5 ms·

This may be a symptom of the hyper-connective world we live in today. You observe this phenomenon elsewhere outside of academia as well. There is an intense pre

by leksak 10y ago

This may be a symptom of the hyper-connective world we live in today. You observe this phenomenon elsewhere outside of academia as well. There is an intense pressure to distinguish oneself from the greater body of people occupying the same industry as yourself. Consider that a vibrant Github profile is essentially a prerequisite to be considered as an engaged professional in software engineering these days. It doesn't matter if you have ethical qualms about Github, or spend most of your day doing... your day job. A 22-year old shouldn't have to fret about not having enough public repositories or not having contributed enough to open-source. I'd expect any craftsman to start producing their best work well into their career, and not at the start of it.

trendia 10y ago
Ethical qualms about Github? Could you explain? Should I worry about hosting my code there?
- leksak 10y ago
  Beyond uptime concerns (which Github manages very well) is the fact that it is closed source, and whenever you use a closed system but that provides very good services for free you yourself is the commodity. Beyond that, dozzie's remark is very much on point, the host could become tainted in the same way as for instance SourceForge. Fortunately, the distributed nature of Git means that you are unlikely to "lose" your source-code. But really you have no control over your code, it is their liberty to censor it if they please. The primary issue, as far as I am concerned, is that while I can clone and go elsewhere is that Github is the centralized hub for (FL)OSS today. Everything lives there. If you are not on there, you are de facto invisible. A better alternative, as far as I am considered, would be a distributed net of "self-hosted" Gitlab instances. I am not comfortable with all the bad press that Github has suffered regarding gender equality, but I feel that I am suffering from vendor lock-in due to the success of the platform.
  - hilop 10y ago
    If GitHub censors your code, or even if they don't, you can publish it elsewhere
    - leksak 10y ago
      Absolutely, but that is part of the ethical qualm. You are perpetuating the use of Github as the dominant (code) social network by using it. If you continue to use it after Github has performed an act of censorship (or anything else that violates your values) and you continue to use the service then that act goes unpunished as they will keep their market share and may repeat that act again. After all, there weren't enough negative consequences to avoid doing so in the future.
- dozzie 10y ago
  Always. It's a third-party place that can sink at any time (if you don't believe this, why would GitHub be any different to what happened to SourceForge or Google Code?). Your code should be hosted under your very own domain first (so called "canonical repository", or how I call it, the "source of truth"), and GitHub or BitBucket should be a replica. Thanks to git's distributed nature, this replica can be two-way, but still a replica.
  - hilop 10y ago
    Google Code shut down, and everything of interest moved. There was no problem.
    - leksak 10y ago
      Despite the altruistic acts of certain people, not every project was migrated. Those that weren't migrated are now lost and a part of our digital heritage died. I consider that a problem.
- gaius 10y ago
  Possibly a reference to https://news.ycombinator.com/item?id=11049067 https://news.ycombinator.com/item?id=11049067
  - OJFord 10y ago
    Changing the de facto social network is so hard though - no matter the bad press on Github's awful diversity policy ("not for whites to lead"; "white women are the problem") or recent research take-down (Gitlab followed suit, then retracted and apologised) or even that its closed-source and Gitlab is (can be) open. If they both started from nothing today, there's no doubt in my mind that we'd all quickly settle on Gitlab, but affecting that shift when Github is such an established default is so much harder.
    - gaius 10y ago
      If they both started from nothing today, there's no doubt in my mind that we'd all quickly settle on Gitlab, but affecting that shift when Github is such an established default is so much harder. Absolutely. Just yesterday I wanted to make a small change to a popular open source package, about 10 lines of C. Forked on Github, then sent a Pull Request. Because that's where the project lives, so that's where everyone goes. And Github has no interest in federating PRs in a way that they could be sent from Gitlab.
      - leksak 10y ago
        To be fair, Gitlab doesn't really have an interest in federating PRs either. But, "we" can make that be a public utility that exists.
    - leksak 10y ago
      Aye, the https://en.wikipedia.org/wiki/Network_effect https://en.wikipedia.org/wiki/Network_effect is a powerful force in dictating human behaviour.
ap22213 10y ago
Today's business requirement that we look productive and emphasize (re)action over observation, thought and planning is the major reason why I often consider leaving the work force entirely (at almost 40). Although it's not often reflected upon these days, innovative, useful software was actually being created before, say, 2005. And, that software wasn't being built at 48-hour hackathons. And, it wasn't being bolted on to an MVP. It was acceptable then (and usually required) that a knowledgeable and skilled team spend a lot of time contemplating requirements and design. I'm proud that I had created important, useful software over the years. Some of it had use for over a decade. Some of it is still in use today. But, I often joke with friends that I never would have been hired today, and I never would have been accepted to the same university program.
- leksak 10y ago
  When I was growing up it seemed to be common knowledge that substantial pieces of work take time, and therefore it isn't likely for one to produce many substantial pieces of work during a lifetime. It remains to be seen if this unmanageable pace that we as an industry have adopted will make us crash and burn hard enough to not repeat our mistakes but as a species, we seemed doomed to repeat ourselves so I am slightly skeptical. I'm glad to hear that you have works of pride.

by leksak 10y ago

trendia 10y ago

Ethical qualms about Github? Could you explain? Should I worry about hosting my code there?

leksak 10y ago
Beyond uptime concerns (which Github manages very well) is the fact that it is closed source, and whenever you use a closed system but that provides very good services for free you yourself is the commodity. Beyond that, dozzie's remark is very much on point, the host could become tainted in the same way as for instance SourceForge. Fortunately, the distributed nature of Git means that you are unlikely to "lose" your source-code. But really you have no control over your code, it is their liberty to censor it if they please. The primary issue, as far as I am concerned, is that while I can clone and go elsewhere is that Github is the centralized hub for (FL)OSS today. Everything lives there. If you are not on there, you are de facto invisible. A better alternative, as far as I am considered, would be a distributed net of "self-hosted" Gitlab instances. I am not comfortable with all the bad press that Github has suffered regarding gender equality, but I feel that I am suffering from vendor lock-in due to the success of the platform.
- hilop 10y ago
  If GitHub censors your code, or even if they don't, you can publish it elsewhere
  - leksak 10y ago
    Absolutely, but that is part of the ethical qualm. You are perpetuating the use of Github as the dominant (code) social network by using it. If you continue to use it after Github has performed an act of censorship (or anything else that violates your values) and you continue to use the service then that act goes unpunished as they will keep their market share and may repeat that act again. After all, there weren't enough negative consequences to avoid doing so in the future.
dozzie 10y ago
Always. It's a third-party place that can sink at any time (if you don't believe this, why would GitHub be any different to what happened to SourceForge or Google Code?). Your code should be hosted under your very own domain first (so called "canonical repository", or how I call it, the "source of truth"), and GitHub or BitBucket should be a replica. Thanks to git's distributed nature, this replica can be two-way, but still a replica.
- hilop 10y ago
  Google Code shut down, and everything of interest moved. There was no problem.
  - leksak 10y ago
    Despite the altruistic acts of certain people, not every project was migrated. Those that weren't migrated are now lost and a part of our digital heritage died. I consider that a problem.
gaius 10y ago
Possibly a reference to https://news.ycombinator.com/item?id=11049067 https://news.ycombinator.com/item?id=11049067
- OJFord 10y ago
  Changing the de facto social network is so hard though - no matter the bad press on Github's awful diversity policy ("not for whites to lead"; "white women are the problem") or recent research take-down (Gitlab followed suit, then retracted and apologised) or even that its closed-source and Gitlab is (can be) open. If they both started from nothing today, there's no doubt in my mind that we'd all quickly settle on Gitlab, but affecting that shift when Github is such an established default is so much harder.
  - gaius 10y ago
    If they both started from nothing today, there's no doubt in my mind that we'd all quickly settle on Gitlab, but affecting that shift when Github is such an established default is so much harder. Absolutely. Just yesterday I wanted to make a small change to a popular open source package, about 10 lines of C. Forked on Github, then sent a Pull Request. Because that's where the project lives, so that's where everyone goes. And Github has no interest in federating PRs in a way that they could be sent from Gitlab.
    - leksak 10y ago
      To be fair, Gitlab doesn't really have an interest in federating PRs either. But, "we" can make that be a public utility that exists.
  - leksak 10y ago
    Aye, the https://en.wikipedia.org/wiki/Network_effect https://en.wikipedia.org/wiki/Network_effect is a powerful force in dictating human behaviour.

ap22213 10y ago

Today's business requirement that we look productive and emphasize (re)action over observation, thought and planning is the major reason why I often consider leaving the work force entirely (at almost 40). Although it's not often reflected upon these days, innovative, useful software was actually being created before, say, 2005. And, that software wasn't being built at 48-hour hackathons. And, it wasn't being bolted on to an MVP. It was acceptable then (and usually required) that a knowledgeable and skilled team spend a lot of time contemplating requirements and design. I'm proud that I had created important, useful software over the years. Some of it had use for over a decade. Some of it is still in use today. But, I often joke with friends that I never would have been hired today, and I never would have been accepted to the same university program.

leksak 10y ago
When I was growing up it seemed to be common knowledge that substantial pieces of work take time, and therefore it isn't likely for one to produce many substantial pieces of work during a lifetime. It remains to be seen if this unmanageable pace that we as an industry have adopted will make us crash and burn hard enough to not repeat our mistakes but as a species, we seemed doomed to repeat ourselves so I am slightly skeptical. I'm glad to hear that you have works of pride.