galaxyLogic

I'm not a Haskeller but I would like to use a language that indeed separates all code into two areas, one in which I can write non-pure functions and one in which I can not.

That would obviously make it easier to analyze and test and verify the program by first focusing on the pure part of it, then on the impure part. I'm not sure if that is possible in "strict languages" but perhaps.

True. Just saying we learn not only by lots of trial and error, but also by trying to imitate others.

Interesting. Those channels seem to all manage side-effects that persist after program execution. What about state-changes that are made and persist into the memory of the running program? Or is there any need for that?

So is there something like IO-Monad in Coalton?

I think that's the greatest feature of Haskell. Divide every program into two parts, one that can have side-effects and one that can not.

People here are saying you must write a lot to get good. Makes sense. But what about also reading a lot to learn from masters?

How do LLMs do it? They don't learn by writing a lot they learn by reading a lot.

I see makes sense. Then it it kinda says that quality of the model is the topicality of its inputs I assume.

> frontier models getting more and more out of date

Why are they getting out of date? Is it because we have new content from the internet that the older models did not have? Or are we simply trying to increase the size of the training data? In other words not more up-todate in terms of time the content was created vs. wanting to use bigger training-input-sets?

Does it matter what is the difference in size of needed inputs for inference vs. training?

When you ask people for an answer you are asking for their help. But if they delegate to AI they are in essence saying they don't want to make the effort to help you. They are making it LOOK LIKE they are doing something to help you by hiding the fact the answer is direct from AI. It is cheating.

"... prior LLM usage is negatively and significantly correlated with written communication and vibe coding performance. We speculate that this relationship is caused either by (i) LLMs having a negative effect on students’ ability to express themselves or (ii) that students who are less proficient at writing are more likely to use LLMs,"

Makes sense right? If you let the AI write and express ideas for you you are not really practicing that skill yourself. It is just like asking someone else do your work for you, you won't learn anything about it.

This interpretation would mean that you could not produce AGPL licensed code which loads some code in dynamically. Like say you couldn't have an AGPL licensed web-browser which sometimes loads pages whose JavaScript's source-code is obscured. ?

I much agree. But I wonder shouldn't the citations all be hyperlinks and thus easy to verify?

> If you want to switch from notepad to Zed, you need to learn how language servers work.

Can you not use Zed without knowing how language servers work?

Or that the AI is the REPL?

(In Haskell) > just adding a simple print somewhere is not going to work without refactor

Interesting. How do people cope with this in practice? Does it mean you can't really use log() -statements for debugging?

The Law of Leaky Abstractions. What is a "leaky" abstraction? How does it "leak"?

I wonder if it should be called "Law of Leaky Metaphors" instead. Metaphor is not the same thing as Abstraction. I can understand a "leaky metaphor" as something that does not quite make it, at least not in all aspects. But what would be a good EXAMPLE of a Leaky Abstraction?

> I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

I think no real human would ask such a question. Or if we do we maybe mean should I drive some other car than the one that is already at the car-wash?

A human would answer, "silly question ". But a human would not ask such a question.

I think the AI can not understand the situation of the student so it does not know what the student does not know. Therefore it can not guide the student through the main hurdles of learning and understanding a topic. Whereas a human tutor was once a human-student, AI never was.

Not so clear from their page but from

https://www.tipranks.com/news/amd-stock-slips-despite-a-majo...

I read:

" In addition to that, the update allows these agents to be turned into desktop apps for multiple operating systems. "

This seems like a new way to create app: Create an (AI) app that creates apps.

> Do you think the people who made it possible, who spent years of their life building and maintaining open source code, will be rewarded?

I think they should be rewarded more than they are currently. But isn't the GNU Public License bassically saying you can use such source-code without giving any rewards what so ever?

But I see your The reward for Open Source developers is the public recognition for their works. LLMs can take that recognition away.

So it's a bit as if Linux Organization told its contributors you can bring in infringing code but you must agree you are liable for any infringement?

But if a lawsuit was later brought who would be sued? The individual author or the organization? In other words can an organization reduce its liability if it tells its employees "You can break the law as long as you agree you are solely responsible for such illegal actions?

It would seem to me that the employer would be liable if they "encourage" this way of working?

So if creative effort was put into writing the prompt, then whoever wrote the prompt should have the copyright to the output produced by ChatGPT?

That would depend on whether those who sold you the software-output, had copyright to it.

Good. I don't doubt your good intensions. Just thinking that even though you would have paid her in the next convenient occasion, she would still have suffered a (semi-) financial loss because of the increased risk she was put under, of not getting her monmey back if you were run over by bus. Cheers :-)

Makes sense to me. But so anybody can take Public Domain code and place it under GNU Public License (by dropping it into a Linux source-code file) ?

Surely the person doing so would be responsible for doing so, but are they doing anything wrong?

What if I'm a sculptor and I design and produce a statue? Shouldn't I still have the copyright to the statue, no matter what kind of machine I used to do the actual sculpting?

But then if AI output is not under GNU General Public License, how can it become so just because a Linux-developer adds it to the code-base?

Maybe she just needed the money. Paying later always has the risk that it will never going to happen. And she helped you, did she get anything in return?

> is LLM generated code a derivative work of the prompts?

Very good question I would think it is. You are just using a mechanical system to transform your prompt to something else, Right?

But, a distiguishing factor may be that:

1. Output of the LLM for the same prompt can vary

2. So you don't really have "control" over what the AI produces

3. Therefore you should not get a copyright to the output of the LLM because you had very little to say about how that transformation (from prompt to code) was made.

I think you're wrong but I may be wrong too, so if you have a link that talks about this formal registration requirement it would be great.