How the GDPR apply to credits?

[Peter]

Does anyone know how GDPR applies to game credits?

Before releasing a game I might want to let a few people test it to get some feedback and then I might want to include them in the "credits" that comes with the game (if they want) but my concern is that the names could be regarded as "personal data" and that I cannot distribute them freely.

If someone later objected to me using their names in the credits I would still want to keep the older versions downloadable. With GPL I think it's even a requirement that I offer the source code for at least three years. And since it's a free open-source game I will not have full control over how the names will be spread and displayed by others.

The GDPR talks about consent as a legal basis but as far as I have understood consent is always revocable (and should be so easily).

[dulsi]

I don't think it applies. If I write a news story about something you did, you can't use GDPR to remove it. (Or at least I don't think you can.) Assuming they agree to have you add them to the credits, it's not personal data. I would still honor requests to remove them later but I don't think you have any legal requirement. I wouldn't purge all historical copies with the name.

[drummyfish]

TFW people are already scared to write names on the Internet, think this society would benefit from a collapse right now. Also funny how this society's MUST GIVE CREDIT collides with the MUST NOT REVEAL IDENTITY urge, what a catch 22 The main thing nowadays is to make sure there is no face of a child anywhere in your work and that your credits have appropriate number of minorities, perhaps that can balance out the otherwise unavoidable sins of living and breathing in this greatly advanced civilization. A final word of advice from someone who's been developing software for years: next time make sure you don't collaborate with anyone so you don't get into trouble, that's what worked best for me to avoid this kind of trouble.

</rant>

[bzt]

Before releasing a game I might want to let a few people test it to get some feedback and then I might want to include them in the "credits" that comes with the game (if they want) but my concern is that the names could be regarded as "personal data" and that I cannot distribute them freely.

No free license expects you to use actual real names. They only say "reasonable attribution", and not "use real names".

To cut the Gordian knot, simply ask your testers what names (or pseudonames, acronyms, nicks, whatever) they would like to be used on your credits list. Simple as that.

With GPL I think it's even a requirement that I offer the source code for at least three years.

With GPL there's no time limit, you must provide the source code unlimited. (GPL says if you distribute your source code only on a physical medium (floppy or CD), then you can't change your offer for three years, not that you can stop offering the source code after three years, eg. you're obliged to make a new offer every three years.)

And since it's a free open-source game I will not have full control over how the names will be spread and displayed by others.

That should not concern you, anyone receiving a copy of your program must follow exactly the same licensing terms and use conditions as you. GPL section 7b: "Requiring preservation of specified reasonable legal notices or author attributions"

TFW people are already scared to write names on the Internet, think this society would benefit from a collapse right now.

Don't you worry, the collapse is already happening in front of your eyes, and it is unstoppable as things stand.

Also funny how this society's MUST GIVE CREDIT collides with the MUST NOT REVEAL IDENTITY urge

Nope, no collision. Let's say I create a software using one of your libraries, and I mention you as "drummyfish". I have credited you, yet I haven't revealed your identity. Works perfectly.

Cheers,
bzt

[Peter]

Thanks for all the replies.

If I write a news story about something you did, you can't use GDPR to remove it. (Or at least I don't think you can.)

I don't know. There seems to be special laws for news media that doesn't apply to the rest of us.

I would still honor requests to remove them later but I don't think you have any legal requirement.

For the latest and future versions, no problem.

I wouldn't purge all historical copies with the name.

Yeah, I have considered the possibility of going back and removing a name from earlier versions, technically I could do that, but it feels wrong to go back and make changes to already released versions like that (the hashes would be different and people would wonder why changes were sneaked in afterwards) and not to mention the extra work. It would still not remove it from copies that might be downloaded elsewhere. This is not something I would like to end up having to do.

Also funny how this society's MUST GIVE CREDIT collides with the MUST NOT REVEAL IDENTITY urge, what a catch 22

In this case there is no one that forces me to give credits. It's mostly just to have something to offer my testers and to make them take it more serious. I don't think I will be able to pay them.

No free license expects you to use actual real names. They only say "reasonable attribution", and not "use real names".

This is not really about licenses. The testers will not be copyright holders.

For copyright notices I would have thought that it was important to use real names to be able to identify who owns the copyright, for legal purposes, but even real names doesn't do that perfectly because there can often be many people with the same name (I once lived on a long street that had two other people with the same first and last name as me) so I wouldn't be surprised if you were right.

To cut the Gordian knot, simply ask your testers what names (or pseudonames, acronyms, nicks, whatever) they would like to be used on your credits list. Simple as that.

It's interesting with a different opinion. What you say is essentially what I had planned to do except that I was thinking about ways to enforce real or at least real-looking names by perhaps rejecting them as testers if they choose a weird looking name. At the end of films you don't see a lot of made up names, do you? Why would a game be different?

I will have to think more about this but it still doesn't avoid the concern that I had unless I force people to use fake names (which I don't consider an option). If people do decide to use their real names I'm still left with the original question about what to do if they later want it removed. I wonder how big movies do in the same situation.

With GPL there's no time limit, you must provide the source code unlimited. (GPL says if you distribute your source code only on a physical medium (floppy or CD), then you can't change your offer for three years, not that you can stop offering the source code after three years, eg. you're obliged to make a new offer every three years.)

Ah, you're right. The thing about three years is only relevant when releasing "object code" (e.g. an executable) on a physical medium such as a CD.

I doubt you have to keep offering the source code forever if you stop distributing the "object code", that would be unreasonable, but that's a different discussion...

And since it's a free open-source game I will not have full control over how the names will be spread and displayed by others.

That should not concern you, anyone receiving a copy of your program must follow exactly the same licensing terms and use conditions as you. GPL section 7b: "Requiring preservation of specified reasonable legal notices or author attributions"

The problem here is the opposite. I don't want to prevent anyone from removing the credits. What I don't want is that I should in some way be obliged to demand that they remove them if the persons in the credits come to me saying they have changed their mind.

[bzt]

What you say is essentially what I had planned to do except that I was thinking about ways to enforce real or at least real-looking names

I'm not sure you have the authority to do that.

by perhaps rejecting them as testers if they choose a weird looking name.

That you can do.

At the end of films you don't see a lot of made up names, do you?

Actually, you do, all the time. Almost every name in every movie's end credits is made up.

For example, does William Pitt, Robyn Fenty, Maurice Micklewhite, Kathleen Ruston, Norma Mortenson, Neta-Lee Hershlag ring a bell? What if I say their screen names, Brad Pitt, Rihanna, Michael Caine, Audrey Hepburn, Marilyn Monroe and Natalie Portman? And not just movies, same everywhere: Freddy Mercury, Bob Dylan, Lady Gaga, Sting...

I will have to think more about this but it still doesn't avoid the concern that I had unless I force people to use fake names (which I don't consider an option).

As I have said, you don't have the authority to do that. You can reject a tester, but you can't tell 'em what nickname to choose, and you can't force them to use their real names either.

If people do decide to use their real names I'm still left with the original question about what to do if they later want it removed.

Well, they simply can't. Once it's on the internet, then it's public for all eternity... (ok, I'm joking here, but here's the thing: if someone gets your game, with license and credits file and all, then no-one, not you, not your testers, not a court can take that away from that user. Even if you change your terms or the credit list later, you can't change retrospectively the games already given away.)

Ah, you're right. The thing about three years is only relevant when releasing "object code" (e.g. an executable) on a physical medium such as a CD.

Nope, you misunderstood. It is relevant if you're offering the source code alongside the object code on a physical medium. Not that it matters, because nobody uses physical medium any more.

I doubt you have to keep offering the source code forever if you stop distributing the "object code", that would be unreasonable, but that's a different discussion...

But you do. If you distribute your program in object format to someone, then that someone can ask for the source any time. That's why everybody is just offering the source on network servers, to come around this restriction. (Even Linux distros that still ship a physical medium like a DVD installer offer the source on the internet, and not on that DVD.)

I don't want to prevent anyone from removing the credits.

Again, you don't have the authority. Besides, all free licenses (with an attribution requirement, like GPL, CC-BY, etc.) requires the preservation of the attribution, so removing the credits is clearly violating the license terms (regardless what you want or don't want).

Use Public Domain or CC0, which don't have an attribution requirement (and face the consequences that your game will be stolen and falsely attributed to someone else, or even worse, someone else will make a fortune by claiming its his and selling it.)

What I don't want is that I should in some way be obliged to demand that they remove them if the persons in the credits come to me saying they have changed their mind.

Don't you worry, you're not. No-one can demand that. Even if a tester changes their mind, they can only demand not to put their names on the credit from now on, but they can't demand to change all copies in retrospect.

Cheers,
bzt

[Peter]

What you say is essentially what I had planned to do except that I was thinking about ways to enforce real or at least real-looking names

I'm not sure you have the authority to do that.

by perhaps rejecting them as testers if they choose a weird looking name.

That you can do.

I was thinking people could apply to become "testers". At the same time they might answer a few questions, one of them being if they want to be included in the credits and what name they want to use. I would only send them the game if they get accepted as testers. If I don't like the name they have chosen, or I don't want them to be testers for other reasons, then I could just decline.

As far as I know there is no legal requirements to mention "testers" so if they later want to change to a different name that I don't accept they would be left with the option of (1) try asking again with a different name, (2) using the previously accepted name or (3) not having the name listed at all.

I'm not saying I will do it this way. I'm just brainstorming. If it's too much trouble I might decide not to have any official testers.

At the end of films you don't see a lot of made up names, do you?

Actually, you do, all the time. Almost every name in every movie's end credits is made up.

I didn't know that. I'm a bit sceptical. I understand some might use a "stage name" but if no one knows it's you then what's the point?

For example, in the Shutter Island credits, are you saying that the "propmakers" Billy Rampey, Michael Smith and Nelson Werntz are likely not their real names?

For example, does William Pitt, Robyn Fenty, Maurice Micklewhite, Kathleen Ruston, Norma Mortenson, Neta-Lee Hershlag ring a bell? What if I say their screen names, Brad Pitt, Rihanna, Michael Caine, Audrey Hepburn, Marilyn Monroe and Natalie Portman? And not just movies, same everywhere: Freddy Mercury, Bob Dylan, Lady Gaga, Sting...

You've got a good point, but they are still real-looking names. It's not like "bzt" or "drummyfish".

I will have to think more about this but it still doesn't avoid the concern that I had unless I force people to use fake names (which I don't consider an option).

As I have said, you don't have the authority to do that. You can reject a tester, but you can't tell 'em what nickname to choose, and you can't force them to use their real names either.

I could choose not to mention them at all.

Ah, you're right. The thing about three years is only relevant when releasing "object code" (e.g. an executable) on a physical medium such as a CD.

Nope, you misunderstood. It is relevant if you're offering the source code alongside the object code on a physical medium. Not that it matters, because nobody uses physical medium any more.

But then the user already have the source code so there is no need for a three years offer. In that case I think it's 6a, instead of 6b, that is relevant.

6. Conveying Non-Source Forms.
You may convey a covered work in object code form under the terms of sections 4 and 5, provided that you also convey the machine-readable Corresponding Source under the terms of this License, in one of these ways:
a) Convey the object code in, or embodied in, a physical product (including a physical distribution medium), accompanied by the Corresponding Source fixed on a durable physical medium customarily used for software interchange.
...

I doubt you have to keep offering the source code forever if you stop distributing the "object code", that would be unreasonable, but that's a different discussion...

But you do. If you distribute your program in object format to someone, then that someone can ask for the source any time. That's why everybody is just offering the source on network servers, to come around this restriction. (Even Linux distros that still ship a physical medium like a DVD installer offer the source on the internet, and not on that DVD.)

But that means you essentially would have to keep offering every old version forever. I don't think that is always what all projects do.

It would also be quite controversial because I might no longer want to pay or be able to afford the cost of hosting the files, electricity, etc. I might want to sell my house and move to a hut in the woods and live from bilberries and fish. Just because I happened to release some software under the GPL I cannot do that? I'm doomed to upheld this "service" for the rest of my life?

Is it section 6d that you interpret this way?

6. Conveying Non-Source Forms.
You may convey a covered work in object code form under the terms of sections 4 and 5, provided that you also convey the machine-readable Corresponding Source under the terms of this License, in one of these ways:
...
d) Convey the object code by offering access from a designated place (gratis or for a charge), and offer equivalent access to the Corresponding Source in the same way through the same place at no further charge. You need not require recipients to copy the Corresponding Source along with the object code. If the place to copy the object code is a network server, the Corresponding Source may be on a different server (operated by you or a third party) that supports equivalent copying facilities, provided you maintain clear directions next to the object code saying where to find the Corresponding Source. Regardless of what server hosts the Corresponding Source, you remain obligated to ensure that it is available for as long as needed to satisfy these requirements.
...

If you are no longer "conveying" the object code then I think it follows that you should no longer have to "convey" the source code.

I'm not a lawyer but I doubt "for as long as needed" means "for ever". It probably just means the user should have a fair chance to download it. It would not be right to give the "object code" away and then give the user only 2 seconds to download the source code.

Update: When I read this again I note that it is written in connection to "server hosts" so I think the intention might just be that you cannot blame the server provider for the source code not being available. E.g. if you host your code on github.com and it goes down for a considerable time you cannot keep offering the "object code" without actually making an effort to come up with an alternative way to offer the source code.

I don't want to prevent anyone from removing the credits.

Again, you don't have the authority. Besides, all free licenses (with an attribution requirement, like GPL, CC-BY, etc.) requires the preservation of the attribution, so removing the credits is clearly violating the license terms (regardless what you want or don't want).

If you are thinking of section 7b ...

Requiring preservation of specified reasonable legal notices or author attributions in that material or in the Appropriate Legal Notices displayed by works containing it;

... then note that this is under "7. Additional Terms" which are terms that I could add to the license if I want but I don't have to. Personally I don't plan to make use of 7b.

There is also a requirement to display "Appropriate Legal Notices" if the original program did so (see 5d) but that seems to only have to contain copyright, warranty and license information.

An interactive user interface displays “Appropriate Legal Notices” to the extent that it includes a convenient and prominently visible feature that (1) displays an appropriate copyright notice, and (2) tells the user that there is no warranty for the work (except to the extent that warranties are provided), that licensees may convey the work under this License, and how to view a copy of this License. If the interface presents a list of user commands or options, such as a menu, a prominent item in the list meets this criterion.

If people do decide to use their real names I'm still left with the original question about what to do if they later want it removed.

Well, they simply can't. Once it's on the internet, then it's public for all eternity... (ok, I'm joking here, but here's the thing: if someone gets your game, with license and credits file and all, then no-one, not you, not your testers, not a court can take that away from that user. Even if you change your terms or the credit list later, you can't change retrospectively the games already given away.)

What I don't want is that I should in some way be obliged to demand that they remove them if the persons in the credits come to me saying they have changed their mind.

Don't you worry, you're not. No-one can demand that. Even if a tester changes their mind, they can only demand not to put their names on the credit from now on, but they can't demand to change all copies in retrospect.

"Can't" legally, or "can't" because it's infeasible?

Even if I have the permission to publish the names, do I have the permission to give that permission to others?

If the GDPR demands that they should be able to revoke their "consent" as simple as they gave it then perhaps I did something illegal when I distributed the names the way I did? Maybe I needed to require that people that republish my game remove the names or ask for permission? These are concerns I have.

It's not just about me, it's also about the people who copy and redistribute my game. I don't want there to be any legal question marks or too much problem for them to have to deal with.

I'm thinking one approach might be to simply be very clear to the tester about what I'm planning to do from the start, and give no promises that the credits will be kept forever (at least not by others). Then in the files I make it very clear for the people who redistributes my game that there is absolutely no legal requirement to keep these credits and that they should feel no shame in removing them if they want to avoid concerns with GDPR or whatever. I'm just brainstorming again...

[PeterX]

Even if I have the permission to publish the names, do I have the permission to give that permission to others?

If you think this way the worries will never end. If you don't believe us that it is legal, then you have only one option: Ask a lawyer. I doubt that we can help you with advice if you think this way.
Edit: Sorry if I sound mean. I just think this can become a bottomless pit.

Being not a lawyer, I think it is legal to give credit in a GPL program.

I'm thinking one approach might be to simply be very clear to the tester about what I'm planning to do from the start, and give no promises that the credits will be kept forever (at least not by others). Then in the files I make it very clear for the people who redistributes my game that there is absolutely no legal requirement to keep these credits and that they should feel no shame in removing them if they want to avoid concerns with GDPR or whatever.

This sounds like a good solution!

[Peter]

Even if I have the permission to publish the names, do I have the permission to give that permission to others?

If you think this way the worries will never end. If you don't believe us that it is legal, then you have only one option: Ask a lawyer. I doubt that we can help you with advice if you think this way.
Edit: Sorry if I sound mean. I just think this can become a bottomless pit.

Sorry if I ramble. For me this is a process to make sense of it all.

Being not a lawyer, I think it is legal to give credit in a GPL program.

Personally I don't think the fact that the game will be GPL-licenced is relevant. I probably shouldn't have allowed the discussion to be sidetracked by it but I found the discussion interesting.

I'm thinking one approach might be to simply be very clear to the tester about what I'm planning to do from the start, and give no promises that the credits will be kept forever (at least not by others). Then in the files I make it very clear for the people who redistributes my game that there is absolutely no legal requirement to keep these credits and that they should feel no shame in removing them if they want to avoid concerns with GDPR or whatever.

This sounds like a good solution!

Maybe... I'm not finished with my game yet so I have some more time to think about it.

[Peter]

I said earlier that I would want to "keep the older versions downloadable" but to be honest this is just a luxury problem. If something happened that legally forced me to remove them it wouldn't be the end of the world as long as it doesn't affect the newer versions negatively.

Assuming GPL do not require me to provide the source code for ever, which I don't think it does. See discussion above.

[bzt]

But then the user already have the source code so there is no need for a three years offer. In that case I think it's 6a, instead of 6b, that is relevant.

Nope, because that's also about distributing on a physical medium. As I've said, nobody uses physical medium any more, so it is actually section 6e that applies:

6e) Convey the object code using peer-to-peer transmission, provided you inform other peers where the object code and Corresponding Source of the work are being offered to the general public at no charge under subsection 6d.

And in 6d:

6d) ...and offer equivalent access to the Corresponding Source in the same way through the same place at no further charge. You need not require recipients to copy the Corresponding Source along with the object code. If the place to copy the object code is a network server, the Corresponding Source may be on a different server (operated by you or a third party)

So if you distribute your game over the internet, then it is enough to offer a downloadable Corresponding Source (and it's fine if the source is separate from your downloadable object).

But that means you essentially would have to keep offering every old version forever. I don't think that is always what all projects do.

Yes, exactly that's what's required, and that's what projects do. But you make this sound a bigger problem than it actually is, code hosting sites like sourceforge, github, gitlab, codeberg etc. make this extremely easy without a fuss. (FYI, the GNU project, the one that originally used GPL, had ftp sites with all available versions like this for example. But they don't do that any more, they have moved to git, like everybody else.)

It would also be quite controversial because I might no longer want to pay or be able to afford the cost of hosting the files, electricity, etc. I might want to sell my house and move to a hut in the woods and live from bilberries and fish. Just because I happened to release some software under the GPL I cannot do that? I'm doomed to upheld this "service" for the rest of my life?

Nope. Nobody said that you must host the source yourself, actually GPL says using a third party free code hosting service is perfectly fine. Just upload the source to sourceforge for example and you can go fishing knowing you don't have unfinished business left behind.

If you are no longer "conveying" the object code then I think it follows that you should no longer have to "convey" the source code.

Yes, but you're responsible for the object code already conveyed, you must provide the source for those.

note that this is under "7. Additional Terms" which are terms that I could add to the license if I want but I don't have to. Personally I don't plan to make use of 7b.

Yep, but section 4 also requires that you

You may convey verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice; keep intact all notices stating that this License and any non-permissive terms added in accord with section 7 apply to the code

Whether those notices include the contributions of the testers is up to section 7, but with or without section 7 Additional terms, you must keep the notices intact.

There is also a requirement to display "Appropriate Legal Notices" if the original program did so (see 5d) but that seems to only have to contain copyright, warranty and license information.

Yes. I think at the end of the day it comes to the question whether you consider testers authors or not (in other words, do you want give credits to them in the notices or not). They contribute to the project considerably that's for sure, but one could argue that they don't write the source code, so no attribution required for them. It's up to the project owner to decide that part, I guess. (I always list testers as contributors in my project's readme (for example) because that's the polite thing to do, and I use the nick they've used for reporting the feedback, nobody complained so far.)

Cheers,
bzt

[Peter]

Ah, you're right. The thing about three years is only relevant when releasing "object code" (e.g. an executable) on a physical medium such as a CD.

Nope, you misunderstood. It is relevant if you're offering the source code alongside the object code on a physical medium. Not that it matters, because nobody uses physical medium any more.

But then the user already have the source code so there is no need for a three years offer. In that case I think it's 6a, instead of 6b, that is relevant.

Nope, because that's also about distributing on a physical medium.

I was referring to the situation you mentioned: "offering the source code alongside the object code on a physical medium"

As I've said, nobody uses physical medium any more, so it is actually section 6e that applies: [...] And in 6d: [...] So if you distribute your game over the internet, then it is enough to offer a downloadable Corresponding Source (and it's fine if the source is separate from your downloadable object).

I agree.

Nope. Nobody said that you must host the source yourself, actually GPL says using a third party free code hosting service is perfectly fine. Just upload the source to sourceforge for example and you can go fishing knowing you don't have unfinished business left behind.

Nothing last forever, not even those big popular code hosting sites will.

Yes, but you're responsible for the object code already conveyed, you must provide the source for those.

You keep saying that but I have still not seen it written in the actual license text.

note that this is under "7. Additional Terms" which are terms that I could add to the license if I want but I don't have to. Personally I don't plan to make use of 7b.

Yep, but section 4 also requires that you

You may convey verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice; keep intact all notices stating that this License and any non-permissive terms added in accord with section 7 apply to the code

Whether those notices include the contributions of the testers is up to section 7, but with or without section 7 Additional terms, you must keep the notices intact.

Your bolding is misleading.

It doesn't say you should keep all notices intact.

What it says is that you should "keep intact all notices stating that this License [...] apply to the code".

I think at the end of the day it comes to the question whether you consider testers authors or not (in other words, do you want give credits to them in the notices or not). They contribute to the project considerably that's for sure, but one could argue that they don't write the source code, so no attribution required for them.

The main thing I would want the "testers" to do is just to play though the game (and hopefully enjoy it) and then send the "recordings" and any additional feedback to me. If I see that many people get stuck at some specific part of the game then I can make it slightly easier, etc.

Of course I would like them to report bugs and also language errors (especially since English is not my native language) but I don't want them to play in an unnatural way in a quest to find bugs. That's not the intention.

I don't really know if this is going to work... Will anyone want to test my game? Will they have the patience to play until the end? Maybe it would be better to include some opt-in recording and feedback sending feature in the official game...

[Peter]

My original plan, before I had thought about testers, were to not mention any names in the game itself (not even my own). Now I had an idea that I could still do it like that and instead only show the testers on my website.

I think there are several advantages with this approach.

1. The names remain under my control and I can easily allow people to opt in and out (regardless of whether this is legally required by the GDPR).
   
   
2. It makes it easier for people who want to distribute slightly modified versions of the game because they don't have to face the dilemma of having to decide how to portray the names.
   
   For example, if the game displayed "GameTitle, a game by MyName" then should they
   
   
   1. add their own name and/or make my name less prominent at the risk of it looking like they are trying to take credit for something they didn't do (or only made a very small part of), or should they
      
      
   2. keep it the way it is at the risk of it looking like it's the original version the way I intended the game to be?
   Not having my name in the game makes the situation much easier. It would be right by default. They could still add my name if they want, for example if they wanted to say that it's "a game based on my game" or it's "my game with modifications" but that would be their decision.

Of course I would have to include names of copyright holders in the files and possibly other places where legally required but that's a different problem.

[bzt]

Nothing last forever, not even those big popular code hosting sites will.

True, but that's not the point. If such a code hosting goes off-line, that's beyond your control. You have uploaded the source and made it available to the general public as the license expected from you, and that's only what matters (from your point of view at least).

Yes, but you're responsible for the object code already conveyed, you must provide the source for those.

You keep saying that but I have still not seen it written in the actual license text.

That's the whole purpose of Open Source.

But if you need the quotes, here you go (from section 1 and 6):

The “source code” for a work means the preferred form of the work for making modifications to it. “Object code” means any non-source form of a work.[...]The “Corresponding Source” for a work in object code form means all the source code needed to generate, install, and (for an executable work) run the object code and to modify the work

You may convey a covered work in object code form under the terms of sections 4 and 5, provided that you also convey the machine-readable Corresponding Source

So yes, if you provide object code for your GPL'd project, then you must also provide all the source code for it.

Your bolding is misleading.

It doesn't say you should keep all notices intact.

What it says is that you should "keep intact all notices stating that this License [...] apply to the code".

Nope, it's not misleading, it is in the part that you cut out elegantly.

The only question is, how do your consider your tester's work, should they be added into the license's notice section or not. You might decide not to, and only list them in a separate file (or not mention them at all, but I think that would be rude). Using a separate file is a perfectly valid solution too (in which case the license's notice part should only contain the copyright holders). Lots of projects do this, specially the ones with lots of contributors. For example, see Battle for Wesnoth, its LICENSE just says "FSF", and merely relies on the github's "Contributors" feature to get an always up-to-date contributors list (and contributors are not listed in the license's notice section). An example from the opposite side, zlib's LICENSE file directly lists all contributors (but there's only two, may I add). Or stb_image (but this last one isn't the best example as it uses a no-attribution license, so listing the contributors isn't required at all, yet there it is).

And to give you a bad example as well, here's a CONTRIBUTORS.txt and another one a'la Google. The concept is the same, separate authors and contributors, but this particular implementation is in direct violation of the GDPR (what else could one expect from Google, one of the world's biggest ad-spammer company?). Seriously, don't do it like this, don't use real names AND email addresses as well. Google might get away with it, but I'm certain you won't.

Cheers,
bzt

[Peter]

If such a code hosting goes off-line, that's beyond your control. You have uploaded the source and made it available to the general public as the license expected from you, and that's only what matters (from your point of view at least).

That is precisely opposite to what 6d says.

Regardless of what server hosts the Corresponding Source, you remain obligated to ensure that it is available for as long as needed to satisfy these requirements.

Yes, but you're responsible for the object code already conveyed, you must provide the source for those.

You keep saying that but I have still not seen it written in the actual license text.

That's the whole purpose of Open Source.

There are plenty of licenses that are considered free/open source that doesn't require that you provide the source code. Of course, if all source code vanished the software would no longer be free/open source but that's just how it is.

Note that the original author can do pretty much whatever he wants because he owns the copyright (assuming it's not a derived work). The license terms only really affect other people who want to distribute and/or create their own derived works based on it.

GPLv3 is a so-called "copyleft" license which essentially means that when you give the software to others you need to also give the source code and the same rights as was given to you. If you had a website where people could download the exe of a GPL-licensed program that had been created by someone else you are responsible for telling people where they can find the source code. If the original author had stopped providing the source code, and it couldn't be found anywhere else, that means you would then be responsible for making the source code available. If you are unable provide the source code, because you cannot find it, you would no longer be allowed to distribute the exe from your website.

Nope, it's not misleading, it is in the part that you cut out elegantly.

What it's saying is that you should preserve all notices that state that the code is being licensed under the GPL. The part that I "cut out" just made it clear that if there are notices saying that some of the "additional terms" in section 7 also apply to the code then you need to preserve those too.

The only question is, how do your consider your tester's work, should they be added into the license's notice section or not.

The testers are not copyright holders so they would not get a copyright notice.

For example, see Battle for Wesnoth, it merely relies on the github's "Contributors" feature to get an always up-to-date contributors list

I guess those "contributors" are people who have uploaded code to the project. My testers would not write or upload any code. It's not that sort of tester.

An example from the opposite side, zlib's LICENSE file directly lists all contributors (but there's only two, may I add).

But those are the copyright holders and are there for legal reasons. I think this is quite different from a list of all the people who have contributed to the project in one way or another.

If you look at the files in the src folder of Battle for Wesnoth you'll see they have similar copyright notices with names in most of the files.

I don't worry about distributing the names of copyright holders, especially if the license demand it. The GDPR lists "processing is necessary for the performance of a contract to which the data subject is party" and "processing is necessary in order to protect the vital interests of the data subject" as lawful purposes for processing. A license is more or less a contract, isn't it? And it is in the persons interest that I keep the copyright notices intact.

It's other names that I'm not required to have that I'm asking about in this thread.

[bzt]

That is precisely opposite to what 6d says.

You have lost the context. I'm not talking about live projects here, I'm talking about old versions and the case when you abandon your project and go fishing for good, because that's what you asked. For an actively developed project you'd look for another network server ASAP obviously.

You keep saying that but I have still not seen it written in the actual license text.

Read again your very own Section 6d quote carefully. You quoted that, not me, so you must have seen it: "you remain obligated to ensure that it [Corresponding Source] is available".

There are plenty of licenses that are considered free/open source that doesn't require that you provide the source code.

Name one. All Free and Open Source licenses demand that you provide the source code, that's what makes them "Open Source". Without source, you can't modify the work, which is one of the fundamental freedoms required by FOSS.

I guess those "contributors" are people who have uploaded code to the project. My testers would not write or upload any code.

The testers are not copyright holders so they would not get a copyright notice.

It's other names that I'm not required to have that I'm asking about in this thread.

Looks like your thread is meaningless as you have already answered your own question.

You don't consider your testers contributors, as you have put it, "testers won't get a notice in the license", so GPL demands absolutely nothing from you notice-wise in this case.

There's nothing for you to worry about, you can delete names from your credits.txt any time (but you can't demand the same from people who have forked your source code, the way as your repo was when they forked (or conveyed the source by any other means) applies. As I've said, no retrospect changes possible in law. Your tester MAY ASK them to change their copy of the credits.txt too, but they are not obliged to follow and do so, at least not by the terms of the license. They might be required to do so because of GDPR, but that's a different question and has nothing to do with GPL).

Cheers,
bzt

[Peter]

You have lost the context. I'm not talking about live projects here, I'm talking about old versions and the case when you abandon your project and go fishing for good, because that's what you asked. For an actively developed project you'd look for another network server ASAP obviously.

It was you that claimed you had to provide the source code forever, not me.

Read again your very own Section 6d quote carefully. You quoted that, not me, so you must have seen it: "you remain obligated to ensure that it [Corresponding Source] is available".

You are taking the quote out of context. You only need to do that if you "convey a covered work in object code form".

There are plenty of licenses that are considered free/open source that doesn't require that you provide the source code.

Name one. All Free and Open Source licenses demand that you provide the source code, that's what makes them "Open Source". Without source, you can't modify the work, which is one of the fundamental freedoms required by FOSS.

I can name a few from that list ...

• The MIT license (also known as the Expat license).
• The Zlib license.
• All the various BSD licenses.
• The Apache License.

The original author would obviously need to release the source code (not because it's required by the license but because it wouldn't be "open source" otherwise) but people who redistribute the software (modified or unmodified) don't need to release the source code (meaning derivative works are not necessarily "open source").

Looks like your thread is meaningless as you have already answered your own question.

Not really. The original question was never really about copyright holders and licenses.

I still don't know the answer to the original question. At least not when it comes the third-party distribution of the names.

What has happened however is that I have gotten a clearer picture and come up with some alternative ways in which I can probably avoid my concerns regardless of how the GDPR apply exactly.

[bzt]

Read again your very own Section 6d quote carefully. You quoted that, not me, so you must have seen it: "you remain obligated to ensure that it [Corresponding Source] is available".

You are taking the quote out of context. You only need to do that if you "convey a covered work in object code form".

And what other options are there? Convey the work in source form in the first place? Either way, you are either obligated to provide the source or you do provide the source...

I can name a few from that list ...

• The MIT license (also known as the Expat license).
• The Zlib license.
• All the various BSD licenses.
• The Apache License.

The original author would obviously need to release the source code, otherwise it wouldn't be open source, but people who redistribute the software (modified or unmodified) don't need to release the source code.

Now you're just obviously trolling, your comment made it clear that you're aware you make no sense here.

All of the licenses you listed demands that your work and its derivatives as well must be modifiable to be Free and Open Source, which implies the source must be available:

• MIT "to deal in the Software without restriction, including without limitation the rights to use, copy, modify" (you can't modify without a source),
• zlib "Altered source versions must be plainly marked as such" (pretty clear that altered source has to be public for that),
• BSDs "Redistributions of source code must retain the above copyright" (directly naming "source code"), and
• Apache "“Source” form shall mean the preferred form for making modifications, including but not limited to software source code" (also directly naming "source code").

If you don't provide the source to the general public, that's called "closed source", and not "open source", end of discussion. (Copyleft licenses such as GPL don't allow you to use open source in proprietary software, while others allow it, but as soon as you do include your software in a proprietary closed source application, you can't say "licensed under a FOSS license" any more, because it becomes proprietary, naturally.)

I still don't know the answer to the original question. At least not when it comes the third-party distribution of the names.

Then I repeat my answer: it is the license that controls how third-party distribution might happen, and your question has absolutely nothing to do with the license, hence your question has nothing to do with third-party distribution.

Cheers,
bzt

[Peter]

And what other options are there? Convey the work in source form in the first place?

That's one option.

Either way, you are either obligated to provide the source or you do provide the source...

If you no longer "convey a covered work in object code form" then you are no longer obligated to provide the source.

Of course this would be up to the courts to decide. I don't think you can just remove it from one day to the next without any warning. But if they have been given plenty of time to download, and they chose not to do it, then I don't think they can come several years later and demand that you give them the source code.

Now you're just obviously trolling, your comment made it clear that you're aware you make no sense here.

Things have become more clear as the discussion has carried on and after refreshing my memory by reading the license texts so I'm more confident about my answers now.

Note that I updated my previous post to make it more clear. Not sure if you saw it...

The original author would obviously need to release the source code (not because it's required by the license but because it wouldn't be "open source" otherwise) but people who redistribute the software (modified or unmodified) don't need to release the source code (meaning derivative works are not necessarily "open source").

All of the licenses you listed demands that your work and its derivatives as well must be modifiable to be Free and Open Source, which implies the source must be available:

Not sure how you can come to that conclusion. The licenses I listed don't even mention "free" or "open source".

The word "free" is present in some places but it doesn't have the same meaning as in the term "free software".

MIT "to deal in the Software without restriction, including without limitation the rights to use, copy, modify" (you can't modify without a source),

Yes, those are the rights that are given to me. I don't need to give the same rights to others that I distrubute the software to.

It's not correct that I cannot modify the software without the source code. I could use a hex editor, a decompiler, etc. This is even more true when the license is applied to something other than code. If it's applied to a PDF the "source code" (using GPLv3's definition of source code: "the preferred form of the work for making modifications to it") would probably be the .odt or LaTeX file from which it was generated. If all I had was the PDF file it wouldn't stop me from making modifications even though it wouldn't be the preferred format.

zlib "Altered source versions must be plainly marked as such" (pretty clear that altered source has to be public for that),

All this line really says is that you should mark "source versions" that have been modified.

It doesn't mean you are not allowed to distribute versions that are not "altered source versions".

BSDs "Redistributions of source code must retain the above copyright" (directly naming "source code"),

The license clearly gives you permission to distribute the software without source code: "Redistribution and use in source and binary forms, with or without modification, are permitted"

The fact that you must keep the copyright notices does not change that.

Apache "“Source” form shall mean the preferred form for making modifications, including but not limited to software source code" (also directly naming "source code").

There is no requirement that you distribute the software in the "preferred form".

If you don't provide the source to the general public, that's called "closed source", and not "open source", end of discussion.

I agree 100%. The license is only part of what makes it open/free. It's a tool that open source software can use.

Just marking your code with an "open source license" without actually giving it to anyone would not automatically make it "open source" from anyone's perspective.

Copyleft licenses such as GPL don't allow you to use open source in proprietary software, while others allow it, but as soon as you do include your software in a proprietary closed source application, you can't say "licensed under a FOSS license" any more, because it becomes proprietary, naturally.

If this is so obvious then I don't see what we have really been arguing about here above.

The license would still apply inside the proprietary software. It might require it to show certain notices for example. The program as a whole would obvious not be FOSS.

Then I repeat my answer: it is the license that controls how third-party distribution might happen, and your question has absolutely nothing to do with the license, hence your question has nothing to do with third-party distribution.

I want to allow free third-party distribution without people having to worry too much about legal questions. Using a free open-source license that people are familiar with is a good way of doing that.

I don't think the law cares whether I use a license or not. I still need to comply with the GDPR, and the people who redistribute my game need to comply with the GDPR.

I don't want people to have to worry about GDPR issues because then they might say "better safe than sorry" and stop distributing my game. I don't want that to happen, not for that reason.

------------------------------------------

If you don't provide the source to the general public, that's called "closed source", and not "open source", end of discussion.

I agree 100%. The license is only part of what makes it open/free. It's a tool that open source software can use.

Actually, I only agree to 95%.

The source code doesn't need to be made available to the "general public". It's enough that it's available to the people who receives the software. Then they can make it available to everyone if they want.

I think the terms "open source" and "free software" are often used synonymously so I'm going to quote both from the Open Source Definition published by the Open Source Initiative and from the Free Software Definition published by the Free Software Foundation.

The rights attached to the program must apply to all to whom the program is redistributed without the need for execution of an additional license by those parties.

Where some form of a product is not distributed with source code, there must be a well-publicized means of obtaining the source code for no more than a reasonable reproduction cost

The first quote makes it pretty clear that the rights only need to apply to the persons who receive the program. The second quote implies that it's alright to only distribute the source code together with the program and since the program is not necessarily available to everyone that means not everyone need to have access to the source code.

A program is free software if the program's users have the four essential freedoms
...

• The freedom to study how the program works, and change it so it does your computing as you wish (freedom 1). Access to the source code is a precondition for this

....

It says that it's the program's users that should have these freedoms. If you are not a user then you don't necessarily need to have access to the source code even though it's free software.


Of course, in this day and age, if someone finds the program useful the chances are very big that it will be made available to the general public very soon.

[bzt]

Anyone reading your post can see that you're contradicting yourself, multiple times. Your post simply does not even deserve to be responded to in detail. You're obviously just arguing for the sake of the argument, and your contradictions make it clear that you actually KNOW that you're wrong.

Therefore I'll only quote one thing from it (which should be obvious to anybody except trolls, BTW):

There is no requirement that you distribute the software in the "preferred form".

Yes, there is, "preferred form" is the source, and without distributing the source you have no right to call that software FOSS any more, period. (And even you wrote at some point that "need to release the source code, otherwise it wouldn't be open source", which directly contradicts what you're saying now...)

You should shake hands with Julius, you're both incapable of admitting your own mistakes, you are both deliberately misinterpreting FOSS licenses and you are both abusing Free and Open Source Software.

There's nothing more I want to say to you.
bzt

[Peter]

I think I see what the misunderstanding is.

The mistake I think you do is that you mix up the definition of being FOSS with the license itself. These two are separate.


What does it mean to be FOSS?

I think the two most important definitions are the Open Source Definition and the Free Software Definition. Different people might prefer one or the other but they are similar and the difference doesn't matter for this discussion.

Note that these definitions are not licenses. They just explain what it means to be "FOSS".


What is a license?

A software license is a set of terms that the author (copyright holder) can apply to the software that tells others under what conditions they are allowed to use and redistribute the software.

It doesn't put any requirements on the author. It's just the author saying, to use my software in this and that way, you have to fulfil these criteria....


What is a FOSS license?

A FOSS license is a license that is suitable for making FOSS software.

It doesn't automatically mean that the software is FOSS if it doesn't live up to the definition of what it means to be FOSS.

Note that it's usually not the license itself that claims that it is a FOSS license. Instead other people have looked at the terms of the license to decide if it qualifies as being a FOSS license and then compiled lists of such licenses. Here are three such lists:
https://www.gnu.org/licenses/license-list.html
https://opensource.org/licenses/
https://docs.fedoraproject.org/en-US/le ... -licenses/

[bzt]

Okay, one last post, just because you made me ROTFL :-)

It doesn't put any requirements on the author.

For example, from GPL: "you remain obligated", "provided that you comply", "you waive", "you disclaim", "your work must carry", "provided you inform", etc. etc. etc. Or do you deny that GPL is a FOSS license like Julius did, perhaps?

What is a FOSS license?

A FOSS license is a license that is suitable for making FOSS software.

It doesn't automatically mean that the software is FOSS if it doesn't live up to the definition of what it means to be FOSS.

Hahahaha, this is hilarious! You do see your own absurdity, right? Do you seriously think that if you don't adhere to the terms and conditions then you won't violate that FOSS license...?

Answer this: why is it then that these are called explicitly "FOSS licenses", and not just "licenses"? Hmmmm? Why is that? What do you think what does "comply" in licenses that comply with the Open Source Definition mean in the first place? Do you seriously think that a license can be an OSI-approved FOSS license without complying to the "the program must include source code, and must allow distribution in source code" requirement?

You already made fool of yourself when you said courts should decide what other forms than "object code" and "source" there are... (hint from GPL "“Object code” means any non-source form of a work.", so by definition no other options can exist). I haven't thought that this is possible, but by claiming that non-FOSS software might use FOSS license, you've just managed to make even bigger fool of yourself. :-) :-) :-)

There's no such thing as closed-source FOSS. If a software does not live up to the definition of FOSS, then that software is not using and cannot use a FOSS license. This is elementary, Watson.

Cheers,
bzt

[Julius]

Or do you deny that GPL is a FOSS license like Julius did, perhaps?

lol, I never did that.

Anyway, lets stop this discussion here, I don't think it is worthwhile to continue.