion of global abiotic and global total oxygen production this represents is not known, but may be significant.

Leaving aside the certainty of yet more cascading collapses of marine life in waters de-oxygenated by deep sea mining: do we want to risk finding out the hard way how significant?

Apparent consensus is we do, but I don’t have to like it, or think these are the plans of sane people who see the big picture.

ML and physics share a belief in the power of their universal abstractions - all is dynamics in spaces at scales, all is models and data.

The belief is justified because the abstractions work for a big array of problems, to a number of decimal places. Get good enough at solving problems with those universal abstractions, everything starts to look like a solvable problem and it gets easy to lose epistemic humility.

You can combine physics and ML to make large reusable orbital rockets that land themselves. Why shouldn’t be able to solve any of the sometimes much tamer-looking problems they fail to? Even today there was an IEEE article about high failure rates in IT projects…

Suggest tagline: “Eminent thought leader of world’s best-funded protoindustry hails great leap back to the design stage.”

Expectations do not spring sui generis into hearts and minds. I don’t know about you, but AFAICT approximately sane and rational folks (ha! Ha!) at least try to derive their expectations from reality.

When most people’s reality is substantially human-defined and abstracted from nature - including a global advertising industry exists to create mass expectations, of economic significance for its clients, often enough to the detriment of their target markets - you can absolutely point the finger at “reality” for pissing on your leg while varieties of Stoic, Buddhist, and HN poster tell you it’s raining.

It’s good to start with ourselves when trying to create change, as that is where the locus of control should lie… but sometimes “reality” is absolutely the reasonable and proximate cause of negative emotion. Saying otherwise feels like anticipatory victim-blaming.

While there is certainly an argument to be made that many contemporary “Western” Pseudo-Christian Superempire nations face a crisis of short-termism, there are also ancient bits of “Western” infrastructure like the Roman aqueducts still in use today - off the top of my head, the Aqua Virgo which supplies Rome’s Trevi Fountain, dated either 19BC or 19AD, I forget; Spain’s Segovia Aqueduct from the first century AD; and the Pont du Gard in Nîmes, from the same period.

Not quite as old, or at the scale of the Dujiangyan system, but still evidence that the “Western” culture did once build for long term. Less ancient, but more indicative, are the European cathedrals built by multiple generations over a century.

As overt corporatism goes, you can do a lot worse than simultaneously embed tech execs in the military command chain, and military commanders in the tech industry.

A health insurance CEO was shot dead in the streets. It’s only one pitchfork but it’s still a pitchfork.

And only weeks before a US election.

Why no mention of a simple wealth tax? The Swiss take 1% of each citizens’ wealth, including unrealised gains on capital market holdings, every year. Paired with a policy of zero cap gains taxes on individuals - unless cap gains are that individual’s sole or major source of income, in which case they are taxable as income - this appears to strike the right balance between raising revenue and encouraging investment, without frightening off too many rich people.

Seems much more straightforward than faffing about with unrealised gains.

Am I to understand that it’s legal and okay for LLM providers to profit massively from training commercial models on copyrighted works, without the rights holders’ permission - but illegal, and unacceptable, for private individuals to access a digital library?

Have spent too little time with Haskell, and am fairly indifferent to pure maths beyond forcible conversion to applied maths against pure mathematicians’ will - but I have an enduring love for jazz. It is freedom incarnate, and the most perfect musical metaphor for life yet known to me.

………………………………………………………………………………… Location: Bali Indonesia currently; footprints in VIC & QLD Australia, Switzerland, U.K., Italy

Remote: preferred; hybrid okay

Willing to relocate: yes

Technologies: C#, Python, SQL flavours, GCP + BigQuery, ASP.NET MVC, bash/zsh, HTML/CSS/JS.

Résumé/CV: https://www.linkedin.com/in/bring-data

email: elliot.grinstead@protonmail.com ………………………………………………………………………………… Data scientist, consultant, and researcher. 6 years turning data into knowledge, and knowledge into commercial automations - or plain old profitable decisions for key stakeholders - at startup, SME, and entreprise scale. Anglo-French bilingual with econ/finance & stats background. Empathic listenership user, in stakeholder requirement elicitation as in junior mentorship. Enjoy public speaking to diverse audiences of different sizes, and levels of technical understanding. Proactive approach to personal and team skill development that balances present OKRs with long-term success goals. …………………………………………………………………………………

“Plutonium:” was added soon after my original comment :P

Fair warning to any folks who went in expecting a read about dimethylmercury: the article is about plutonium.

Two kinds of debate: one where the expected outcome is a refinement of ideas, another where the expected outcome is an agreement on future behaviours (“alignment”). If you approach someone who only converses to negotiate alignment, they are likely to take your seeking of peer review feedback as a mindgame designed to make them do more of what you want to do.

For me, this explains why some compulsively managerial types are incredibly averse to conceptual abstraction in conversation: it comes off as a bamboozlement attempt!

Simple example:

Alice: “I need you to do X.” Bob: “OK, I’ll do X.”

Not much negotiating, but Bob has aligned to Alice’s request.

More subtle:

Carla: “Do you think we should do X or Y?” Danny: “I’m leaning X but Y might have the following benefits…” Carla: “I’m leaning Y for those exact reasons. Why do you think X?”

And so on, until alignment is reached. Hope this helps!

Reminds me of Joscha Bach’s definition of a nerd, from his Lex Friedman podcast episode:

<paraphrase> “The definition of a nerd is someone who thinks conversation is for peer reviewing ideas.” “What do non-nerds think conversation is for?” “Negotiating alignment.” </paraphrase>

A good difference to keep in mind when we want to bend parallel conversations with the alignment negotiators to intersection.

Interesting, thank you. One remark:

> Modernization is hard, time consuming and very expensive. These regulations are a good driver for all that, but it takes time. Having been in some of these boardrooms there is not typically a cartel of evil executives figuring out how to screw patients. A lot of the time it is a group of relatively normal people trying to hold up a technological and business process house of cards with chewing gum, paper clips and hope.

It bears questioning how, in ”the richest country on Earth”, something as important as a hospital ends up held together with spit and twine.

Inside this richest country, the richest individuals and companies have been capturing ever-larger shares of GDP. Perhaps instead of growing private fortunes, this money could have gone towards modernising these regulated industries?