Re: Why GNU/Linux is not accepted: an observation
* Akira Urushibata [2019-11-12 15:09]: > On Saturday I attended an open source event in Fukuoka, western > Japan. I visited the booth of an organization named LinuC which > conducts exams and issues certificates to those who pass. > > I had glanced through at one of the textbooks they recommend. It > said that Linux started in 1991. It did not make clear that the Linux > kernel was built upon existing GNU software. > > I pointed out that Linux, strictly speaking, is just the kernel and > different from what is commonly referred to Linux. This remark made > the booth attendant visibly uncomfortable: he started wading and > gasping for air. He was troubled because when things are presented > this way, the shallowness of one's understanding of crucial system > components becomes impossible to disguise. Ironically it is the > knowledge of these very components which lie outside the kernel > that examinations like LinuC measure. > > I had noticed that their textbook mentions neither sed nor awk. It is > hard for me to envision someone claiming UNIX proficiency with no > experience with sed and awk. You could use this information: https://en.wikipedia.org/wiki/History_of_Linux#Linux_under_the_GNU_GPL In the "Notes for linux release 0.01", Torvalds lists the GNU software that is required to run Linux:[18] Sadly, a kernel by itself gets you nowhere. To get a working system you need a shell, compilers, a library etc. These are separate parts and may be under a stricter (or even looser) copyright. Most of the tools used with linux are GNU software and are under the GNU copyleft. These tools aren't in the distribution - ask me (or GNU) for more info. And then you could point out that GNU and GNU Compiler Collection existed long before Linux kernel. Sending corrections to editors is fine. Jean
Re: GNU Kind Communication Guidelines versus "social contract" or Codes of Conduct
Andreas Enge, 06/11/19 11:33: If you start by equating two unrelated concepts, nothing useful can come out of a discussion. I don't know about equating, but there is a clear connection. When you establish some rules with some sanctions, and a process to enforce them, this process needs to be depend on somebody or something. If this something is not held accountable to the same values and methods that the project normally operates on, it's easy to have a conflict. A common example is a self-appointed enforcer attached to an entity accountable only to itself (as many USA foundations are). If you add it on top of a project whose community is a do-ocracy or democracy, based on some values, there is no way to make sure the enforcer respects the community's values. (Forgive the following analogy.) Many software projects have a "constitution" but said constitution has no teeth. If you add a "criminal law" and there is no way to hold the respective executive/judicial power in check, whatever values the "constitution" proclaims are no longer worth the paper they're printed on. The traditional GNU structure at least is coherent because there is a single source of legitimacy. In Debian, as far as I understand, everything is under the project leader, who is however elected. In Wikimedia there's a self-appointed legal entity with its own bureaucracy (Wikimedia Foundation) and a separate community with its own values and processes. These things are not easy to get right. Federico
Re: Why "GNU/Linux" is not accepted: an observation
On Saturday I attended an "open source" event in Fukuoka, western Japan. I visited the booth of an organization named LinuC which conducts exams and issues certificates to those who pass. I had glanced through at one of the textbooks they recommend. It said that Linux started in 1991. It did not make clear that the Linux kernel was built upon existing GNU software. I pointed out that "Linux", strictly speaking, is just the kernel and different from what is commonly referred to "Linux." This remark made the booth attendant visibly uncomfortable: he started wading and gasping for air. He was troubled because when things are presented this way, the shallowness of one's understanding of crucial system components becomes impossible to disguise. Ironically it is the knowledge of these very components which lie outside the kernel that examinations like LinuC measure. I had noticed that their textbook mentions neither sed nor awk. It is hard for me to envision someone claiming UNIX proficiency with no experience with sed and awk. Opening the Software Toolbox by Arnold Robbins https://www.gnu.org/software/coreutils/manual/html_node/Opening-the-software-toolbox.html How well one understands the above document, I believe, is a good measure of UNIX proficiency. Awk is not discussed in depth here, but if you wish to solve real-life problems with this approach, you need it. One who adheres to the "It all started with one email in 1991" story lacks a firm grasp of the operating system. And people with this level of understanding are making decisions on what is important and what is not in "Linux." I understand that Richard Stallman wants students to learn Lisp using GNU and Emacs as the working environment. On the other hand he is along with Arnold Robbins an original author of GNU Awk.
