Live data from Hacker News

Post-GPT Computing

grady.io

441–450 of 504 posts

Re: Post-GPT Computing

#441
post #394

Earlier quoted context omitted.

So wait, just to be clear, you deployed production code in a language you don't use regularly? And this is a good thing? This is supposed to take programming jobs? HN is incredible.

Who said anything about production code in a language I don't use regularly? Internal tools that automate 3 workflows we'd been doing manually. 2 node scripts and a super simple web app exposed on our private network.

'Production' doesn't mean 'external'; it means 'actively in use'.

Re: Post-GPT Computing

#444

Earlier quoted context omitted.

There will never be anything free. We always have to pay with labor. But maybe we should share the fruits of labor more equally instead of giving most of it to someone who has some documents they got from their parents.

It is always easy to demand more from the people who are better off than you and attribute their success on luck, be it heretical or not. I am not going to say that I haven't been more privileged than most of the people on the planet. My ancestors made my country into what it is which gave me free education, low corruption, and in general a good start in life. However there are also a lot of people who had the same s…

> heretical

Do you mean "hereditary"? :D

Re: Post-GPT Computing

#445
post #192

Earlier quoted context omitted.

[flagged]

it begins with everyone has enough to begin with, and whatever is importent is available as public services. there are different flavors to do this, the currently best implementation is to have very high&progressive taxes that fund a strong public sector, that in turn makes people feel safe and allows them to pursue higher education (doctors, teachers, ...) that again are needed for all this. Sure you can work to hav…

> but not 100x more than your neighbor, nobody is worth that.

Hard disagree. Have you seen the average human, and compare him to somebody like John Carmak, Andrej Karpathy, etc. They create value at least 100x as big as the avg person

Re: Post-GPT Computing

#446

Earlier quoted context omitted.

>Cutting out that middle band of ability and resources is a surefire recipe for social unrest. I don't know, the US has been pushing that envelope for 40+ years and people are still paying taxes...

> I don't know, the US has been pushing that envelope for 40+ years and people are still paying taxes... And people can push the "meth consumption" envelope for years before they finally die from it, too. Getting away with unsustainable practices for X amount of time doesn't prove they're sustainable and won't end in collapse. It just means collapse can take more than X amount of time.

Nothing is sustainable. This is a concept invented to thwart questioning the environmental impacts of industrial activity. "Oh we can do it 'sustainably'".

To claim AI tools will strip out the wealth of the middle class and cause unrest can be disproved by the fact that it was beaten to it by the US corporations by over a generation.

The "middle class" was an invention of post war US to reward the soldiers who fought in Europe and the Pacific. Things are just reverting to a more long (as in multiple centuries) trend behavior of economies. Few haves, lots of have-nots and barely anything in between.

Re: Post-GPT Computing

#447

Earlier quoted context omitted.

I think your parent is on to something: it won't be the programmers trusting GPT a little too much, it will be the PMs, and there won't be a programmer :-)

I see an Idiocracy inspired future for programming. People claim that AI can write code so they start firing programmers. Universities stop software engineering programs as there is no one taking the courses. People stop writing blogs or stackoverflow. Software engineers either move to other fields or start living offgrid. No new innovation or new line of code written by human. Meanwhile, software quality get worse b…

It seems like you're assuming that the only way for it to get better is more human-written code. Also, why would software quality get worse over time, as opposed to stagnating at worst? Attempting to use it for more complex applications?

There is already so far beyond enough data available online for an AI to be a super-human senior software engineer/architect/whatever. Because there's enough for the humans, and an AI can do it better in theory. It just needs to be a good enough model.

Barring the most extreme edge cases, it's never going to be worse than it is right now.

Re: Post-GPT Computing

#448
post #25

Earlier quoted context omitted.

I mean this in the least snarky, most sympathetic way possible: it's time to level up. Countless roles have had to do this, it's now our turn.

I've been leveling up since I started as a helpdesk rep at 18, becoming a Windows sysadmin, then a junior dev, and now a senior dev. College was not an option for me for reasons totally outside of my control, as my parents decided to not do their taxes for a number of years and I was unable to get any financial aid (grants or loans) whatsoever. I scratched and clawed, read tons of books, blogs, spent extra time polis…

[dead]

Re: Post-GPT Computing

#449
post #265

Earlier quoted context omitted.

One of the reasons I'm not blown away is that everything we've tasked it to do has resulted in rather terrible answers. A lot of them outright didn't work. When they did, the way GPT created our solutions were often in a way that wouldn't work well over time. Unfortunately you wouldn't necessarily know that unless you really know your tools. In many ways, this isn't too different from people performing brute-force pr…

Are people actually allowing it to perform actual software development? State your assumptions. Read and summarize the pervious documents generate a data flow diagram. generate a data model. Get it to inquire about use cases and requirements generate tests for these uses cases and requirements.

I've played with getting it to ask me about my requirements. I think that was on ChatGPT3.5? It wasn't great, but it wasn't horrible.

Re: Post-GPT Computing

#450

Last week I used ChatGPT for the first time for a real world task at work. It was a self-contained lambda function to perform some admin tasks, so it seemed like an ideal fit. Although the experience was good, it's far from the end of programmers. This was my experience: * Although ChatGPT is pretty good at generating code, it kept making simple mistakes such as calling non-existing APIs or introducing bugs. Some of…

It already has wider attention span than humans, if anything it'll dominate at precisely more complex input very quickly with evolution counted in weeks not millions of years.

Let's wait for "And now... long term memory is all you need" paper.

Post reply on HN