4 ms·

I think people should do the opposite and move their code to GitHub if it helps Copilot. It's the single most important invention of my lifetime. Its existence

by Kiro 4y ago

I think people should do the opposite and move their code to GitHub if it helps Copilot. It's the single most important invention of my lifetime. Its existence trumps any ideological or legal hurdles.

bayindirh 4y ago
I'd rather keep my code away from a company and feature which disregards the licenses I prefer for my codebases, thank you.
- williamcotton 4y ago
  You might be surprised about the limitations of your copyright claims! https://en.wikipedia.org/wiki/Idea%E2%80%93expression_distinction https://en.wikipedia.org/wiki/Idea%E2%80%93expression_distin... https://en.wikipedia.org/wiki/Abstraction-Filtration-Comparison_test https://en.wikipedia.org/wiki/Abstraction-Filtration-Compari...
  - bayindirh 4y ago
    Copilot is using GPL licensed code snippets as starting points or for adding to other snippets as derivations. As a result, it's injecting GPL code inside to another code, or providing GPL code as is. At that point the code needs to be GPL licensed, or it's a breach of license. It's plain and simple. Triviality cause is hard to argue, because you can code a nice "trade secret", or an algorithm worthy of its paper in ~25 lines (at least I did). So Copilot's use of GPL licensed code is not fair use in my eyes. BTW, all my public code is (A)GPLv3+ licensed. I want my code to stay open, not closed.
    - williamcotton 4y ago
      Beyond notions of de minimis is the important distinction that copyright does not apply to utilitarian features! There are also notions of expressions that were already in the public domain and some copyright cases even refer to this as prior art.
      - bayindirh 4y ago
        You boldly assume that all GPL code is trivial and all the functions included in these codebases can be claimed to be on public domain. There are many GPL licensed codebases (e.g. Linux Kernel, GNU Octave, GNU R, etc.) which contains serious research and novel ideas and implementations. These codebases are esp. licensed that way to keep that research in the open. I have such a codebase which I was planning to open source with a GPL3.0 license, but I postponed that plans because of GitHub's Copilot shenanigans. As I said, the code contains many novel, yet short snippets which are worthy of their own research paper. Downplaying the research embedded in open codebases is just harmful to the discussion.
    - Kiro 4y ago
      Is looking at GPL code and rewriting it from scratch in your own style a license breach? For me, Copilot always uses the context and writes new code based on your own code as input. It uses the same patterns and code style as your own code.
      - bayindirh 4y ago
        Yes. When you write the same algorithm, even with a different style, it's a direct derivation of the code at hand. Derivative works of GPL licensed code are still GPL licensed. Moreover, Copilot doesn't always derive code, but provides the code as-is, reproducing it comment by comment [0], and with a wrong license while at that. [0]: https://twitter.com/mitsuhiko/status/1410886329924194309 https://twitter.com/mitsuhiko/status/1410886329924194309
        williamcotton 4y ago
        Only the expressive elements of an algorithm fall under copyright. Generally speaking math equations and algorithms can’t be restricted by copyright. Expressive, non-technical comments in code are most definitely protected by copyright.
        bayindirh 4y ago
        So, I can use a code from a leaked source code as I wish, and duplicate all the algorithms and methods for achieving something. Nice. Well, we're so naive for making clean room development for so long, and writing EULA's with sentences all in caps stating that this is a trade secret and you can't do anything with it. BTW, I want to reiterate that I'm trying to keep things open, not closed. So I'm not trying to copyright and protect it from being modified. This is what GPL is actually. So, the research is already there, in the form of a paper. You can re-implement that. But, my implementation is under GPLv3. So I'm not copyrighting/patenting the algorithm. It's certain expression I've implemented as a code, and opened under GPL. As a result, if you derive anything from my implementation, that's GPL too. If you want that algorithm, it's on paper. Go implement that. I don't care.
      - 4y ago
slickdork 4y ago
Why should we provide free data sets for github just to let them privatize gains?
nonrandomstring 4y ago
> trumps any ideological I don't think that word means what you think it means. The word "ideological" gets bandied about a lot here as a pejorative drop-in for "things I don't support". Perhaps we could be a bit less dramatic when considering the plurality of values people hold.
- dang 4y ago
  I use the word 'ideological' all the time here (https://hn.algolia.com/?dateRange=all&page=0&prefix=true&query=by%3Adang%20ideological&sort=byDate&type=comment https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...). Your comment got me thinking about what I mean by it. I guess I mean something like: a fixed cluster of ideas that people apply to quickly make good/bad judgments, often accompanied by strong emotions. Ideology tends to be generic (the same ideology gets applied to lots of things), predictable (a fixed idea lets one know where one stands in a predictable way), emotional (the need to know this is strongest on high-stakes issues), and binary (each ideology organizes itself in opposition to an opposing one). I use the term in moderation comments because these qualities are bad for HN discussion. The 'fixed' aspect makes discussion predictable, the 'generic' aspect makes it shallow, and the binary/emotional aspect makes it inflammatory. A bad trifecta for curious conversation.
  - nonrandomstring 4y ago
    These are very good parameters for encouraging positive discourse. and averting dull discussion that descends into a game of Top-Trumps - I play my ideological card, you play your idealogical card... ad nauseum. I'm not sure whether predictable, emotional and polar qualities are necessarily intrinsic to "ideologies" in the broad sense. Novel, calmly reasoned and nuanced commentary could be offered in support of positions so labelled. But I get your drift on "touchy topics that people are prone fly off the handle over". FWIW I think this has more to do with stated identity than abstract modes of thought these days. As a moderator I guess one gets a sense of the dynamics of a thread and when it's going to tip in a bad direction. As I see it, we've come through, and remain in, an interregnum for perhaps 50 years, in which the word has become dirtied by the failures of political and religious practices. And also by incessant and needy "wars" on <the next terrible thing>. A thematic analysis of news from the past 20 years would surely place "ideology" in close proximity to words like "radicalised" or "extremist". That is parochial. As a sceptic I'm possessed of a meta-ideology - that ideologies in themselves are healthy. We can have ideologies like economics, belief in space-faring exploration, benevolent technological society, self-determination in health and preventative medicine, and other powerful ideas which, in this forum, would be inarguably deemed positive. Perhaps, unless we propose abandoning thought as a way of being (in a Buddhist sense) ideology is just the inescapable waters in which we swim. I guess what bugs me is if people use "ideology" as a bare, dismissive accusation that attempts to cut-off further dialogue. If everything and anything, including scientific method, can be an "ideology" it ultimately adds nothing to the conversation. respects
carapace 4y ago
Although I don't agree with you I up-voted the comment because I think you're raising an important point. Automating code generation is part of the self-accelerating process of the techno-singularity, eh?
masukomi 4y ago
It could be easily argued that nuclear bombs are one of, if not the, most significant inventions in the past century, and their existence continues to have a singnificant geopolitical impact. I think it's safe to say that the vast majority of humanity is ideologically opposed to their use. BUT, they're one of, if not _the_ most important technological invention that currently effect us. So, that trumps ideological and legal hurdles, and thus we should use them.... right? I would like to presume you would disagree with that assertion. With that in mind, would you like to reconsider your justification for continuing to use GitHub? To be clear, I'm not arguing that you shouldn't use GitHub. I'm arguing that the importance, or impact, of an invention is irrelevant to its continued use. What's important is the short, and long term _effect_ of continuing to use it.
sgift 4y ago
All I see when I look at Copilot is code written by the most stupid coworker that could ever exist - and now, my job is to review it. I really hope we had better inventions in your lifetime than that.
xigoi 4y ago
Thanks, but I don't want a huge company to sell my code without giving me anything back.

by Kiro 4y ago

bayindirh 4y ago

I'd rather keep my code away from a company and feature which disregards the licenses I prefer for my codebases, thank you.

williamcotton 4y ago
You might be surprised about the limitations of your copyright claims! https://en.wikipedia.org/wiki/Idea%E2%80%93expression_distinction https://en.wikipedia.org/wiki/Idea%E2%80%93expression_distin... https://en.wikipedia.org/wiki/Abstraction-Filtration-Comparison_test https://en.wikipedia.org/wiki/Abstraction-Filtration-Compari...
- bayindirh 4y ago
  Copilot is using GPL licensed code snippets as starting points or for adding to other snippets as derivations. As a result, it's injecting GPL code inside to another code, or providing GPL code as is. At that point the code needs to be GPL licensed, or it's a breach of license. It's plain and simple. Triviality cause is hard to argue, because you can code a nice "trade secret", or an algorithm worthy of its paper in ~25 lines (at least I did). So Copilot's use of GPL licensed code is not fair use in my eyes. BTW, all my public code is (A)GPLv3+ licensed. I want my code to stay open, not closed.
  - williamcotton 4y ago
    Beyond notions of de minimis is the important distinction that copyright does not apply to utilitarian features! There are also notions of expressions that were already in the public domain and some copyright cases even refer to this as prior art.
    - bayindirh 4y ago
      You boldly assume that all GPL code is trivial and all the functions included in these codebases can be claimed to be on public domain. There are many GPL licensed codebases (e.g. Linux Kernel, GNU Octave, GNU R, etc.) which contains serious research and novel ideas and implementations. These codebases are esp. licensed that way to keep that research in the open. I have such a codebase which I was planning to open source with a GPL3.0 license, but I postponed that plans because of GitHub's Copilot shenanigans. As I said, the code contains many novel, yet short snippets which are worthy of their own research paper. Downplaying the research embedded in open codebases is just harmful to the discussion.
  - Kiro 4y ago
    Is looking at GPL code and rewriting it from scratch in your own style a license breach? For me, Copilot always uses the context and writes new code based on your own code as input. It uses the same patterns and code style as your own code.
    - bayindirh 4y ago
      Yes. When you write the same algorithm, even with a different style, it's a direct derivation of the code at hand. Derivative works of GPL licensed code are still GPL licensed. Moreover, Copilot doesn't always derive code, but provides the code as-is, reproducing it comment by comment [0], and with a wrong license while at that. [0]: https://twitter.com/mitsuhiko/status/1410886329924194309 https://twitter.com/mitsuhiko/status/1410886329924194309
      - williamcotton 4y ago
        Only the expressive elements of an algorithm fall under copyright. Generally speaking math equations and algorithms can’t be restricted by copyright. Expressive, non-technical comments in code are most definitely protected by copyright.
        bayindirh 4y ago
        So, I can use a code from a leaked source code as I wish, and duplicate all the algorithms and methods for achieving something. Nice. Well, we're so naive for making clean room development for so long, and writing EULA's with sentences all in caps stating that this is a trade secret and you can't do anything with it. BTW, I want to reiterate that I'm trying to keep things open, not closed. So I'm not trying to copyright and protect it from being modified. This is what GPL is actually. So, the research is already there, in the form of a paper. You can re-implement that. But, my implementation is under GPLv3. So I'm not copyrighting/patenting the algorithm. It's certain expression I've implemented as a code, and opened under GPL. As a result, if you derive anything from my implementation, that's GPL too. If you want that algorithm, it's on paper. Go implement that. I don't care.
    - 4y ago

slickdork 4y ago

Why should we provide free data sets for github just to let them privatize gains?

nonrandomstring 4y ago

> trumps any ideological I don't think that word means what you think it means. The word "ideological" gets bandied about a lot here as a pejorative drop-in for "things I don't support". Perhaps we could be a bit less dramatic when considering the plurality of values people hold.

dang 4y ago
I use the word 'ideological' all the time here (https://hn.algolia.com/?dateRange=all&page=0&prefix=true&query=by%3Adang%20ideological&sort=byDate&type=comment https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...). Your comment got me thinking about what I mean by it. I guess I mean something like: a fixed cluster of ideas that people apply to quickly make good/bad judgments, often accompanied by strong emotions. Ideology tends to be generic (the same ideology gets applied to lots of things), predictable (a fixed idea lets one know where one stands in a predictable way), emotional (the need to know this is strongest on high-stakes issues), and binary (each ideology organizes itself in opposition to an opposing one). I use the term in moderation comments because these qualities are bad for HN discussion. The 'fixed' aspect makes discussion predictable, the 'generic' aspect makes it shallow, and the binary/emotional aspect makes it inflammatory. A bad trifecta for curious conversation.
- nonrandomstring 4y ago
  These are very good parameters for encouraging positive discourse. and averting dull discussion that descends into a game of Top-Trumps - I play my ideological card, you play your idealogical card... ad nauseum. I'm not sure whether predictable, emotional and polar qualities are necessarily intrinsic to "ideologies" in the broad sense. Novel, calmly reasoned and nuanced commentary could be offered in support of positions so labelled. But I get your drift on "touchy topics that people are prone fly off the handle over". FWIW I think this has more to do with stated identity than abstract modes of thought these days. As a moderator I guess one gets a sense of the dynamics of a thread and when it's going to tip in a bad direction. As I see it, we've come through, and remain in, an interregnum for perhaps 50 years, in which the word has become dirtied by the failures of political and religious practices. And also by incessant and needy "wars" on <the next terrible thing>. A thematic analysis of news from the past 20 years would surely place "ideology" in close proximity to words like "radicalised" or "extremist". That is parochial. As a sceptic I'm possessed of a meta-ideology - that ideologies in themselves are healthy. We can have ideologies like economics, belief in space-faring exploration, benevolent technological society, self-determination in health and preventative medicine, and other powerful ideas which, in this forum, would be inarguably deemed positive. Perhaps, unless we propose abandoning thought as a way of being (in a Buddhist sense) ideology is just the inescapable waters in which we swim. I guess what bugs me is if people use "ideology" as a bare, dismissive accusation that attempts to cut-off further dialogue. If everything and anything, including scientific method, can be an "ideology" it ultimately adds nothing to the conversation. respects

carapace 4y ago

Although I don't agree with you I up-voted the comment because I think you're raising an important point. Automating code generation is part of the self-accelerating process of the techno-singularity, eh?

masukomi 4y ago

It could be easily argued that nuclear bombs are one of, if not the, most significant inventions in the past century, and their existence continues to have a singnificant geopolitical impact. I think it's safe to say that the vast majority of humanity is ideologically opposed to their use. BUT, they're one of, if not _the_ most important technological invention that currently effect us. So, that trumps ideological and legal hurdles, and thus we should use them.... right? I would like to presume you would disagree with that assertion. With that in mind, would you like to reconsider your justification for continuing to use GitHub? To be clear, I'm not arguing that you shouldn't use GitHub. I'm arguing that the importance, or impact, of an invention is irrelevant to its continued use. What's important is the short, and long term _effect_ of continuing to use it.

sgift 4y ago

All I see when I look at Copilot is code written by the most stupid coworker that could ever exist - and now, my job is to review it. I really hope we had better inventions in your lifetime than that.

xigoi 4y ago

Thanks, but I don't want a huge company to sell my code without giving me anything back.