4 ms·

I didn’t really get the impression here that Jared wanted a social network. He just wanted a better mechanism for discovering good quality and interesting proje

I didn’t really get the impression here that Jared wanted a social network. He just wanted a better mechanism for discovering good quality and interesting projects. Popularity is often one good signal.

callmekit 6y ago
Quote from the article: "I’d also be interested to see more social network-like features on GitHub".
marcus_holmes 6y ago
I'm not sure Github stars were/are ever an indicator of "good quality" or even "interesting". And I think the Web Development community could do with a decent break from "new" as well. What would be great is if we could just prune some of the gajillion frameworks and libraries already out there, and spend some time to make the ones we keep decently robust and elegant.
- majewsky 6y ago
  GitHub stars suffer from the same problem as likes on Twitter or Facebook. A social network would actually need three separate kinds of positive interactions: - a like ("I want to let the creator know I like this.") - a share ("I want my followers to know about this.") - a bookmark ("I want to be able to find this again later.") Most social networks only have the first two. GitHub, even worse, only has one. (I personally use stars exclusively as bookmarks, btw.)
  - IggleSniggle 6y ago
    Yes, exactly. I'm more likely to star a small project that I find interesting even if flawed because I don't want to lose track of it. I don't really understand why I would need or want to star React or Nodejs for example. There is zero chance I lose track of those projects.
    - ProZsolt 6y ago
      For me, this is a positive thing. With this, I can find some niche things that I can't find otherwise. People star it because it provides some value to them. I wouldn't "like" project that are broken or half done, but I star it if it can solve my problem. On the other side, when somebody stars my project I will invest more energy into it because I see that it is useful to somebody else, not just me. Most of my projects are sparsely tested, because usually I'm the only user and it is not mission-critical. When it's broken I usually know where. When I see starts appearing on my project I add more tests and handle edge cases better, so new users have a better experience. If they would only bookmark it, I wouldn't know if somebody wants to use my software. And they would also have a bad experience, and maybe drop it after evaluating as a piece of crap.
      - IggleSniggle 6y ago
        I agree.

callmekit 6y ago

Quote from the article: "I’d also be interested to see more social network-like features on GitHub".

marcus_holmes 6y ago

I'm not sure Github stars were/are ever an indicator of "good quality" or even "interesting". And I think the Web Development community could do with a decent break from "new" as well. What would be great is if we could just prune some of the gajillion frameworks and libraries already out there, and spend some time to make the ones we keep decently robust and elegant.

majewsky 6y ago
GitHub stars suffer from the same problem as likes on Twitter or Facebook. A social network would actually need three separate kinds of positive interactions: - a like ("I want to let the creator know I like this.") - a share ("I want my followers to know about this.") - a bookmark ("I want to be able to find this again later.") Most social networks only have the first two. GitHub, even worse, only has one. (I personally use stars exclusively as bookmarks, btw.)
- IggleSniggle 6y ago
  Yes, exactly. I'm more likely to star a small project that I find interesting even if flawed because I don't want to lose track of it. I don't really understand why I would need or want to star React or Nodejs for example. There is zero chance I lose track of those projects.
  - ProZsolt 6y ago
    For me, this is a positive thing. With this, I can find some niche things that I can't find otherwise. People star it because it provides some value to them. I wouldn't "like" project that are broken or half done, but I star it if it can solve my problem. On the other side, when somebody stars my project I will invest more energy into it because I see that it is useful to somebody else, not just me. Most of my projects are sparsely tested, because usually I'm the only user and it is not mission-critical. When it's broken I usually know where. When I see starts appearing on my project I add more tests and handle edge cases better, so new users have a better experience. If they would only bookmark it, I wouldn't know if somebody wants to use my software. And they would also have a bad experience, and maybe drop it after evaluating as a piece of crap.
    - IggleSniggle 6y ago
      I agree.