Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GitHub Copilot

copilot.github.com

Post reply on HN
Print view View original

571–580 of 1001 posts

Re: GitHub Copilot

#571

Post by ayush--s »

wow I'm going to have lots of opinions about this.

1. A lot of people on this thread are concerned about licensing issues with GPL etc. I am sure Github will restrict the beta until it figures out that stuff.

2. I wonder if eventually our corrections to the code suggested by the model would be used to feedback to the model, and if that'll lead to a differential pricing - If I let it see my code, I get charged lesser.

3. I believe a mini-GPT-3 model is where it's at. GPT-3 (and similar) models look to be to too big to run locally. I've been using TabNine for past year or so & it gives me anywhere between 5-10% productivity boost. But one of the main reasons why it works so well is because it trains on my repo as well. TabNine is based off GPT-2 from what i've heard.

4. prediction: Microsoft is probably going to milk GPT-3. Expect a bumpy ride.

5. In all likeliness, this would be a great tool to make developers productive, rather than take their jobs - at least at levels that are more than just code-coolie.

6. Eventually all tasks with enough data around it will see automation using AI.

Re: GitHub Copilot

#572

Post by TeMPOraL »

hintymad wrote: post #486
inimino wrote: post #385

I'm amazed to see how positive the overall response is to this idea. Almost as if programmers think that writing programs is the worst part of the job and ready to be automated away. As someone more aligned with the Dijkstra perspective, this seems to me like one of the single worst ideas I've ever seen in this domain. We already have IDEs and other tools leading to an increase in boilerplate and the acceptance of it…

Somehow I don't see people discuss this kind of tools from the perspective of managing essential complexity versus accidental complexity. Maybe copilot just increases the abstraction level of coding, so we can treat generated code as a building block, just like we nowadays rarely needs to care about how to write assembly code or how a balanced tree works?

> Maybe copilot just increases the abstraction level of coding, so we can treat generated code as a building block

At this point it doesn't, and we can't, because Copilot is just a fancy autocomplete. The code is there, first class, in your file. It doesn't introduce new concepts for you, it just tries to guess what you mean by function signature + descriptive comments, and generates that code for you.

Re: GitHub Copilot

#574

Post by jka »

throwaway675309 wrote: post #454
inimino wrote: post #385

I'm amazed to see how positive the overall response is to this idea. Almost as if programmers think that writing programs is the worst part of the job and ready to be automated away. As someone more aligned with the Dijkstra perspective, this seems to me like one of the single worst ideas I've ever seen in this domain. We already have IDEs and other tools leading to an increase in boilerplate and the acceptance of it…

I mean you say this, but you and most likely the majority of programmers rely on dozens of repositories, packages and libraries with likely zero deep understanding of it (and at the very least haven't read the source code of ) so I don't really understand the difference here. The advantage of something like this is that instead of having to go to stack overflow or any number of reference sites and copy pasta it can j…

I'd be willing to bet a reasonable amount that there's a large future for "subtractive software development" (maybe a slightly misleading or unfair term, since it'd include bugfixes).

Once we have multiple proven technologies that handle each of the functional areas that we collectively need, then we'll start to find greater benefit in maintenance, bugfixes, and performance improvements for those existing technologies and their dependencies than we find writing additional code and libraries.

Re: GitHub Copilot

#575

Post by drusepth »

fabiospampinato wrote: post #108
fzaninotto wrote: post #64

Earlier quoted context omitted.

It shows the suggestions line by line, and only shows the best guess. It's not more intrusive than Intellisense. You can actually see all the code blocks Copilot is thinking about if you want to, but that is indeed a distraction.

The problem I see with that is that's not possible for it to understand well which code is the best, GPT-3 is trying to mimic human writing in general, the thing is most human code is garbage, if this system was able to understand how to make code better you could keep training it until you had perfect code, which is not what the current system is giving you (a lot of the times anyway).

>if this system was able to understand how to make code better you could keep training it until you had perfect code

Based on the FAQ, it looks like some information about how you interact with the suggestions is fed back to the Copilot service (and theoretically OpenAI) to better improve the model.

So, while it may not understand "how to make code better" on its own, it can learn a bit from seeing how actual devs do make code better and theoretically improve from usage.

Re: GitHub Copilot

#576

Post by hbosch »

inimino wrote: post #385

I'm amazed to see how positive the overall response is to this idea. Almost as if programmers think that writing programs is the worst part of the job and ready to be automated away. As someone more aligned with the Dijkstra perspective, this seems to me like one of the single worst ideas I've ever seen in this domain. We already have IDEs and other tools leading to an increase in boilerplate and the acceptance of it…

If we didn't need programmers to do the programming, that would be a perfect world.

Re: GitHub Copilot

#577

Post by visarga »

asdfman123 wrote: post #309
fzaninotto wrote: post #40

I've been using the alpha for the past 2 weeks, and I'm blown away. Copilot guesses the exact code I want to write about one in ten times, and the rest of the time it suggests something rather good, or completely off. But when it guesses right, it feels like it's reading my mind. It's really like pair programming, even though I'm coding alone. I have a better understanding of my own code, and I tend to give better na…

Pack it all up, boys, programming's over. Hello, AI. Anyone want to hire me to teach your grandma how to use the internet?

> Anyone want to hire me to teach your grandma how to use the internet?

Only for the first time to train a model for that.

Re: GitHub Copilot

#578

Post by TeMPOraL »

jokethrowaway wrote: post #567
anyonecancode wrote: post #566

Earlier quoted context omitted.

> I wonder how long it will be until someone sets up the rest of the feedback loop and starts putting up github projects made of nothing but code from this tool, and it can start to feed on itself. This is my actual hoped-for endgame for the ad based internet. At some point Twitter, FB, etc will be exclusively populated by bots that post ads, and bots that simulate engagement with those ads to drive up perceived valu…

The most unlikely thing you mentioned is that we will be able to tax huge corporations.

You jest, but the parent is close to what I believe is a possible scenario - the one Nick Bostrom calls "a Disneyland with no children".

No tax, no flying cars, eventually not even humans around - just AI-driven companies endlessly trading with each other, in a fully-automated, self-contained, circular economy, from which sentient beings were optimized away.

Re: GitHub Copilot

#579

Post by nindalf »

anyonecancode wrote: post #566
inimino wrote: post #385

I'm amazed to see how positive the overall response is to this idea. Almost as if programmers think that writing programs is the worst part of the job and ready to be automated away. As someone more aligned with the Dijkstra perspective, this seems to me like one of the single worst ideas I've ever seen in this domain. We already have IDEs and other tools leading to an increase in boilerplate and the acceptance of it…

> I wonder how long it will be until someone sets up the rest of the feedback loop and starts putting up github projects made of nothing but code from this tool, and it can start to feed on itself. This is my actual hoped-for endgame for the ad based internet. At some point Twitter, FB, etc will be exclusively populated by bots that post ads, and bots that simulate engagement with those ads to drive up perceived valu…

I invite you to try automating this and let us know what happens. Try creating, let’s say 1000 accounts and try liking or posting and see what happens. I’ve seen that system at work and doubt you’d get very far.

More than that, you misunderstand how advertisers prioritise their money. They pay for outcomes. If they notice that over the past couple of months they’ve been receiving mostly bot traffic, they stop advertising. Not everyone all at once, but enough that revenue begins to decline. An ad based business that cares about the long term will do it’s best to weed out the inauthentic engagement.

Post reply on HN