Comment by derefr | Hacker News Reader

Comment by derefr | Hacker News Reader

Parent Full thread

I think you misread something? What you just gave is the definition of the term “source available.”

I’ll restate. My descriptivist definition of the term “open-source”, was that something is “open-source” if you are free to do at least one of either forking the thing for your own use, or modifying the thing and pushing those modifications upstream. In other words, “open-source” software is software where you’re free to notice a flaw in the program and, by some mechanism, get it fixed in such a way that other people can benefit from the fix.

And yes, the OSI definition of “Open Source” is an organization enforcing an arbitrary definition, because it disagrees with many uses of the term “open source” — specifically, the OSI considers GPL-licensed software to not be capital-letters “Open Source”, despite everyone conventionally considering it regular-adjective “open-source”. (The FSF does a similar brain-damaged thing with the term “Free Software”, which has not-much to do with what people would call “free software” — but at least “free software” isn’t a very common term in modern parlance; people aren’t constantly talking about “free software” with the FSF then misconstruing their words as if they had said “Free Software.” Which is exactly what the OSI does when people talk about “open-source software”, and then the OSI misconstrues their statements as statements about “Open Source software.”)

Yes, this is exactly why pedants use the term “FOSS” — i.e. “Free and Open Source Software”, a union-ing of the set of software that is FSF-jargon “Free Software” with the set of software that is OSI-jargon “Open Source Software.”

Regular people, who aren’t trying to be sycophants toward these two organizations, on the other hand, just use the uncapitalized adjective open-source to refer to this larger set.

> specifically, the OSI considers GPL-licensed software to not be capital-letters “Open Source”

What? OSI disagrees with you.

Yes, here's a list of licenses approved by the OSI as open source, which includes the GPL and AGPL:

https://opensource.org/licenses/alphabetical

"Free software", "open source software", and "free and open source software" are often used interchangeably because they refer to software that is largely protected by the same set of licenses. The terms differ by emphasizing different aspects of the philosophy behind FOSS.[1]

[1] https://www.gnu.org/philosophy/open-source-misses-the-point....

> My descriptivist definition of the term “open-source”, was that something is “open-source” if you are free to do at least one of either forking the thing for your own use, or modifying the thing and pushing those modifications upstream.

Respectfully, this definition of "open-source" does not match common use, even with a loose colloquial interpretation in mind. As commonly understood, an "open-source" license requires the ability to acquire the source, and to modify it, and to redistribute it.

Your definition, quoted above, would call open-source a license that allowed for personal modifications to source but forbade redistribution of those modifications. Your definition would also call open-source a license that required all modifications of the source to be made through a single point of central control. I think both of these cases fail even loose understandings of what "Open Source/open-source" is. I do think they both describe "source available".

You draw a distinction between "open-source" as a colloquial term and "Open Source" as defined by the OSI, but I don't think this isn't very compelling or useful. Sure, language evolves and people use terms incorrectly all the time, but that doesn't make erroneous use correct. In particular, "Open Source" is used as a legal criteria for a set of software licenses, so precision in its definition is important. Also, I have not personally come across any cases where there was any material difference between colloquial use of the term open-source and precise definition of Open Source as defined by the OSI. Do you have anything to cite that justifies a distinction here?

So I beg you to reconsider the idea that the OSI is "some organization who just decided to name themselves after the term" and the idea that "Open Source" is a term "defined loosely by how people use it." In point of historical fact, the term was first used by the people behind the organization, who formed it to popularize it as a new concept, and an alternative to "Free Software": https://web.archive.org/web/20021001164015/http://www.openso.... The OSI is not some organization formed to co-opt and redefine the term for their own purposes after it was already in common use.

I do agree with you that often times people think of specific project management structures when talking about open-source development. I.e. the debates about what makes for a successful open-source project, cultivating a community, making it easy for community members to contribute, etc. There can be non-open-source projects using open-source development methodologies, sure, but it doesn't make them "open-source" even in a colloquial sense. The sole criteria is always the license granted to users of the software.

Reply on news.ycombinator.com