I don't think "plateaued" is the right word, but I do feel like there's been something like a logistic curve compression in the difference between smaller and larger models as the field evolves. For inference at least, the scale of practical difference between a single high-VRAM GPU or SFF UMA box, a whole rack, and a whole data center seems to be falling far short of what we might have imagined just a few years ago. The conversations I've heard have largely turned away from breathless anticipation of the next frontier model and toward attempts at hard-nosed evaluation of which tokens are worth the cost.

I think it’s more that pushing frontier is extremely costly and there is no free lunches in same way as 2024.

The difference is still vast, it's just that the smaller stuff is "good enough" for many things now.

Maybe, though that's another kind of progress in itself. Very impressive progress!

It would absolutely make sense for someone who enjoys cycling to buy a pedal-assist e-bike, especially if they e.g. moved to a hillier area, sustained an injury, or just started having their strength/stamina decline with age.

(incidentally, the proliferation of unregistered/unlicensed motorcycles marketed as "e-bikes" has done enormous damage to the viability of e-bikes as a general category)

It's called metaphor. Please reserve your erm akshully's for someone else.

That appears to be using your metaphor and pointing out how being a cyclist and using an e-bike aren’t mutually exclusive. (I do know multiple avid cyclists that own e-bikes as one of several bikes they own and ride).

Ripping off the metaphor for a moment, it was basically extending what you said to point out that being a coder and using LLMs aren’t mutually exclusive.

(Asking them to reserve erm akshully’s for someone else came across as rather dismissive sounding of their thoughts, which from a former bystander in this thread appeared legitimate and just a continuation of using your own metaphor).

It's actually a valid extension of your metaphor that introduces nuance that substantially weakens your point

I doubt that's the only factor, but I agree that it's almost certainly a factor. The modern "tech" industry has filled our waking hours with unceasing demands to do something. Buy this product, watch this video, listen to this podcast, react to this social media post; even "practice self-care" has somehow turned into an obligatory virtue-cultivating task. We obviously had advertising before, but this is a different beast, something akin to living in a casino.

As Bo Burnham put it, "apathy's a tragedy and boredom is a crime".

True, but it's also the case that in some places and subcultures we're already at a point where meat with certain qualifications in some contexts is a Veblen good, and the rhetoric around it is very much a quasi-mystical blend of notions of purity and vitalism (see e.g. paleo/primal/ancestral diet influencers). That edge, almost by definition, can't be produced in a lab, and if we were really being purely rational/utilitarian about these things, I doubt we'd be eating as much meat as we are today.

I love a good cheeseburger. I’ve enjoyed ones from Michelin restaurants and from local holes in the wall; I’m not choosy. To me, the taste and texture of a cheeseburger is one of the best testing things on the planet.

I do not eat cheeseburgers to feel manly or dominant. I eat them because they are delicious. If I could have a plant-based or otherwise non-meat burger that tastes as good as the real thing, I’d eat it. Impossible burgers are getting close and I’ve had my fair share of them, but they’re not there yet. They’re near enough that I’m confident we’ll get there soon.

What really matters is that there's a coherent policy for assigning addresses, which is absolutely compatible with having multiple DHCP servers on the same subnet (e.g. each server configured to hand out non-overlapping IP ranges).

It's less about the IP ranges they're giving out not overlapping and more about the requests that they respond to not overlapping (or at least that they don't contradict one another). The classic 'someone plugged a consumer wifi router into the office/university network' problem often is made worse because the consumer wifi router gives out different addresses.

I'm not a copyright lawyer, but my understanding is that you do not own a license when you buy a CD or cassette, you own a copy (strictly speaking, a "phonorecord" in US law). Copyright doesn't regulate the sorts of typical uses that the owner of a copy would engage in, so a license is not required. When you see those warnings on VHS/DVD/Blu-ray that they're licensed only for private home viewing, what they're telling you is that you don't have a license to do anything controlled by copyright such as broadcasting, publicly exhibiting, or reproducing it.

Software is significantly more complicated and gets into issues of physical permanence/transience and specific exemptions/limitations to normal copyright law (see e.g. 17 USC 117).

Right; it's not political in the sense of "vote for candidate/party/policy XYZ", it's political in the sense that Omarchy is distinguished by being a vehicle for DHH's opinions, and by promoting Omarchy one can promote the general idea that DHH is a respectable guy with good opinions while still remaining firmly situated in "technical" subject matter.

Also, the book that popularized "hello world" was very consciously advancing a programming-first pedagogical agenda, as opposed to building up a theoretical foundation first:

> The only way to learn a new programming language is by writing programs in it.

The problem is that once you draw "the line", then what? Warehouse a bunch of helpless people in overcrowded prisons? Bus them to some other less-draconian municipality? Forcibly drug them into quasi-submission and then turn them loose again until the depot injection wears off? Or perhaps we should "just kill 'em", as Fox & Friends cohost Brian Kilmeade once suggested?

If they are committing crimes, they should be imprisoned for an appropriate period of time as a deterrent to crime, a punishment for their actions, and to facilitate rehabilitation (even though evidence shows this doesn't really happen).

If they are mentally ill and unable to care for themselves or pose a threat to the community because of their mental illness, they should be institutionalized and treated. If they are able to be released into the community after a period of time, checks should be performed to ensure compliance with treatment/medication.

The most important part of incarceration is removing what is a quantifiable risk to the public from the public square. Everything else is secondary. The point of a state having police powers is to ensure for the common good, it is not to continually indulge those who wish to damage that common good. To the extent that the state is responsible for e.g. punishment and rehabilitation, these are ancillary responsibilities that are (or were) downstream from removing threats to the common good from the public square.

Seems like, after all of these police and prosecutorial "reforms", we are back to deriving why many state powers exist from first principles. Fun to watch in real time.

"I've rarely seen an unhoused people do harm" is an absurd statement.

It's an old study, but Martell in Base-rate estimates of criminal behavior by homeless mentally ill persons in New York City found that "The rate of violent crimes was 40 times higher and the rate of nonviolent crimes 27 times higher in the homeless population".

Most people don't like mentally ill homeless people who commit violent crime at a rate 40 times higher than average. Remind me what right in the Constitution enshrines your ability to crack the skull of a 70 year old nurse on the ground, then wander off and go to sleep? https://abcnews.com/US/70-year-nurse-killed-random-assault-b...

> It's an old study, but Martell in Base-rate estimates of criminal behavior by homeless mentally ill persons in New York City found that "The rate of violent crimes was 40 times higher and the rate of nonviolent crimes 27 times higher in the homeless population".

It is indeed old (1995 [0]) and that's not what the paper studied or concluded. First, the study did not gather data on the homeless population, but on mentally ill people already in the criminal justice system:

* "The authors analyzed data from ... records of 77 homeless defendants and 107 domiciled defendants referred for psychiatric examination by the criminal and supreme courts ..."

Second, the paper's goals were to gather needed data on homelessness among mentally ill people in the criminal justice system, and to examine the interactions between homelessness and crime among mentally ill people:

* "The study aimed to estimate the prevalence of homelessness among mentally disordered offenders entering the criminal justice andforensic mentalhealth systems, to compare base rates of arrest for violent and nonviolent criminal charges among homeless and domiciled persons with mental illness, and to examine patterns in the categories of victims chosen ..."

Third, the conclusion was that, among mentally ill people generally (not only those charged [1]) those homeless were ~35x more likely to be charged with crimes than those domiciled, and it was somwhat even more likely that those crimes would be violent or against strangers (there's more detail in the paper). The paper says nothing about crime among the homeless population generally.

> absurd

> crack the skull of a 70 year old nurse

Do pejoratives and outrage make it more persuasive or less? More importantly, do they make it more likely or less likely that you find and express truth?

I think much less likely, because it means 1) You are dismissing a lot of information, 2) You are not examining the matter thoughtfully (being outraged), and 3) You likely have little better argument.

That also makes it less persuasive to people who are thinking critically.

[0] When crime rates were much higher. The paper: Martell, D. A., Rosner, R., & Harmon, R. B. (1995). Base-rate estimates of criminal behavior by homeless mentally ill persons in New York City. Psychiatric Services, 46(6), 596–601. https://doi.org/10.1176/ps.46.6.596

[1] Note that they gathered data only on mentally ill people charged with crimes and not on the general population of mentally ill people. The latter group's data they took from the same author's prior research.

Can you provide any evidence that homeless people are equally or less likely to commit crimes, compared to the general population?

Look, if you make a claim - that homeless people are dangerous - then the burden of proof is on you.

A scientist doesn't say, 'there's life on Venus, prove me wrong'. A lawyer doesn't say 'you burnt down the house, prove me wrong'. They have to prove their claim. Otherwise, nobody responds - in science they'd be ignored, in court the defense doesn't even need to say a word.

Otherwise, we just don't know. I gave you the respect of taking your argument seriously, and of finding and reading your citation, even as you were hurling abuse and outrage (and maybe hadn't read it yourself, it seems). If you will step up now and contribute, you'll find a welcome reception and an interesting life and discussion. You can contribute. Someone else probably cited that quote and paper, and didn't read it, and look where it left you. You can do better than them, you can find some actual truth - and we'd both learn things, I'm certain.

But now you just don't know, admit it. Your assertion is possible of course, but IME it's also possible otherwise - homeless people are among the most vulnerable. For example, is the crime rate among homeless people more or less than the crime rate in corporate executive offices or on Wall Street or in the offices of elected representatives - I could see it either way. I just don't know.

You gave an anecdote that "I've rarely seen an unhoused people do harm". I'm saying do you have any evidence beyond your assertion from personal experience that homeless people are safe?

You're making a claim that homeless people are safe, a claim that flies in the face of most people's experience and common sense.

It also flies in the face of the data:

"In 2022 the San Diego County District Attorney’s Office reported that homeless individuals were 514 times more likely to be arrested and charged with a felony and were often facing multiple criminal cases. The data shows substantially higher rates of robbery, residential burglary, assault, arson, and vandalism. " - https://www.sdcda.org/content/MediaRelease/Homeless%20Data%2...

"Los Angeles Police Department crime data from 2018 to 2021 shows that homeless individuals, while only 1 percent of the population, are suspects in 6 to 8 percent of all crimes and 11 to 15 percent of violent crimes." - https://abc7.com/feature/homeless-crime-los-angeles-data-res...

You're welcome to admit you are wrong about the precious "unhoused" who are committing assault at a rate 130x times greater than normal people in San Diego, or you can keep trying to deflect about Wall Street. $20 says you're the same type of person that thinks a pitbull is a safe nanny dog.

The contempt seems like a pattern - toward homeless people, toward me, toward facts and anything and everyone else that don't fit what you want. It's not persuasive or interesting, and contempt seems pretty strongly correlated with not knowing, learning (essential to knowing), or caring about the actual facts or anyone else. It's also not respectful and incompatible with freedom - that other people can validly do and think things that make you uncomfortable.

Why would anyone take these comments seriously? Why would I spend my time on them? I think I will be moving on. I hope you have a good rest of your day.

> hey should be imprisoned for an appropriate period of time as a deterrent to crime, a punishment for their actions, and to facilitate rehabilitation (even though evidence shows this doesn't really happen).

And to prevent them from committing crime for some time. That might need to be a very, very long time. It is the most important reason to lock people up, sometimes.

Right now, many places in the US aren’t even “treating the symptoms”.

We can start there, then move to address the more structural, long-term problems that contribute to these issues.

An action that reduces drug use by 5% a decade from now is fine, but taking a criminal off the street today yields far greater present value.

If they're doing lots of vandalism then yes put them in prisons. Get rid of 90% of drug charges to make room.

You get a much better score if the imprisonment is trying to rehab them, of course.

Forced rehab if needed and then somewhere they can work to pay their way, learn the skills to help themselves. There's a ton of manual labor that needs doing, potholes to be fixed.

Oh, so forced labour then?

If someone’s forcing society to pay for their living expenses, it only seems fair for society to force them to labor to cover the expenses.

I’m sure the next objection will be “what, so you’re going to concentrate them in some kind of camp?” Yes because that’s the smart thing to do with a large scale problem. Use scalable solutions.

>If someone’s forcing society to pay for their living expenses,

But the mentally unwell person didn't force society to pay for their living expenses. Society decided it couldn't stand them living their life the way they were and locked them up.

Society pays for locking people up because it is done for the benefit of society. The point is that a functioning society is more productive than an unorganized one, and so can afford to do that kind of thing.