10 ms·

Fork if you don't like it. That's the whole point of free software. Edit: I find it disturbing that so many people want to erase hacker culture from the inter

Fork if you don't like it. That's the whole point of free software. Edit: I find it disturbing that so many people want to erase hacker culture from the internet. Consensus is a Bay Area idea, one that came quite late to the world of computers, and one that many of us disagree with.

MBCook 8y ago
That’s the really odd thing in this situation. The maintainers all seem to want to remove it, and the project governance says the maintainers are supposed to decide based on the quorum. RMS just over ruled that. Because it was his joke and he felt like it. Which seems like it goes against the whole “community“ thing.
- zeth___ 8y ago
  If you read the waiver you have to sign before contributing anything to the fsf you surrender all your copyrights to them. This is so they can keep moving software into newer GPL versions. The maintainers are just mechanics working on someone else's car and have as much say in choosing the interior decor. The beauty of the system is that if they get sick of that state of affairs they get an identical copy of the car for free and can then do whatever they want to it.
  - josefx 8y ago
    >This is so they can keep moving software into newer GPL versions. That is already an optional part of the GPL itself afaik.
- gpm 8y ago
  The maintainers don't own the project though, GNU (RMS) does. If the maintainers want to control the project they are maintaining it does seem that they need to fork it and continue under a new name. (Or at least that's what I got from this thread. Certainly it's possible that GNU/RMS/Whoever originally had rights has passed those rights to the maintainers, but it doesn't happen simply by virtue of them maintaining it.)
lsh 8y ago
No, that's just another option when it comes to free software. Why not make an effort to cohere rather than disperse? Stronger together and all that. RMS is pulling the authority card, which is always a dodgy thing to do, and people want an egalitarian project rather than an authoritarian one. Both sides need to learn when to pick a fight and when to let something pass.
- zeth___ 8y ago
  Because GNU has always worked that way and given us some of the best software ever written.
  - geofft 8y ago
    Worth noting that basically every Linux distro for many years used the eglibc fork, until the former glibc maintainer stepped down, and it's hardly the only GNU project that is what it is today because it managed to reabsorb a fork by admitting a mistake. GCC comes to mind. GNU is what it is today largely because of the name recognition, and people's willingness to get stuff incorporated into GNU because of that name recognition. eglibc was very conservative about patches because it aimed for reunification eventually, but it didn't have to be.
- t3soro 8y ago
  > some people want an egalitarian project rather than an authoritarian one Plenty of us prefer the authoritarian model, rather than the chaotic model of arbitrary consensus.
  - MBCook 8y ago
    Which is totally fair. But that’s not how this project is set up. And that’s what makes this so odd.
sneak 8y ago
Forking out any vestigal rms authority would benefit everyone. They’re bikeshedding over whether or not to include a comment on a controversial topic. It is not relevant to the functioning of the library and is just a distraction. Rip it out and ban rms for perpetuating such idiotic drama.
- danmg 8y ago
  RMS isn't the one bikeshedding it.
  - olliej 8y ago
    he's the one forcing it -- he's not an active member of libc anymore, he's not a maintainer, but he decided that his personal joke from 20 years ago absolutely must remain in the documentation because he's petty. He sounds like the one in need of a safe space where nothing ever changes
    - danmg 8y ago
      It's more like he's preventing bikeshedding by stopping them from making irrelevant non-technical decisions.
      - josefx 8y ago
        There was already a consens before he decided to contribute his own opinion to a triviality. By doing so he prolonged the discussion at least a bit. Forcing your opinion about a pointless and otherwise done issue on others seems like textbook bikeshedding to me. At leaast now he can honestly say he is still actively involved with glibc.
justizin 8y ago
> Consensus is a Bay Area idea [citation needed] Merriam-Webster says the first known use was 1843, at https://www.merriam-webster.com/dictionary/consensus https://www.merriam-webster.com/dictionary/consensus. An SCM change, by very nature, doesn't "erase" anything, it only creates a record of change. The repository will always have the history.
readme 8y ago
Excellent idea! Just rename it political-libc for all the people who want politics mixed in with their C library.
- favorited 8y ago
  You realize it's the current version of glibc which has the political joke? The patch that RMS is opposing here removed a political reference from the docs.
- johannes1234321 8y ago
  The FSF and its GNU project are a political endeavor. For a less-political (non-political doesn't exist) libc one might fork into linux-libc or similar. (I'm not saying that's appropriate thing to do, as I'm just a lurker enjoying the free work by others in glibc's regard, but wanted to make the point)
  - dragonwriter 8y ago
    > The FSF and its GNU project are a political endeavor. But neither has anything about abortion policy in it's public declaration of goals. So, while some political messages might make sense in GNU technical docs, I don't see why a warning making an oblique reference to the Mexico City policy is not only acceptable but so critical that central leadership needs to overrule the specific project's maintainers to preserve it.
  - throwajay87492 8y ago
    These constant attempts to assert that "everything is political" are getting really old. All people have the capacity to be political. Nothing else is. I wish I could understand who taught you people that insistence of stupid horseshit is the key to achieving your goals.
    - johannes1234321 8y ago
      First of all I didn't say that "everything is political" but that in each organisation there is a degree of politics. Maybe the issue is in the word "politics" - this is not about left vs right, but social collaboration. Some people like each other more than others, some are more trustworthy to others and often the way I propose a change and whom I talk to has impact of the change being accepted - those are politics. Also technical decisions have an impact on different people and in subtle ways impact users (i.e. if a kernel switches to a high-res framebuffer mode to print boot messages and developers assume they have the width for formatting messages this impacts visually impaired people, willingly or not -- just made the example up to show subtle things) Politics aren't bad and are an effect of different people being present in the same space, while of course there is a difference between FSF/GNU/RMS with a strong focus on a political mission and some other projects primarily focussing on a technical mission.
geofft 8y ago
"Hacker culture" is a myth - not in the sense that it's false, but in the sense that it's a set of stories that people like because of their storytelling power, and its exact historical truth is not particularly relevant. As a myth, it tends towards better remembering things that make better stories, like great men inventing things through sheer force of will. And as a myth, it's a story about our origins, which were no doubt messier than we would like to remember. That our present does not seem as awesome as the myth is an unavoidable effect of myths. But there a few historically true things that are relevant here: first, that an important part of "hacker culture" unequivocally originated in the Bay Area (Fairchild Semiconductor, Homebrew Computer Club, BSD, etc.), and second, that "consensus" has been an important idea in this culture for a while - an obvious example being the IETF's "rough consensus and running code".

MBCook 8y ago

That’s the really odd thing in this situation. The maintainers all seem to want to remove it, and the project governance says the maintainers are supposed to decide based on the quorum. RMS just over ruled that. Because it was his joke and he felt like it. Which seems like it goes against the whole “community“ thing.

zeth___ 8y ago
If you read the waiver you have to sign before contributing anything to the fsf you surrender all your copyrights to them. This is so they can keep moving software into newer GPL versions. The maintainers are just mechanics working on someone else's car and have as much say in choosing the interior decor. The beauty of the system is that if they get sick of that state of affairs they get an identical copy of the car for free and can then do whatever they want to it.
- josefx 8y ago
  >This is so they can keep moving software into newer GPL versions. That is already an optional part of the GPL itself afaik.
gpm 8y ago
The maintainers don't own the project though, GNU (RMS) does. If the maintainers want to control the project they are maintaining it does seem that they need to fork it and continue under a new name. (Or at least that's what I got from this thread. Certainly it's possible that GNU/RMS/Whoever originally had rights has passed those rights to the maintainers, but it doesn't happen simply by virtue of them maintaining it.)

lsh 8y ago

No, that's just another option when it comes to free software. Why not make an effort to cohere rather than disperse? Stronger together and all that. RMS is pulling the authority card, which is always a dodgy thing to do, and people want an egalitarian project rather than an authoritarian one. Both sides need to learn when to pick a fight and when to let something pass.

zeth___ 8y ago
Because GNU has always worked that way and given us some of the best software ever written.
- geofft 8y ago
  Worth noting that basically every Linux distro for many years used the eglibc fork, until the former glibc maintainer stepped down, and it's hardly the only GNU project that is what it is today because it managed to reabsorb a fork by admitting a mistake. GCC comes to mind. GNU is what it is today largely because of the name recognition, and people's willingness to get stuff incorporated into GNU because of that name recognition. eglibc was very conservative about patches because it aimed for reunification eventually, but it didn't have to be.
t3soro 8y ago
> some people want an egalitarian project rather than an authoritarian one Plenty of us prefer the authoritarian model, rather than the chaotic model of arbitrary consensus.
- MBCook 8y ago
  Which is totally fair. But that’s not how this project is set up. And that’s what makes this so odd.

sneak 8y ago

Forking out any vestigal rms authority would benefit everyone. They’re bikeshedding over whether or not to include a comment on a controversial topic. It is not relevant to the functioning of the library and is just a distraction. Rip it out and ban rms for perpetuating such idiotic drama.

danmg 8y ago
RMS isn't the one bikeshedding it.
- olliej 8y ago
  he's the one forcing it -- he's not an active member of libc anymore, he's not a maintainer, but he decided that his personal joke from 20 years ago absolutely must remain in the documentation because he's petty. He sounds like the one in need of a safe space where nothing ever changes
  - danmg 8y ago
    It's more like he's preventing bikeshedding by stopping them from making irrelevant non-technical decisions.
    - josefx 8y ago
      There was already a consens before he decided to contribute his own opinion to a triviality. By doing so he prolonged the discussion at least a bit. Forcing your opinion about a pointless and otherwise done issue on others seems like textbook bikeshedding to me. At leaast now he can honestly say he is still actively involved with glibc.

justizin 8y ago

> Consensus is a Bay Area idea [citation needed] Merriam-Webster says the first known use was 1843, at https://www.merriam-webster.com/dictionary/consensus https://www.merriam-webster.com/dictionary/consensus. An SCM change, by very nature, doesn't "erase" anything, it only creates a record of change. The repository will always have the history.

readme 8y ago

Excellent idea! Just rename it political-libc for all the people who want politics mixed in with their C library.

favorited 8y ago
You realize it's the current version of glibc which has the political joke? The patch that RMS is opposing here removed a political reference from the docs.
johannes1234321 8y ago
The FSF and its GNU project are a political endeavor. For a less-political (non-political doesn't exist) libc one might fork into linux-libc or similar. (I'm not saying that's appropriate thing to do, as I'm just a lurker enjoying the free work by others in glibc's regard, but wanted to make the point)
- dragonwriter 8y ago
  > The FSF and its GNU project are a political endeavor. But neither has anything about abortion policy in it's public declaration of goals. So, while some political messages might make sense in GNU technical docs, I don't see why a warning making an oblique reference to the Mexico City policy is not only acceptable but so critical that central leadership needs to overrule the specific project's maintainers to preserve it.
- throwajay87492 8y ago
  These constant attempts to assert that "everything is political" are getting really old. All people have the capacity to be political. Nothing else is. I wish I could understand who taught you people that insistence of stupid horseshit is the key to achieving your goals.
  - johannes1234321 8y ago
    First of all I didn't say that "everything is political" but that in each organisation there is a degree of politics. Maybe the issue is in the word "politics" - this is not about left vs right, but social collaboration. Some people like each other more than others, some are more trustworthy to others and often the way I propose a change and whom I talk to has impact of the change being accepted - those are politics. Also technical decisions have an impact on different people and in subtle ways impact users (i.e. if a kernel switches to a high-res framebuffer mode to print boot messages and developers assume they have the width for formatting messages this impacts visually impaired people, willingly or not -- just made the example up to show subtle things) Politics aren't bad and are an effect of different people being present in the same space, while of course there is a difference between FSF/GNU/RMS with a strong focus on a political mission and some other projects primarily focussing on a technical mission.

geofft 8y ago

"Hacker culture" is a myth - not in the sense that it's false, but in the sense that it's a set of stories that people like because of their storytelling power, and its exact historical truth is not particularly relevant. As a myth, it tends towards better remembering things that make better stories, like great men inventing things through sheer force of will. And as a myth, it's a story about our origins, which were no doubt messier than we would like to remember. That our present does not seem as awesome as the myth is an unavoidable effect of myths. But there a few historically true things that are relevant here: first, that an important part of "hacker culture" unequivocally originated in the Bay Area (Fairchild Semiconductor, Homebrew Computer Club, BSD, etc.), and second, that "consensus" has been an important idea in this culture for a while - an obvious example being the IETF's "rough consensus and running code".