I agree with parts about entitlement. The expectation of support and treatment of open source software as if it was proprietary is a real problem.
But, the authour makes a similar mistake - they conflate open source software with source-available (proprietary) software. As an example, I strongly disagree with this part:
When software is open-source, it is open-source, not necessarily free and open-source (FOSS), and even if it is FOSS, it might still have a restrictive licence. The code being available in and of itself does not give you a right to take it, modify it, or redistribute it.
If you replace it with this version, I am happy:
When software is source-available, it is source-available, not necessarily open source or free and open-source (FOSS). The code being distributed under a source available license does not give you a right to take it, modify it, or redistribute it.
I think it’s really important that we keep a clear delineation between free/open source software on one side, and source-available (proprietary software) on the other.
A lot of companies are trying to co-opt and blur the meaning of the term so they can say “seeing the source was always the point, none of the other freedoms mattered”, in order to sell you proprietary licenses.
Open source gives you the right to take, modify and redistribute it. Source available does not. And that’s ok, just please don’t blur the terms together.
even if it is FOSS, it might still have a restrictive license
Likewise, this is definitionally untrue. The whole purpose of FOSS is to give you the four freedoms.
Maybe it sounds a bit like a conspiracy theory but with how often people make this “mistake” i really believe it’s a deliberate effort to undermine the meaning of open-source
I once had an argument in another community on here about something very similar. And they told me I was wrong. The mods deleted my posts.
I posted the links and the definition/requirements for FOSS as compared to just open source.
They kept telling me i was talking about open source and not libre.
The links and definitions and requirements I posted:
From Richard Stallman, from the site whose creators developed the rules and requirements for FOSS, GNU.org, and from the itsFOSS site which, indeed, references and links to the first 2.
The definitions also explicitly state the difference and uniqueness of each and compares them to the nonstandard open source (source available) labels.
I unjoined that community and found a less ignorant one.
The mods are definitely wrong, and they shouldn’t be deleting posts here. But you could also be wrong, I don’t have the original posts to go off of, but I do have this one.
FOSS is not the same as Free Software, it’s a combination of Free Software and Open Source software, meaning it applies to both. In long form, it’s Free and Open Source Software, meaning it applies to things applying to one or the other, and not necessarily both.
If you mean Free Software (i.e. the FSF/GNU definition), then use that term. If you mean Open Source (i.e. the OSI definition), then use that term. If you’re not sure which you mean, but you know you mean one of the two, use the term FOSS. If you just mean the source is available but it doesn’t necessarily fit the the Free Software or Open Source definitions, use the term “source available” and leave it at that.
Most FOSS licenses are both Free and Open Source (i.e. they meet the definition of both), but not all. Many Open Source licenses are incompatible w/ Free Software licenses, for example the Apache 2.0 license is incompatible with the (L)GPL < v3 in some cases.
In general:
FOSS - preferred when the software is either free software or open source software
source-available - preferred when the software is not FOSS, but you do have access
proprietary - use for either source-available or non-source-available software
Free Software - use when referring to copy-left software (yes, it applies to more, but let’s keep things simple)
Yes, just because the source code is available doesn’t mean it is licensed for others to take. Now we have AI tools that have scraped the web for all of its content and won’t see the difference between source available, open source , and free and open source. It is possible that those who use AI tools could be unknowingly using code without the license to do so.
The code being distributed under a source available license does not give you a right to take it, modify it, or redistribute it.
I would add in “automatically” there. Source-available licenses could give you the right to modify it and redistribute it, or it may not. For example, the Unreal Engine is proprietary and covered by a source-available model, and you can redistribute it to other license holders (must accept certain terms and conditions), and even then w/ restrictions. So you can take it, modify it, and redistribute it, but there are a handful of very important asterists there.
Basically, if you don’t recognize the license as one of the major ones (Apache, BSD, MIT, (L|A)GPL, etc), then treat it as source-available w/ no rights other than reading it until you actually read and understand the license.
even if it is FOSS, it might still have a restrictive license
Likewise, this is definitionally untrue. The whole purpose of FOSS is to give you the four freedoms.
That depends on your definition of “restriction.”
If you use a copyleft license, you could be restricted from using proprietary plugins, since you’d have to release the source of those plugins if a user accesses it in a covered way, and you don’t have the right to do that. That sounds like a restriction to me, and it doesn’t apply to more permissive licenses like BSD or MIT.
The intent of copyleft is to ensure freedoms for the recipients of derivatives of your works. In software that means the users of forks. Copyleft restricts you to the same license (or a compatible one) to prevent you adding more restrictions. ““More permissive”” software licenses can be redistributed with the same license but often it’s a more restrictive license (e.g. MIT -> proprietary).
You can’t remove the license, say MIT, or add any restrictions to the code. You can combine it with other works and distribute the combined work under different terms, provided you still abide by the terms of the license (e.g. source distributions need to retain the license, binary distributions don’t).
Copyleft forces modified distributions to have the same terms as the copylefted software. If you have some MIT code and some GPL code and you combine it, the combined work is GPL. Likewise if you combine proprietary code and GPL, the combined work is GPL, and that’d regardless of how it’s distributed.
Permissive licenses protect the source as it was when it was combined into the other work. Copyleft licenses enforce distributed changes to the source are made available under the same license. Which one you want depends on what “freedoms” you want. Do you want the freedom to use the source however you want, including with proprietary code? Then you want a permissive license. Do you want to ensure that any changes to the code are always available? Then you want copyleft code.
With the minimal amount of work added the combined work can now have added restrictions. They’re pushover licenses.
Devs are free to choose whatever license they want but in the pathfinding problem of interacting with others then “protecting the source” is the wrong target node. Copyleft is a tool to help people.
Let’s use an example of the MIT license, perhaps the most permissive (license text below).
This license only applies to the source code, so it allows you to distribute executables under any terms you like. So the combined work here is not under the terms of the MIT license, but whatever terms the distributor wants.
If the combined work is distributed in source form, then any modifications you make are under the MIT license unless you make it explicit which parts are licensed differently. If users don’t like the license terms for your changes, they can remove them and be subject to only the terms of the code they keep. You cannot change the terms of existing code.
Since the MIT license doesn’t obligate users to release their modifications or combined work, it’s more attractive to businesses, and businesses frequently upstream their changes to reduce their own maintenance burden. A lot of changes are not upstreamed though.
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Choice of license depends strongly on how you want others to use it. If you want it to appeal to businesses as a library (and attract fixes from their full-time devs), a permissive license is a good bet. If you want it to be a community developed thing or perhaps an application, copyleft can make a lot of sense. I’ve like all FOSS licenses and I’ve released projects under everything from the MIT license to the AGPL.
My preferred license barring any other considerations is the MPL (copyleft at the file level), but there are usually other considerations.
My only ask is that people use a standard license instead of DIYing one (or not using one). You can also change your license at any time, provided you’re the only contributor or every other contributor agrees (otherwise you’d need to selectively remove their contributions).
I agree with parts about entitlement. The expectation of support and treatment of open source software as if it was proprietary is a real problem.
But, the authour makes a similar mistake - they conflate open source software with source-available (proprietary) software. As an example, I strongly disagree with this part:
If you replace it with this version, I am happy:
I think it’s really important that we keep a clear delineation between free/open source software on one side, and source-available (proprietary software) on the other.
A lot of companies are trying to co-opt and blur the meaning of the term so they can say “seeing the source was always the point, none of the other freedoms mattered”, in order to sell you proprietary licenses.
Open source gives you the right to take, modify and redistribute it. Source available does not. And that’s ok, just please don’t blur the terms together.
Likewise, this is definitionally untrue. The whole purpose of FOSS is to give you the four freedoms.
Maybe it sounds a bit like a conspiracy theory but with how often people make this “mistake” i really believe it’s a deliberate effort to undermine the meaning of open-source
I think many people fail to wrap their head around the concept since almost everywhere else nothing is really free.
Yeah, freedom has become vanishingly rare
I once had an argument in another community on here about something very similar. And they told me I was wrong. The mods deleted my posts.
I posted the links and the definition/requirements for FOSS as compared to just open source.
They kept telling me i was talking about open source and not libre.
The links and definitions and requirements I posted:
From Richard Stallman, from the site whose creators developed the rules and requirements for FOSS, GNU.org, and from the itsFOSS site which, indeed, references and links to the first 2.
The definitions also explicitly state the difference and uniqueness of each and compares them to the nonstandard open source (source available) labels.
I unjoined that community and found a less ignorant one.
The mods are definitely wrong, and they shouldn’t be deleting posts here. But you could also be wrong, I don’t have the original posts to go off of, but I do have this one.
FOSS is not the same as Free Software, it’s a combination of Free Software and Open Source software, meaning it applies to both. In long form, it’s Free and Open Source Software, meaning it applies to things applying to one or the other, and not necessarily both.
If you mean Free Software (i.e. the FSF/GNU definition), then use that term. If you mean Open Source (i.e. the OSI definition), then use that term. If you’re not sure which you mean, but you know you mean one of the two, use the term FOSS. If you just mean the source is available but it doesn’t necessarily fit the the Free Software or Open Source definitions, use the term “source available” and leave it at that.
Most FOSS licenses are both Free and Open Source (i.e. they meet the definition of both), but not all. Many Open Source licenses are incompatible w/ Free Software licenses, for example the Apache 2.0 license is incompatible with the (L)GPL < v3 in some cases.
In general:
Go to gnu.org
I linked GNU’s definition of Free Software already.
Yes, just because the source code is available doesn’t mean it is licensed for others to take. Now we have AI tools that have scraped the web for all of its content and won’t see the difference between source available, open source , and free and open source. It is possible that those who use AI tools could be unknowingly using code without the license to do so.
I would add in “automatically” there. Source-available licenses could give you the right to modify it and redistribute it, or it may not. For example, the Unreal Engine is proprietary and covered by a source-available model, and you can redistribute it to other license holders (must accept certain terms and conditions), and even then w/ restrictions. So you can take it, modify it, and redistribute it, but there are a handful of very important asterists there.
Basically, if you don’t recognize the license as one of the major ones (Apache, BSD, MIT, (L|A)GPL, etc), then treat it as source-available w/ no rights other than reading it until you actually read and understand the license.
That depends on your definition of “restriction.”
If you use a copyleft license, you could be restricted from using proprietary plugins, since you’d have to release the source of those plugins if a user accesses it in a covered way, and you don’t have the right to do that. That sounds like a restriction to me, and it doesn’t apply to more permissive licenses like BSD or MIT.
The intent of copyleft is to ensure freedoms for the recipients of derivatives of your works. In software that means the users of forks. Copyleft restricts you to the same license (or a compatible one) to prevent you adding more restrictions. ““More permissive”” software licenses can be redistributed with the same license but often it’s a more restrictive license (e.g. MIT -> proprietary).
You can’t remove the license, say MIT, or add any restrictions to the code. You can combine it with other works and distribute the combined work under different terms, provided you still abide by the terms of the license (e.g. source distributions need to retain the license, binary distributions don’t).
Copyleft forces modified distributions to have the same terms as the copylefted software. If you have some MIT code and some GPL code and you combine it, the combined work is GPL. Likewise if you combine proprietary code and GPL, the combined work is GPL, and that’d regardless of how it’s distributed.
Permissive licenses protect the source as it was when it was combined into the other work. Copyleft licenses enforce distributed changes to the source are made available under the same license. Which one you want depends on what “freedoms” you want. Do you want the freedom to use the source however you want, including with proprietary code? Then you want a permissive license. Do you want to ensure that any changes to the code are always available? Then you want copyleft code.
With the minimal amount of work added the combined work can now have added restrictions. They’re pushover licenses.
Devs are free to choose whatever license they want but in the pathfinding problem of interacting with others then “protecting the source” is the wrong target node. Copyleft is a tool to help people.
Your first paragraph is a bit misleading.
Details
Let’s use an example of the MIT license, perhaps the most permissive (license text below).
This license only applies to the source code, so it allows you to distribute executables under any terms you like. So the combined work here is not under the terms of the MIT license, but whatever terms the distributor wants.
If the combined work is distributed in source form, then any modifications you make are under the MIT license unless you make it explicit which parts are licensed differently. If users don’t like the license terms for your changes, they can remove them and be subject to only the terms of the code they keep. You cannot change the terms of existing code.
Since the MIT license doesn’t obligate users to release their modifications or combined work, it’s more attractive to businesses, and businesses frequently upstream their changes to reduce their own maintenance burden. A lot of changes are not upstreamed though.
MIT license text
Copyright © 2025 <copyright holders>
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Choice of license depends strongly on how you want others to use it. If you want it to appeal to businesses as a library (and attract fixes from their full-time devs), a permissive license is a good bet. If you want it to be a community developed thing or perhaps an application, copyleft can make a lot of sense. I’ve like all FOSS licenses and I’ve released projects under everything from the MIT license to the AGPL.
My preferred license barring any other considerations is the MPL (copyleft at the file level), but there are usually other considerations.
My only ask is that people use a standard license instead of DIYing one (or not using one). You can also change your license at any time, provided you’re the only contributor or every other contributor agrees (otherwise you’d need to selectively remove their contributions).