My (probably flawed) comparison is to the term "fair use". Yes, you can play all sorts of games to make those 2 words mean almost anything you want, but at the end of the day that term is defined by law, not by pedantry.
"open source" has an accepted definition, and it's damaging to society to try to undermine it.
If you think I'm exaggerating, please remember that you can probably thank open-source software for the growth of the Internet, the availability of previously restricted secure encryption and thousands of tools that you probably use to earn a living.
If I would ever release it with restrictions, I would call it "source available" or something like that.
I think as a community, we should also have a strict term for a project with source code, but too restrictive to be called open source.
From my search, "Source Available" was the best terminology, unless someone knows a better one.
EDIT: maybe "Awesome Source" ;-)
Not "open source". That term has a meaning, and it's very important that meaning doesn't get diluted.
Edit: typo
Also, given that the original engine is now free as well doesn't this change actually make the license a bit more GPL like? If you modify the Defold engine you can only give it away or keep it to yourself, basically.
I'm sure no one uses the git Version Control System. Oh wait, that's open source software under the GPL.
Your assertions are easily proven false. Sure, a lot of organizations prefer permissive licenses, but saying no one does the GPL is absurd.
AFAIK there could also be some issues sending software to the AppStore if it includes GPL licensed code, but LGPL licensed binaries should be ok to include (from what I understand). Of course this is from a closed source standpoint, my clients don’t generally like to open source their projects.
Personally GPL sounds like a minefield to me so as a developer I avoid GPL licensed code if possible and much prefer BSD licensed code.
I still feel like the asset flipping issue is a legitimate concern though.
> 9. License Must Not Restrict Other Software
> The license must not place restrictions on other software that is distributed along with the licensed software. For example, the license must not insist that all other programs distributed on the same medium must be open-source software.
When I base software on GPL software, then my software is distributed along with that GPL software. According to this phrase, GPL might not place restrictions on my software. Yet it does.
I think it is very much related. If my linker cannot link GPLed software because of the license of this software, then this software is restricting my linker. As per the definition of opensource.org, GPLed software is not open source.
Again,
> 9. License Must Not Restrict Other Software
GPL obligations apply in the latter case, OSI definition talks about the first case.
Further, as an example, software running on Linux is clearly different than the Linux kernel. Nobody would say that an application and the OS kernel is the same software. OSI doesn't make a distinction between library calls and kernel calls.
https://opensource.org/licenses/GPL-2.0 https://opensource.org/licenses/GPL-3.0
Even AGPL is approved. https://opensource.org/licenses/AGPL-3.0
The GPL requires two things, mainly, if you're going to use something GPL in a closed source package.
1) Attribution. You can't just take and not comment.
2) Modifications should be made available under the same license.
this means that if you're using GPL software and including it as a shared object, then you're fine.
If you need to modify that software to make it work for you (including fixing bugs) then you need to make those elements available.
This is different than insisting that anything that touches GPL code be GPL also.
Some proprietary licenses do have provisions for this, which makes them GPL incompatible.
edit For brevity I ignored dual licensing. If you release your library under both GPL and Apache, then things are of course different.
On top of that, they even use "free and open source", which refers to certain moral freedoms rather than license mechanics.
EDIT: IMHO it's in the best interest of everyone working on such licenses to come up with a new positive term and position that, instead of burning goodwill by trying to co-opt the "open source" label. (E.g. if I remember correctly, in the discussions around cloud software, "fair software" was one label used. Nice and positive word, not stepping on existing communities toes). And in reverse I hope people would be accepting of attempts with such new licenses if they keep the messaging straight.
- The OSI claims to have coined the term in 1988 https://opensource.com/article/18/2/coining-term-open-source....
- Here is a use of the term, 7 times, and as the main object, from 1996 http://www.xent.com/FoRK-archive/fall96/0269.html
- Here is a use of the term, as a proper noun, from 1993 https://groups.google.com/forum/#!msg/comp.os.ms-windows.pro...
On a separate note the OSI's original mission statement was (https://web.archive.org/web/20000408035603/http://opensource...)
> The Open Source Initiative's mission will be to own and defend the Open Source trademark, to manage the www.opensource.org resources, to develop branding programs attractive to software customers and producers, and to advance the cause of open-source software and serve the hacker community in other appropriate way
Which they since abandoned because "there is virtually no chance that the U.S. Patent and Trademark Office would register the mark "open source"; the mark is too descriptive" (and one suspects because the mark was already in use in trade of software, see the xent.com link above) https://opensource.org/pressreleases/certified-open-source.p...
They have no special rights to the term open source. I wish people (including the OSI) would stop pretending otherwise. You can think that it's a good idea to use their definition without relying on falsehoods.
But the fact is, if it was a phrase in current use, you wouldn't just have one or two examples. There would be thousands. Where are all the other mentions of Caldera's "Open Source" software?
The reality is this: https://books.google.com/ngrams/graph?content=%22+open+sourc...
Can you spot where the OSI appeared? The reason they couldn't get their trademark is because they had been so successful in promoting the term.
Though, I am curious now who came out with something around 1980. Might have to dig into that one.
In the body it is used (hyphens not added)
2x: “open-source” not followed by the word code.
3x: “open-source code”.
1x: “open source code”.
Moreover it’s not just a throwaway phrase but a label they are using for their model
Caldera believes an open source code model benefits the industry in many ways.
They continue to be extremely interested in DOS and support our open-source technology direction
Caldera’s OEM and Channel Partners can utilize the open-source code models for DOS and Linux to create
(I love it when I can copy and paste old arguments ;))
> But the fact is, if it was a phrase in current use, you wouldn't just have one or two examples. There would be thousands. Where are all the other mentions of Caldera's "Open Source" software?
It doesn't have to be extremely popular to be in use. The truth of the matter is that there wasn't much "open source" or "source available" software. The other truth of the matter is I have finite patience for trawling through old mailing lists looking for the term open source.
> The reality is this: https://books.google.com/ngrams/graph?content=%22+open+sourc....
As you yourself note there is a bump dating to... 1975... the truth of this is the OSI used a term already in use and made it more popular, not that they invented it.
> Can you spot where the OSI appeared? The reason they couldn't get their trademark is because they had been so successful in promoting the term.
No, it was because the term is merely descriptive. It would also be because the term was already in use if the PTO noticed that (but they often don't - and I haven't tracked down the documents to check if they did).
That's outright nonsense. Just for comparison, Linux had already been around for the better part of a decade by that point, Red Hat had incorporated in 1993, and GNU/FSF might as well have been wrapping up the tour for their fifth album. To say that there wasn't much of the stuff in 1998 is either either ignorance, delusion, or both.
Open source software projects were a bajillion years old by that point, but nobody was calling them that, what with the term not yet having been invented and all. They were (begrudgingly!) "free software" until it was decided that mozilla.org would be a thing.
The phrase open source software has an extremely well-known meaning. Uses like this, which fraudulently claim to be open source software even though they fail to comply with the normal definition as used for over 20 years, is simply an exploitative fraud.
If they want to join the open source community, they would be very welcome. But trying to say that they're part of a community they're not is not okay. It's especially dangerous to the people who are not familiar with the law and might make big decisions for themselves based on falsehoods.
Today, trying to use that phrase in the way they're using it is deceptive.
I understand that you have nothing but loathing and contempt for my approach to software development and you are welcome to promote your philosophy but you will find it impossible to bend simple language to your wishes as long as there are people who see things completely differently.
What words would you like me to use for open source software that you're not going to trample over in 5 years when it becomes convenient for advertising your software?
Option a) Makeup a non-generic not already in use term and trademark it.
Option b) Just say "licensed under <well_known_license>".
What you're required to do to get the exclusivity you're looking for is well established in the rules we are all required to play by known as "the law".
Otherwise, I can say something like this and expect you to understand me: Most people believe that an onlooker underhandedly ignores an omphalos, but they need to remember how non-chalantly the cigar about an espadrille laughs out loud. When a menagé à trois for a labyrinth rejoices, a starlet living with a clodhopper feels nagging remorse. An unsightly impresario is darling. When the surly menagé à trois starts reminiscing about lost glory, a curmudgeonly bubble ceases to exist.
Or I can redefine water to mean any clear liquid, including sulfuric acid, and sell that to you, and claim you should've understood what I meant when you die.
The words "open source" have a plain pre-existing meaning. "Water" does not.
Now to go ahead and make your argument for you - this use of the term open source might be like if I went ahead and sold you (having to force this a bit, since I've failed to find a good example of a noun) insurance against kicking the bucket, and then I tried to define it as insurance against literally kicking buckets. Possibly you can find a better example, but that's the best I could come up with and
- "the bucket" clearly distinguishes the term from being merely descriptive
- Selling insurance against kicking buckets makes no sense, unlike "open source software" where "open" means "anyone can look at it"
- Kicking the bucket is a phrase with a much longer history than OSI defined open source software.
Go find your own words. "Shared source" has been used for literally decades to mean Defold's exact situation - why would you not use that?
Yes you can, some people will hate it some people will like it, just like anything else
Language is useless when it can't be used to create shared understanding.
Ok so for the purpose of "shared understanding", let use my preferred definition of open source then.
This is actually a rather good description of what the OSI did. They took a term with plain meaning that was already in use and tried to define it to mean something else.
Now people are getting mad when other people use the phrase "Donal Trump is a space alien" to literally mean that "Donald Trump is an alien from space".
(I can only assume that people have previously said that Donal Trump is a space alien... there are a lot of strange people in the world after all)
- The OSI claims to have coined the term in 1998 https://opensource.com/article/18/2/coining-term-open-source...
- Here is a use of the term, 7 times, and as the main object, from 1996 http://www.xent.com/FoRK-archive/fall96/0269.html
- Here is a use of the term, as a proper noun, from 1993 https://groups.google.com/forum/#!msg/comp.os.ms-windows.pro...
> "open source" was not in common use at the time that OSI decided to use it.
But I suspect that if you search reddit/hn discussions in that period thoroughly you'll find a few cases... feel free if you want to waste your time.
Oops, that was a typo on my part, the OSI claims to have invented it in 1998 not 1988, updated my original post here too
See the source: https://opensource.com/article/18/2/coining-term-open-source...
Epic, Microsoft, IBM, and hundreds of other companies, big and small, have been careful to avoid this issue specifically because there is an existing definition and it causes confusion.
If this is what happened, then they succeeded.
Absolutely nobody is arguing in good faith that there exists a community of people using the term “open source software” to mean something specific to them, which is different from how the OSI defines the term. This community of people, if it ever existed, certainly does not exist anymore. It’s all corporate shills trying to argue in bad faith that they are allowed to call something “open source”. If it was really a confusing problem, they would call it something else; there are enough terms available. But it’s important to them to be able to fool people that it is OSI Open Source when in fact it isn’t, so they call it “open source”.
And then we have the literal-minded people who claim that word definitions are a question of dictionaries, historical precedent or etymology. These people are helping absolutely no-one, except the aforementioned corporations. Language is defined by current general use, and in current language, “open source software” means the OSI definition.
Probably so
> It’s all corporate shills trying to argue in bad faith that they are allowed to call something “open source”.
I've seen basically 0 evidence of this. As far as I've seen it's primarily people like me arguing that you can't lambast people for using words to mean what they mean (even if the specific combination of words has a now common second meaning), and people who naively use the term without understanding that there's a hoard of rabid programmers who insist that the phrase "open source" has been imbued with special meaning and using it otherwise is to summon the devil.
No corporate shill who has an understanding of the situation would be dumb enough to use the term and get into that fight.
> No corporate shill who has an understanding of the situation would be dumb enough to use the term and get into that fight.
These are not people who made an innocent mistake. The product owner of Defold is commenting in this very thread (without any disclaimers, I might add). They want people to believe that it is really OSI “open source”, when in fact it isn’t.
It certainly didn't help them to use the term open source here, instead of everyone discussing the game engine they're releasing completely for free to the public with no strings attached and source code attached they're discussing the meaning of the term open source and accusing them of fraud.
The engine was released for free with no strings attached - you're free to implement whatever you like using the engine.
The source code was released, with very minor strings attached.
See my comment above for links: https://news.ycombinator.com/item?id=23234029
Clicking on "view page source" is more akin taking a picture of the Eiffel Tower, while "open source" would be if France posted the architectural plans in the open. But just because the plans are out in the open doesn't mean that someone doesn't hold copyright to them.
There's already a word for this, "source available", and by your own words it's a great fit for this scenario.
"Open-source software" is a term, it has its meaning. Diluting it for the sake of marketing does not serve the community well.
When someone makes a blatantly false claim, we have another word for it: "fraud".
A Rose by Any Other Name May smell as sweet, but lying is still not okay. The phrase "open source software" has a meaning. If you just mean source available, say that instead.
Because then it is indeed fraud!
Otherwise it isn't.
Fraud is the intentional deception to secure unfair or unlawful gain, or to deprive a victim of a legal right. The government does NOT try to officially define every phrase in the language, nor does it need to. That's not how it works. The issues are, (1) does there appear to be intentional deception, and (2) Would that deception result in unfair/unlawful gain or the loss of the legal right? Issue 2 is true by definition (open source software as the term is generally used always allows people to use the software for any purpose, including commercial users, and this doesn't provide that right, so by definition the victim is losing a legal right). So we're really only asking the first issue - is there intentional deception? I presume you'd argue that there's no deception. Only a court can decide that for sure, but if someone uses a phrase likely to mislead most people, that's at least getting dangerously close.
So no, a government doesn't have to define the term. But even if you think that governments have to use the term that way... well, governments do define the term "open source software" just like OSI does. Again, I'll focus on the US, but this is by no means limited to the US.
The US Office of Management and Budget (OMB) memo of August 8, 2016 (M-16-21) "Federal Source Code Policy: Achieving Efficiency, Transparency, and Innovation through Reusable and Open Source Software" https://www.whitehouse.gov/sites/whitehouse.gov/files/omb/me... defines the term "open source software" as follows:
"Open Source Software (OSS): Software that can be accessed, used, modified, and shared by anyone. OSS is often distributed under licenses that comply with the definition of “Open Source” provided by the Open Source Initiative (https://opensource.org/osd) and/or that meet the definition of “Free Software” provided by the Free Software Foundation (https://www.gnu.org/philosophy/free-sw.html)" Notice that it can be "used by anyone" and it specifically references the OSI's definition. That memo included a mandate to release a certain amount of code as OSS - which meant the government had to define the term, and yes, they used the normal definition for it.
That OMB memo cites the US Department of Defense (DoD)'s official policy on open source software that was released in 2009: https://dodcio.defense.gov/Portals/0/Documents/FOSS/2009OSS.... - it says in 2.2.b.1.iv "Open source licenses do not restrict who can use the software or the fields of endeavor in which the software can be used." So the US DoD thinks that commercial use is by definition allowed by open source licenses. This wasn't even the first government memo about open source software; they had another one in 2003.
I've been citing executive branch policies, but it's also in US law. The National Defense Authorization Act (NDAA) of 2018 at https://www.congress.gov/bill/115th-congress/house-bill/2810... section 875 requires the DoD to participate more fully in a "pilot program for open source software", and cites the policy and definition of OMB M-16-21 ("Federal Source Code Policy: Achieving Efficiency, Transparency, and Innovation through Reusable and Open Source Software") dated August 8, 2016. Yes, that's a US law, and it by reference defines "open source software" using the OSI definition. That is, commercial use must be allowed.
So yes, the US government DOES have a definition of open source software, and it requires permission for any field of endeavor, just like OSI's definition does. It's been that way for almost 20 years.
Caveat: I am NOT a lawyer. But I've cited my sources, look them up if you have questions.
Please see this statement: https://twitter.com/defold/status/1262744466311360517
What this is about is the potential for naive copies where someone takes the codebase and tries to make essentially the exact same product with no significant changes. That would be a source of confusion and competition and not really serve anyone.
It is great to be enthusiastic about open source, but objecting to people sharing source code is rarely a good idea. I openly challenge anyone to make a functioning, useful fork of this project that would leave enough in place for this clause to apply.
You are talking about game engines and their complexities, which makes it hard to comply. There are compliant full fledged operating systems out there (I am sure you have heard of a few) that are actually "open source". So I am not sure why any company should get away with using the well known term in a misleading way.
Let's talk about the operating system angle. When I went to work at NeXT in 1993 there was a lot of GPL code in the system. By 1999 as Mac OS X approached release there was a frantic effort to remove all GPL code in favor of BSD or Apache licensed alternatives or in some cases nothing at all was favorable. If Mac OS had continued to make use of GPL code and components that could have made a big difference to the whole software development community in a strategic sense. Instead this kind of zealotry resulted in barriers. Not only do you not have the control you imagine, but your ongoing efforts to draw these lines makes everything that much more difficult and unpleasant.
It includes "No Discrimination Against Fields of Endeavor"
"You can not commercialise original or modified (derivative) versions of the Defold editor and/or engine" does not meet (6).
I'm not even going to start on the use of the term "free".
[0] https://opensource.org/osd