Hacker News Reader

> The existence of the factory did not wipe out the boutique craftsman.

But... it did? It's not like there are lot of horse carriages out there or, for a more recent example, plenty of Cobol developers. Or for a closer example, there used to be plenty of shoemakers when I was young and yet now I know of exactly one.

> The process still exists as-is in its prior form, like the woodcarvers of yesteryear. Nothing that you desire (the process/journey) has been lost. It still exists.

At the risk of stating the obvious: just because I can be a woodcarver as a hobby it doesn't mean it's the same as making a living out of woodcarving because doing something I consider enjoyable for a living is not the same as doing it exclusively as a hobby. If I fire a Cobol programmer knowing full well they'll have to work at their local supermarket until retirement, telling them "you can still program in your free time" will not bring them comfort.

Horse carriages were replaced by a superior technology, that has literally zero bearing on the discussion and actively weakens your argument. Also they are still made to demand, but that demand shrunk dramatically when people stopped needing horses.

The entire analogy is flawed though, and this is why: a market still exists for handmade objects. Factories make cheap clothing/objects, but handmade bags/woodworking are still deeply valued and there are people who still make their living in woodworking and other fields. Factories did bring some professions to near-extinction, but that doesn’t mean there is nothing left of them. In fact, many utilize factory techniques and tools in their work to ease costs, like lathes.

As for COBOL, it was mostly phased out because it’s Hell to program for and other languages like Python allow for faster dev time (for both good and bad). COBOL programmers still exist, and you’re honestly unlikely to fire that COBOL programmer because you need them to maintain mission-critical infrastructure. However, because you’re not actively building new COBOL programs, you also don’t need any more than that one. For compliance reasons it also needs to be a human, because a lot of surviving COBOL code is medical or financial in nature.

>a market still exists for handmade objects

Sure the top 0.1% of the profession can survive, that isn't great comfort to the 99.9% that lose their job though.

Handbag maker used to be a business owner in the large market of handbags.

Now he is in the small and very limited market of handmade leather objects.

Considering all the unemployed former handbag makers, competition would be fierce in that niche market.

Alternative is working in a handbag factory below previous pay grade doing a loosely related job.

It’s not clear if we’ll be given the factory option.

A craft wasn't ripped away from you: The process still exists as-is in its prior form, like the woodcarvers of yesteryear. Nothing that you desire (the process/journey) has been lost. It still exists.

Agree, I don't buy the notion that mathematics is a craft, it's a pursuit of knowledge. A better analogy might be cartography where people spent a lot of effort mapping unknown regions and now we can take pictures from space. This is both faster and probably more accurate and there are basically no unknown regions left so there's less motivation to do it anymore.

That said, there's nothing stopping anybody from mapping out an area today if they get a kick out of it. The gripe with mathematics is less about the loss of a craft and more about a superior solution replacing the old approach as the best option; the old approach very much still exists though.

> the old approach very much still exists though.

You are willfully misinterpreting the comment. You know exactly what they mean, which is that the meaning behind the craft is gone. The value of performing the craft is not in itself, but in the feeling of creating something of value, something useful to other people.

there is value in the craft itself, and in creating value. now one half of the bargain is gone. you feel like you're just playing games now and not contributing anything (which tbf, math was before to some degree anyway).

I don't get this attitude that there's no point in doing something because a computer can do it better. That just tells me that you don't enjoy the "craft" and only enjoy the part where you get credit for doing something.

It's like playing chess because you want to win and don't care about the part where you play the game. Computers are better at chess too but people still play it because they enjoy it. Does nobody practice mathematics just because they enjoy it?

the issue is that there are no value adds from humans anymore if the LLM is smarter than everyone

> the issue is that there are no value adds from humans anymore if the LLM is smarter than everyone

The value adds (from the view of humans wanting it) come from:

1) a human understanding concept X & thus can communicate about X; or

2) finding alternate routes to X that make it easier for humans to get there.

(1) is human-centric & thus non-replaceable, since the requirement of a human is what gives it the value add.

(2) is similar to how there can be many paths to deriving the Pythagorean theorem. None of the existing paths devalue the others, but they each allow those with different perspectives to reach the same destination.

right, so mathematicians will be communicating someone else's discoveries instead of making their own. notwithstanding whether that's an interesting job or not (I know, it's not part of the discussion), why wouldn't LLMs be able to do that with enough RLHF?

And why would a human be able to come up with a proof an LLM can't? I get your point but the new paths and perspectives could probably all be derived by prompting "prove xyz statement" a few times in fresh chats.

I don't know about you, but I got a lot of satisfaction from finding proofs for my math problem sets despite the fact that they were "solved problems"

> I don't know about you, but I got a lot of satisfaction from finding proofs for my math problem sets despite the fact that they were "solved problems"

Good for you, but that's not a job, and no one is going to award you a PhD for doing that.

99.99% of valuable work is a version of something that's been done before, including in math. Lots of people get phds for successfully applying known stuff to new contexts, at least in my fields. It still takes pretty deep understanding to be able to grasp something well enough to apply it in a useful way, whether you derived it for the first time ever, found it in a textbook, or asked chadbbt

"Applying known stuff to new contexts" is creative, original work, and is quite different from "working out the answers to a problem set", as mentioned in the comment that I am replying to. Don't move the goalposts.

Moreover, that's precisely the kind of work that AIs can do. So why would anyone pay you to do that, whether or not you find it satisfying?

I'm not moving the goalposts -- you just changed the subject so the discussion is different.

> I'm not moving the goalposts -- you just changed the subject

That's not an accurate representation of this conversation.

Me too - as a depressed PhD student, the only math I've been doing lately is problem sets for courses, because they're fun and help me pass the class.

But the fact of the matter is that the purpose of my program and of research isn't to work on solved problems but to solve new ones. And the machine seems to be better than me in basically every way in this regard.

Except that you have innate human dignity and the machine is a hunk of plastic and (semi)metal.

sure, but we're talking about math here.

It's an interesting thought experiment: how many human lives would cost you one LLM?

Statistically correct more of the time doesn’t equal smarter though. It doesn’t equate to more creative either. Honestly if independent thought was to become extinct due to LLM’s what would be the point of society anyways? We’re not talking about shoes or handbags, we’re talking about the thing that makes us human. So weird that we in the tech industry are so willing to ruin humanity for the sake of a sick IPO.

> We’re not talking about shoes or handbags, we’re talking about the thing that makes us human.

Exactly. That's why people are upset, and why more people are going to be upset as AI subsumes more parts of society. The handbags are just an analogy.

I wouldn't say that independent human thought will become "extinct," but rather that it will be severely devalued. And I don't see any technical reason why AIs can't replace every aspect of human thought, including creativity.

> So weird that we in the tech industry are so willing to ruin humanity for the sake of a sick IPO.

Not all of us. But the people who are making the decisions about AI have their own set of values.

There's an analogy here in woodworking -- rich people still go to craftspeople when they want e.g. a table that doesn't look like everyone else's. Same way people need to work with software engineers / web designers now, if they want to have a website that doesn't look like everyone else's.

Sure. And how many such skilled craftspeople are there per capita, compared to the number of craftspeople when all tables were made by hand?

Absolutely, it becomes a much smaller niche. Most people just want a table. Only a few want something unique.