4 ms·

read up on free software movement vs open source thing. you are only looking at "open source" when the argument being that the software should be free for users

read up on free software movement vs open source thing. you are only looking at "open source" when the argument being that the software should be free for users, not the developers in the middle. when i write AGPL code, i expect "end users" to be given source of that code in whatever circumstance they might get to use the software. now here comes a developer who uses my code, puts it in a private fork in your words and denies his/her users access to source code, what happened to my giving that dev license to use the code? what you are referring to us those MIT/BSD license where the developer is expected to heartlessly work to build software which can be forked by anyone, used in a proprietary code and users cant do anything about it. you are talking about freedom of the "developer" while free software people talk about freedom of end users. obviously "fake open source" is there because something like SSPL is not a "open source license" but it is in essense the superset of AGPL and that is what Free software guys want. you use this code, the users MUST get source, the dev has no say in that

cassonmars 5y ago
I’m not really buying the notion that SSPL isn’t open source, and the more I read the discourse around free software and the intentions thereof (_user_ freedom), I fail to see how it doesn’t count. I know a lot of folks love to quip that the OSI won’t accept SSPL, but they don’t get to control the use of the term since it was never theirs to begin with, and the source is absolutely open and available. The stipulation is that you can’t pull an Amazon where you get to play “I’m not touching you” games with the software by wrapping the service with value adds via proxy to escape having to contribute back. That feels in line with the intentions that classical distributed software like GPL intended, for that the emergence of SaaS enabled barn-door-wide loopholes.
- nix23 5y ago
  You don't understand the difference of Open Source and Free Software
  - ttybird 5y ago
    I believe that they do. This attack is unwaranted and without substance or explanation.
  - cassonmars 5y ago
    The way I understand it is that the spirit of Free Software is _user_ freedom - they are free to use the software in any way they wish without restrictions. SSPL meets this bar no differently than any other license with provisions requiring open sourcing code. Open Source, per the OSI, has a particular point of view, in one clause they feel SSPL fails the test: their interpreted ambiguity of what triggers related code to also require the same license felt like it was too infectious — i.e. it could not live harmoniously in an ecosystem of deployed software because of their fear that mere interactions could be considered a triggering event. The reality is that is not true. Not only does the license very clearly outline the condition (basically it eliminates the proxy loophole AWS has historically used), the creators of the license also created an accompanying FAQ providing further clarifications of intent. I find the strong emotional reaction people get when someone asks why SSPL doesn’t count is quite odd. I find the fact that literally nobody (besides the OSI in extremely weak argumentation that the license self-clarifies on, the authors repeatedly clarified on, and the accompanying FAQ removes all doubt) can provide a coherent argument in opposition even more strange. If I want to provide an equally vitriolic remark, I think the reality of why OSI rejected it, and why some people feel so strongly opposed, is ironically the same suspicions they casted on the genesis of the license (MongoDB has a business interest in this license): look at the members of the OSI — there are a number of IaaS people, who of course would be opposed to recognizing a license that stops their grift.
jillesvangurp 5y ago
As an occasional OSS contributor and maintainer of a few minor Github projects; being able to give my users and contributors the freedom to do as they want rather than as I want is precisely why I prefer MIT style permissive licenses. I'm under no pressure by anyone to give them that freedom. If somebody finds a use for code that I wrote, more power to them! Exactly what I want. And yes, I put some thought into this and definitely read up on this. For me the AGPL is actually the least free OSS license. I avoid depending on any software that uses it because I don't want to have to consult a lawyer when dealing with dependencies. I mistrust companies wielding it to be perceived as open when in reality all they want is for you to buy their software under a different license. In a way what Elastic does is more honest. They are simply saying, this software is ours and this is what you are allowed to do with it (i.e. it's a permissive license). But it's straightforward to understand and it seems many of their non paying users are OK to continue to use their software under that license. As a very infrequent contributor, I of course regret the license change though. IMHO the true value in OSS communities is joint copyright ownership of the developers working on it. If there are enough copyright holders, it precludes licensing changes. That combined with a permissive enough license lets people innovate freely. Anything with widely enough distributed ownership is almost impossible to restrict the usage off commercially as long as the license is permissive enough. The lack of distributed ownership is how Elastic was able to re-license and it is why Linux is such a success. Unintentionally (Richard Stallman definitely did not like the form this success took), GPLv2 provided enough freedom through some loopholes in the license that companies could use it to build proprietary products. Products like Android, which are mostly closed source and worth quite a lot. Attempts to prevent further commercial success in the form of GPLv3 and the AGPL never really affected Linux because by the time those came into existence, re-licensing was already a mission impossible even if Linus Torvalds would have wanted to; which as I understand it he was never really that much of a proponent of to begin with.

cassonmars 5y ago

I’m not really buying the notion that SSPL isn’t open source, and the more I read the discourse around free software and the intentions thereof (_user_ freedom), I fail to see how it doesn’t count. I know a lot of folks love to quip that the OSI won’t accept SSPL, but they don’t get to control the use of the term since it was never theirs to begin with, and the source is absolutely open and available. The stipulation is that you can’t pull an Amazon where you get to play “I’m not touching you” games with the software by wrapping the service with value adds via proxy to escape having to contribute back. That feels in line with the intentions that classical distributed software like GPL intended, for that the emergence of SaaS enabled barn-door-wide loopholes.

nix23 5y ago
You don't understand the difference of Open Source and Free Software
- ttybird 5y ago
  I believe that they do. This attack is unwaranted and without substance or explanation.
- cassonmars 5y ago
  The way I understand it is that the spirit of Free Software is _user_ freedom - they are free to use the software in any way they wish without restrictions. SSPL meets this bar no differently than any other license with provisions requiring open sourcing code. Open Source, per the OSI, has a particular point of view, in one clause they feel SSPL fails the test: their interpreted ambiguity of what triggers related code to also require the same license felt like it was too infectious — i.e. it could not live harmoniously in an ecosystem of deployed software because of their fear that mere interactions could be considered a triggering event. The reality is that is not true. Not only does the license very clearly outline the condition (basically it eliminates the proxy loophole AWS has historically used), the creators of the license also created an accompanying FAQ providing further clarifications of intent. I find the strong emotional reaction people get when someone asks why SSPL doesn’t count is quite odd. I find the fact that literally nobody (besides the OSI in extremely weak argumentation that the license self-clarifies on, the authors repeatedly clarified on, and the accompanying FAQ removes all doubt) can provide a coherent argument in opposition even more strange. If I want to provide an equally vitriolic remark, I think the reality of why OSI rejected it, and why some people feel so strongly opposed, is ironically the same suspicions they casted on the genesis of the license (MongoDB has a business interest in this license): look at the members of the OSI — there are a number of IaaS people, who of course would be opposed to recognizing a license that stops their grift.

jillesvangurp 5y ago

As an occasional OSS contributor and maintainer of a few minor Github projects; being able to give my users and contributors the freedom to do as they want rather than as I want is precisely why I prefer MIT style permissive licenses. I'm under no pressure by anyone to give them that freedom. If somebody finds a use for code that I wrote, more power to them! Exactly what I want. And yes, I put some thought into this and definitely read up on this. For me the AGPL is actually the least free OSS license. I avoid depending on any software that uses it because I don't want to have to consult a lawyer when dealing with dependencies. I mistrust companies wielding it to be perceived as open when in reality all they want is for you to buy their software under a different license. In a way what Elastic does is more honest. They are simply saying, this software is ours and this is what you are allowed to do with it (i.e. it's a permissive license). But it's straightforward to understand and it seems many of their non paying users are OK to continue to use their software under that license. As a very infrequent contributor, I of course regret the license change though. IMHO the true value in OSS communities is joint copyright ownership of the developers working on it. If there are enough copyright holders, it precludes licensing changes. That combined with a permissive enough license lets people innovate freely. Anything with widely enough distributed ownership is almost impossible to restrict the usage off commercially as long as the license is permissive enough. The lack of distributed ownership is how Elastic was able to re-license and it is why Linux is such a success. Unintentionally (Richard Stallman definitely did not like the form this success took), GPLv2 provided enough freedom through some loopholes in the license that companies could use it to build proprietary products. Products like Android, which are mostly closed source and worth quite a lot. Attempts to prevent further commercial success in the form of GPLv3 and the AGPL never really affected Linux because by the time those came into existence, re-licensing was already a mission impossible even if Linus Torvalds would have wanted to; which as I understand it he was never really that much of a proponent of to begin with.