win311fwg

Perhaps you have confused software engineer with coder? LLMs are most definitely the final death knell for anyone still working as a coder, but that profession was on the way out for decades already. The earlier comment was clear about software engineer.

...you then take it to the customer. And wait, and wait, and wait, and maybe eventually they'll get around to trying out what you created, and if you're lucky they'll even give you some feedback to loop back with to try again. They will only try your application a few times before giving up on you, so even if the feedback loop is immediate like you claim (its not; not even close), you aren't afforded many iterations. A software engineer's job is, of course, to get ahead of the customer using judgment, anticipation, and dealing ambiguity to minimize how many round trips are required to satisfy the customer.

If AI can replace the software engineer, why can't it replace all "white collar" work?

Some people do wear all sorts of uncomfortable clothes and shoes, but they do so when they want to reject the fashion. Fashionable people blend into the wall, so naturally some people gravitate towards anti-fashion in order to differentiate themselves from the rest of the heard.

I do agree that it has become fashionable to practice anti-fashion on special occasions, although that isn't the type of fashion we've been talking about.

Don't all clothes? I buy bargain rack t-shirt and jeans and get decades out of them just fine.

In the performance economy it doesn't matter how rich the consumer is, but what the consumer wants to consume does matter. Because the buyer and consumer are disjointed in the performance economy, the consumer isn't swayed by cost. However, if the consumer prefers AI performances, which they quite easily could, then humans are in trouble. But so far it seems human performances remain what the consumer wants to consume.

(Technically manufacturing and knowledge work could have established the same separation, but since the bright minds went to performance nobody thought to before it was too late)

If we take "write the specification of how to approach the problem" at literal value, you might be narrowly thinking of a low-level compiler, or what sometimes gets called a transpiler? However, compilers encompass a much wider spectrum than that. Generally they enable users to operate at higher levels of abstraction.

If the compiler is writing the specification, what purpose does the person serve? That is not what I'm seeing out in the wild, though. Perhaps what is happening in research labs aside, compilers, even those implemented as an LLM, don't do anything until the user specifies their intent.

In fact, you literally recognized that in your earlier comment, asking about my action to specify my problem. Not sure why the story is changing now? There is probably a day where the systems solve your problems before you even know you have them, but we're not there yet. The technology hasn't advanced that much in the span of a couple of hours.

> asking clarifying questions as needed.

Sure. Even the earliest, most primitive compilers would seek clarification if there was something that it felt required more information. That's expected.

> If you're using LLMs as you describe, you're using them ineffectively.

LLMs are just compilers so they can't do anything else. Maybe you're thinking of some technology that is layered on top of a compiler like an IDE, REPL, or something to that effect?

Sure. I wrote out specification that described the problem. It compiled that into a description of how to deal with it.

> People aren't just using AI for writing code

Compilers have never been about producing code in particular, though, even if that is where they are most commonly used. A LaTeX compiler, for example, outputs books. To think they are about code exclusively is a grave misunderstanding of what a compiler is.

(No, compilers are not deterministic)

The tech crowd used to say that only the "dumb" ones did that. They said those who were "smart like us" were much more adaptable, learning to code and whatnot to find new, often even better careers. It will be interesting to see if the emperor has clothes.

Assuming no new paradigms on the hardware side that no longer fit into our current programming models, do they? Rails' claim to fame was eliminating "XML sit-ups" to improve the developer experience, but who cares how many "sit-ups" LLMs need to do?

Drain plug is in a stupid location that I don't often feel like dealing with. I do change the oil myself in all my other machines with much more pleasant for oil change designs, though. So I guess what you are saying is that AI isn't ergonomic enough yet, but will that hold true much longer?

How? I do remember the big protests in the 90s quite well, certainly, but that was 30-some-odd years ago and when, as we saw earlier, there were 300% more diary farmers. Pressure doesn't last forever, especially as you grow into insignificance.

From what I see the pressure comes from the general public, mostly from having built up this idea that supply management keeps "hormone laden" milk out of the market, combined with the people of Quebec seeing any kind of federal "meddling" as an attack on their people.

When longstanding party favourite Maxime Bernier ran for leader of the CPC, the as you say the nominally free market party, on the platform of eliminating supply management it wasn't just dairy farmers who laughed him out of the room. It was everyone. Same thing happened with the early PPC party, before eventually pandering to the crazies to try and find any kind of support. It was quite clearly non-farmers who wanted no part in seeing an end to supply management.

> because the farmers were fuming at higher freight rail prices against other commodities

Words are only worth so much, granted, but when this occurred it was framed as being necessary for the sake of the entire Canadian economy and to preserve Canada's reputation as a trustworthy supplier. The message sent was abundantly clear that it wasn't being done to appease farmers. How do you translate that into farmers having political sway?

Again, the largest component is crop insurance, which is modelled essentially the same as it is in Canada. ~60% of the premium is paid for by the government in both cases. The payout is dependent upon acres.

> The number of US dairy farms is down almost 80% since the 1990s

And 74% in Canada, for what it is worth.

> If you're a Canadian dairy farmer

Not anymore. I am among the 74%. The double-edged sword with the stability offered by supply management is that it makes borrowing much easier (effectively government-guaranteed cashflow), which pushes the price of everything higher, so the economics aren't all that great unless you sold your $100,000 per acre farm in the Netherlands or Germany to pay for it.

> the US system looks like it encourages production above all else

I am not sure I would say that. Co-ops with quota-like systems are not uncommon in the US. What their system does allow is that one guy who thinks he can undercut the prevailing market to slide in, while Canada has a legal path to put a halt to such behaviour. Maybe that's something favourable, but why only in dairy, poultry, and eggs? Isn't that something that every market would want?

> Canadian farmers have a lot more of a direct influence on their government

I do remain a Canadian farmer, though, so I would love for you to expand upon that thought.

> Because it's political!

This is quite a deviation from the bulk commodity/farming practice explanation before. Why the change in tune?

It's essentially the same system, with the primary subsidy being paying 60% of crop insurance premium in Canada and 62% in the US. Of course since crop insurance is acre-based, those with the largest landholdings capture most of the subsidy, but that is equally true in Canada. It's probably true that the US has some larger farms, given the much larger arable landbase (and one of Canada's largest farms just having recently gone bankrupt!), which is what I expect you are trying to say, but I'm not sure how pertinent that is as it is still an equal payout on a relative scale.

> they are not bulk commodities that ship around the world like grains do.

Neither is milk from other animals (e.g. goats), other types of meats (e.g. pork, beef), etc. but none of these have the same system. Moreover, until 2009 tobacco was also supply managed under the same system and it is a bulk commodity that is shipped around the world. So I'm not sure your theory, as fun as it is, holds.

> Canada also doesn't allow a lot of the farming methods (including hormones) that the USA does.

It's a mixed bag. Canada puts higher value on animal welfare, but is much more lax when it comes to things like pesticides. The US doesn't allow hormone use in poultry, so that doesn't really explain what purpose supply management serves either.

There are no real limitations outside of dairy, poultry, and eggs; same as on the US front. If it is so difficult why is there nothing in place, aside from those three categories? Are these the only food categories that matter to people? What about the food that the animals in those categories eat?

> but the US directly and indirectly subsidizes production.

Since the Obama era, the US and Canada have had essentially the same subsidy system, save dairy, poultry, and eggs, which, as you pointed out, get their own special rules. But as you pointed out they are still subsidized indirectly via things like corn subsidies, which is where Canada and the US are now effectively on the same page (that wasn't always the case, of course, but the reforms happened a long time ago now and what went on 30 years ago doesn't mean much anymore). When cooler heads prevailed the US played fair in aligning their system to Canada's, so why isn't the same being willing to be offered in kind? Changing the diary, poultry, and egg system doesn't mean giving up subsidies, if the people want to continue to offer subsidies. The US model already offers a way to provide that. Canada already accepts that for other food products, so there is nothing unprecedented there.

Of course, it's not really about the farmers/production, is it? It is about protecting Saputo, Lactalis, and Agropur's triopoly.

Those who argue copyright infringement have a somewhat compelling claim, but stolen is a little out there. Especially since text I have written has provably never been stolen, making "every piece" impossible.

The underlying trouble with the copyright infringement claim is that it forgets that copyright was extended by the people as a construct to help foster innovation so it expected that the same people would ignore/revoke copyright claims where the product of its use is seen as a greater innovation. Copyright is but a human-created tool, not some natural law of the universe.

I understand also that not everyone sees AI as a worthy innovation, but societies do not function on the whims of a few people.

To be fair, I've seen no nation where the people make their will known. Does a democracy depend on actual action or only theoretical possibility? It seems from my observance that most people consider democratic nations to be democratic not because they practice as such but because it is believed that the people have the opportunity to exercise democracy if they ever choose to.

Which also solves the alignment problem with those who do not enjoy seeing LLMs write software. But that brings us back to: Aligned to whom?

Your 'correction' is incorrect. A corporation did not carry out an attack. Humans did. And, as it happens, those humans were acting as agents to OpenAI, so the original title technically got it right. It is a poor title as anyone who doesn't give it much thought might mistake a human agent for an LLM agent so your symbolic effort to improve upon it is warranted, but sadly you missed the mark.

If we knew it was employees that did it then "OpenAI employees carried out an attack on RubyGems." would work, but since we don't know who did it "agent" is better in the sense that it also encompasses contractors, board members, etc. Of course, if we knew who did it then "<Person's name> carried out an attack on RubyGems" would be the way.

The joys of English.

Does OpenAI being considered "too big to fail" lead us down the wrong path? Yes.