I disagree with the premise that open source is something that should rise above a developer's need to eat. Spending your time creating software and then releasing it for free is something to be...
I disagree with the premise that open source is something that should rise above a developer's need to eat.
Spending your time creating software and then releasing it for free is something to be applauded, but I do not think it should be expected.
Most people do not have the money/energy to spend large amounts of time creating free software.
I've seen many times where a project will come to a halt because the creator cannot afford to spend the time it requires to develop instead of working and no one is willing to pick up the slack or pay them to continue (such as RES which just announced it's in maintenance mode).
But I strongly disagree with the "walled garden" approach many are taking for it, the same could be done by more ethical means like advertising, affiliate marketing, repaying the sponsor in other terms (like putting out a brief marketing video for them on youtube as is usually done).
This bit reminds me of the reaction anytime a game modder decides to put some of their content behind a paywall.
There'll be threads full of people complaining and asking why they didn't just put some ads up or add a donation link and long comment chains agreeing that those would've been the right moves.
Then the developer will come in with a response like "I've had a donation page linked on all my content for 2 years and I've gotten $12".
the same could be done by more ethical means like advertising
How is the tracking and user data collection of modern ads more ethical than simply and openly charging an upfront fee for your software?
This move is totally against the spirit of GPL/MIT/Apache licensed software which is all about openness and software freedom.
As parsley said, those licenses are about making the source available to end users.
Charging for software does not change this, it just limits your end users to people willing to pay for the software.
This is where the OP lost me. While I don't entirely like the idea of charging a fee for accessing OSS software, I like it a lot more than advertising. Advertising in general is terrible, it...
How is the tracking and user data collection of modern ads more ethical than simply and openly charging an upfront fee for your software?
This is where the OP lost me. While I don't entirely like the idea of charging a fee for accessing OSS software, I like it a lot more than advertising. Advertising in general is terrible, it twists market incentives into all sorts of weird directions.
“Free software” means software that respects users' freedom and community. Roughly, it means that the users have the freedom to run, copy, distribute, study, change and improve the software. Thus, “free software” is a matter of liberty, not price. To understand the concept, you should think of “free” as in “free speech,” not as in “free beer.” We sometimes call it “libre software,” borrowing the French or Spanish word for “free” as in freedom, to show we do not mean the software is gratis.
You may have paid money to get copies of a free program, or you may have obtained copies at no charge. But regardless of how you got your copies, you always have the freedom to copy and change the software, even to sell copies.
Realistically, I don't think this happens enormously often, at least for the small projects that would benefit from sponsor-only type arrangements. FairEmail is GPLv3, with a public repository,...
Realistically, I don't think this happens enormously often, at least for the small projects that would benefit from sponsor-only type arrangements.
FairEmail is GPLv3, with a public repository, and asks for payment to unlock full features, in addition to being somewhat nagware. Full sources are publicly available, and so, of course, it would be easy to circumvent the nag-screens and feature locks and release a full version, but while I know someone did, their fork seems mostly defunct and far less popular.
RCU is a fully paid AGPLv3 program. Links to sources and binaries are sent via mailing lists: payment purchases a year's subscription to those mailing lists, and neither the binaries nor the sources are publicly available. Yes, you can redistribute the sources, but I haven't seen them being posted frequently as circumvention, and I doubt the ability to get the software legally from others has pulled away many customers (parts of the source have shown up in several other projects for other purposes, but the author points to these as a good thing). The author appears to work for the FSF, so I assume they know what they're doing with regards to the GPL and software sales.
Conversely: CentOS was entirely this model. I do agree that for most people, for smaller projects, there is little incentive to look beyond the author for downstream copies.
Conversely: CentOS was entirely this model.
I do agree that for most people, for smaller projects, there is little incentive to look beyond the author for downstream copies.
You can't leave something with a known security bug up for 6 months that you've fixed in the main version. This means that you have to update everything that's actually required for the...
You can't leave something with a known security bug up for 6 months that you've fixed in the main version.
This means that you have to update everything that's actually required for the functioning of the free version immediately, and so it's essentially free for almost everyone in practice.
I'm not sure what you mean? It's often possible to backport security fixes to an older branch without also including newer features. It's a hassle though.
I'm not sure what you mean? It's often possible to backport security fixes to an older branch without also including newer features. It's a hassle though.
...Not really? Business model aside, I'm fairly sure it is perfectly valid to only share source with customers, at least under GPL. On the business side, I kind of dislike how completely different...
This move is totally against the spirit of GPL/MIT/Apache licensed software which is all about openness and software freedom.
...Not really? Business model aside, I'm fairly sure it is perfectly valid to only share source with customers, at least under GPL.
On the business side, I kind of dislike how completely different business (from the customer perspective) imitate each other. It gives the feeling that at the C-suites level github is not different from twitch, patreon or pornhub.
I think that if there are developers out there who want to embrace this model and have a project that is successful enough to make money using it, then more power to them. I don't understand how...
I think that if there are developers out there who want to embrace this model and have a project that is successful enough to make money using it, then more power to them. I don't understand how having more options for indie/FOSS developers to get paid for their work could be seen as anything but a step in the right direction.
And personally I loathe most forms of ads and marketing, so if this move means that more products and services move to a pay-with-your-wallet-not-your-privacy model then I'm all for that, too.
Looking at the official announcement page, it seems like they are basically now offering sponsorships on private repositories, which is... good? Like others have said, this does not violate...
Looking at the official announcement page, it seems like they are basically now offering sponsorships on private repositories, which is... good?
Like others have said, this does not violate GPL/MIT/Apache licenses. But what really gets me is that we've seen the years and years of abandoned open source repositories, both large and small, from burned out developers working on projects endlessly. Sometimes I get burned out at work and I get paid for it - I can't imagine having the stress of paying bills on top would really impact my projects.
The author makes this point
The pandemic has broken the financial backbone of many individuals as it is
but the issue is that very, very often it is large wealthy tech businesses that are using the free software without paying a dime and at the same time expecting absolute perfection. These organisations can absolutely afford to pay licenses, but refuse to. By giving a developer or a team a means of putting their foot down and demanding compensation, it is effectively changing the tide - if you want this code, pay up. That is completely fair.
I guess the counterargument can be made which is if you want to be paid, then explicitly make a license demanding payment and attach it to your code repository. Firstly, if someone wanted to just download the source code without paying, they could do that, and I'd imagine the proof that the entity stole it is now in the code creator(s) side to prove. Even if the entity wanting the code wasn't that bad, if they were large enough they could hire lawyers to find an escape clause in the license to make them exempt. Forcing the entity to sponsor to get the code, regardless of license, will allow the entity to get the code while the author(s) get their money.
The author then makes - to me at least - a massive stretch, by saying:
Imagine a world tomorrow where every single website or blog on the Internet is put behind a paywall, a world where money is the sole motivation to create content and the size of people's pockets determines what gets published.
But that hasn't happened where every single website/blog on the Internet is put behind a paywall. Hell, some news websites did put themselves behind a paywall, while other news websites used that as an advantage to keep their websites "unpaywalled" but financed by other means. Turns out options is a good thing, and for a community that is so starved of resources - financial or otherwise - it makes sense to give the open source community more ways of financing their work that they've been giving away for free for collectively years now.
This might be relevant - basically, let's have the cake and eat it too: You can impose via license that big businesses have to negotiate a license for your product. For everyone else, it's still...
but the issue is that very, very often it is large wealthy tech businesses that are using the free software without paying a dime and at the same time expecting absolute perfection. These organisations can absolutely afford to pay licenses, but refuse to. By giving a developer or a team a means of putting their foot down and demanding compensation, it is effectively changing the tide - if you want this code, pay up. That is completely fair.
This might be relevant - basically, let's have the cake and eat it too: You can impose via license that big businesses have to negotiate a license for your product. For everyone else, it's still open source. I like the idea a lot. Sadly, the FOSS I'd like to develop doesn't lend itself to this business model. For some other software, including the giant stack of libraries in ecosystems like python or rust, this might actually be a good model.
Ok that is super cool. The fact that a lawyer has worked on it gives me great confidence in the license - it just needs exposure! But thank you for sharing, very cool!
Ok that is super cool. The fact that a lawyer has worked on it gives me great confidence in the license - it just needs exposure!
I think the author of this article doesn’t understand the feature GitHub is releasing. The article reads like they think only ONE sponsor is possible, OR that sponsorship prices will be too high...
I think the author of this article doesn’t understand the feature GitHub is releasing. The article reads like they think only ONE sponsor is possible, OR that sponsorship prices will be too high for individuals to afford.
Projects can have multiple sponsors and often sponsorship is ANY amount. Effectively this turn into “pay what you think it’s worth”, which is a bet positive IMO
I disagree with the premise that open source is something that should rise above a developer's need to eat.
Spending your time creating software and then releasing it for free is something to be applauded, but I do not think it should be expected.
Most people do not have the money/energy to spend large amounts of time creating free software.
I've seen many times where a project will come to a halt because the creator cannot afford to spend the time it requires to develop instead of working and no one is willing to pick up the slack or pay them to continue (such as RES which just announced it's in maintenance mode).
This bit reminds me of the reaction anytime a game modder decides to put some of their content behind a paywall.
There'll be threads full of people complaining and asking why they didn't just put some ads up or add a donation link and long comment chains agreeing that those would've been the right moves.
Then the developer will come in with a response like "I've had a donation page linked on all my content for 2 years and I've gotten $12".
How is the tracking and user data collection of modern ads more ethical than simply and openly charging an upfront fee for your software?
As parsley said, those licenses are about making the source available to end users.
Charging for software does not change this, it just limits your end users to people willing to pay for the software.
This is where the OP lost me. While I don't entirely like the idea of charging a fee for accessing OSS software, I like it a lot more than advertising. Advertising in general is terrible, it twists market incentives into all sorts of weird directions.
I'm guessing the author didn't read this page https://www.gnu.org/philosophy/free-sw.en.html.
The goodnews/badnews aspect is that anyone who buys is free to redistribute the repo, provided they remove any trademarks (if relevant).
Realistically, I don't think this happens enormously often, at least for the small projects that would benefit from sponsor-only type arrangements.
FairEmail is GPLv3, with a public repository, and asks for payment to unlock full features, in addition to being somewhat nagware. Full sources are publicly available, and so, of course, it would be easy to circumvent the nag-screens and feature locks and release a full version, but while I know someone did, their fork seems mostly defunct and far less popular.
RCU is a fully paid AGPLv3 program. Links to sources and binaries are sent via mailing lists: payment purchases a year's subscription to those mailing lists, and neither the binaries nor the sources are publicly available. Yes, you can redistribute the sources, but I haven't seen them being posted frequently as circumvention, and I doubt the ability to get the software legally from others has pulled away many customers (parts of the source have shown up in several other projects for other purposes, but the author points to these as a good thing). The author appears to work for the FSF, so I assume they know what they're doing with regards to the GPL and software sales.
Conversely: CentOS was entirely this model.
I do agree that for most people, for smaller projects, there is little incentive to look beyond the author for downstream copies.
One strategy I’ve really liked is to provide a 6 month old version for free and the up to date version on a subscription service.
You can't leave something with a known security bug up for 6 months that you've fixed in the main version.
This means that you have to update everything that's actually required for the functioning of the free version immediately, and so it's essentially free for almost everyone in practice.
I'm not sure what you mean? It's often possible to backport security fixes to an older branch without also including newer features. It's a hassle though.
Depends on the content. For game mods that might not matter.
...Not really? Business model aside, I'm fairly sure it is perfectly valid to only share source with customers, at least under GPL.
On the business side, I kind of dislike how completely different business (from the customer perspective) imitate each other. It gives the feeling that at the C-suites level github is not different from twitch, patreon or pornhub.
I think that if there are developers out there who want to embrace this model and have a project that is successful enough to make money using it, then more power to them. I don't understand how having more options for indie/FOSS developers to get paid for their work could be seen as anything but a step in the right direction.
And personally I loathe most forms of ads and marketing, so if this move means that more products and services move to a pay-with-your-wallet-not-your-privacy model then I'm all for that, too.
Looking at the official announcement page, it seems like they are basically now offering sponsorships on private repositories, which is... good?
Like others have said, this does not violate GPL/MIT/Apache licenses. But what really gets me is that we've seen the years and years of abandoned open source repositories, both large and small, from burned out developers working on projects endlessly. Sometimes I get burned out at work and I get paid for it - I can't imagine having the stress of paying bills on top would really impact my projects.
The author makes this point
but the issue is that very, very often it is large wealthy tech businesses that are using the free software without paying a dime and at the same time expecting absolute perfection. These organisations can absolutely afford to pay licenses, but refuse to. By giving a developer or a team a means of putting their foot down and demanding compensation, it is effectively changing the tide - if you want this code, pay up. That is completely fair.
I guess the counterargument can be made which is if you want to be paid, then explicitly make a license demanding payment and attach it to your code repository. Firstly, if someone wanted to just download the source code without paying, they could do that, and I'd imagine the proof that the entity stole it is now in the code creator(s) side to prove. Even if the entity wanting the code wasn't that bad, if they were large enough they could hire lawyers to find an escape clause in the license to make them exempt. Forcing the entity to sponsor to get the code, regardless of license, will allow the entity to get the code while the author(s) get their money.
The author then makes - to me at least - a massive stretch, by saying:
But that hasn't happened where every single website/blog on the Internet is put behind a paywall. Hell, some news websites did put themselves behind a paywall, while other news websites used that as an advantage to keep their websites "unpaywalled" but financed by other means. Turns out options is a good thing, and for a community that is so starved of resources - financial or otherwise - it makes sense to give the open source community more ways of financing their work that they've been giving away for free for collectively years now.
This might be relevant - basically, let's have the cake and eat it too: You can impose via license that big businesses have to negotiate a license for your product. For everyone else, it's still open source. I like the idea a lot. Sadly, the FOSS I'd like to develop doesn't lend itself to this business model. For some other software, including the giant stack of libraries in ecosystems like python or rust, this might actually be a good model.
Ok that is super cool. The fact that a lawyer has worked on it gives me great confidence in the license - it just needs exposure!
But thank you for sharing, very cool!
I think the author of this article doesn’t understand the feature GitHub is releasing. The article reads like they think only ONE sponsor is possible, OR that sponsorship prices will be too high for individuals to afford.
Projects can have multiple sponsors and often sponsorship is ANY amount. Effectively this turn into “pay what you think it’s worth”, which is a bet positive IMO