Did using AGPL benefit Zed with job offers and fame and preventing companies from being sold without disclosing they were using Zed’s software, or practically anything else he wrote about?
Zed’s “waning fade” is pretty much self-chosen in my reading. He doesn’t give any talks anymore and seems pretty happy with the current way things are.
This is Zed at his best. Recent experiences have led me to a similar position as Zed’s, albeit for different reasons. BSD/MIT is typical in the Haskell community, but I’m not sure I’ll keep licensing my future OSS work like that.
I will keep existing projects on their current licenses.
this post and the ensuing discussion across the blogosphere have made me a proponent of the gpl for anything that could have commercial value downstream.
I arrived at this position because of people demanding our Haskell book be free & creative commons despite:
My coauthor being a not-wealthy homeschooling mother of two who works on the book in her spare time
My spending about half of the book’s development only taking light contract work or no work at all, burning my own savings so that I could focus on it
People’s attitude that they are entitled to our labor despite most of them making more money than myself or Julie has made us very hesitant to release the book as creative commons.
This and other things has me leaning towards keeping future OSS projects I start GPL. I won’t harass anyone else I work with into doing it, because quite frankly I think every programmer has had that experience and it’s infuriating. (cf. what I wrote above about creative commons)
inb4randos: I am not open to, nor am I interested in, debating the merits/demerits of our book being made CC. We have discussed the issue at length with serious & committed proponents of the commons and they admitted All Rights Reserved probably makes more sense for our needs.
agreed, cc books are awesome, but that should be each author’s individual and unpressured choice. it’s a shame some people don’t feel any cognitive dissonance in demanding someone else give their work away just because they would prefer that.
I’m really glad that you’re charging for the book, and I really hope it’ll be worthwhile for you both financially. It means that it’s fair for the reasons you mentioned, but also the readers end up with a better product (unlike a lot of the half-arsed free books on many topics out there).
People just have a weird sense of entitlement, and zero understanding of how much effort goes into a book.
Interestingly, for my book, I’ve only had one silly price demand that I recall (“it shouldn’t be any more than $5” – umm, why?).
This and other things has me leaning towards keeping future OSS projects I start GPL.
The problem is that a lot of software is cloud-hosted these days. The GPL does not bring you much there, because the software is not distributed. The AGPL addresses this problem, but is unfortunately incompatible with some other FLOSS licenses.
I used to use the GPL for the reasons stated, but the interesting thing is that a company actually contacted me to purchase (and paid for) a license for one of my Apache 2 Licensed projects. This project had a NOTICE file and the Apache License requires that its contents are displayed in the application or documentation. They basically paid me for not having to put this notice in their software.
Here’s my counterpoint to his argument: I don’t use GPL licensing on my open source software because it imposes restrictions on the use of my software that I don’t want to impose on my users. I write software because I want people to use it, not because I want to make it difficult to use.
In the end it comes down to your motivations I guess. The author laments, “I Don’t Want To Be Ignored Again”. If your main motivation in writing open source software is fame I guess that’s an important consideration for you, but I don’t think I’m alone in not giving a toss about fame and not feeling that writing software is all about having your ego stroked.
I write open source software because I like writing it and because I want to use it myself. Either way I’m happy to share with others. If I GPL my work it sets the legal bar too high for most commercial organizations so they just won’t use it. It’s no skin off my nose for them to use it so why not just let them by going BSD instead?
Absolutely. I’ve made the same argument you have; I open-source stuff in the hope that it will help people, and also because I find it freeing to give up ownership.
I’ve previously seen a lot of argument for the GPL on the basis that it helps people more than BSD does, which… is a legitimate difference of opinion, I suppose.
What is new to me about this author’s argument is that it’s a choice to use the GPL, with the motivation of wanting recognition! I disagree with it, but I still find it new and interesting. I’m highly sympathetic to concerns about exploitation, which were also discussed here at Throwing in the towel. Seeing those as aligned with the GPL is fascinating.
I agree with the theoretical arguments that GPL would help more people. In practice it hasn’t worked out that way - when I worked on GPL code no-one used it, when I work on BSD[1] code companies use it and some of them contribute back. If there were other organizations producing code then maybe the GPL model could work, but even if for-profit companies are only contributing 1/10 or 1/100 of what they write, that’s still more code than the rest of the world produces at all.
I don’t think GPL will give you fame either, at least in the long term. All it means is that your code will be replaced. libedit displaces readline, LLVM displaces GCC - when you release a good BSD library there’s no reason for anyone to ever replace it, whereas even if you release an excellent GPL library, if it does something useful then in the long term there will be enough people who need a BSD version that eventually one will be written and people will switch to that.
(I also think Shaw’s ego is vastly out of proportion with his ability, and the fact that he felt he was getting less fame than he deserved doesn’t at all convince me that this was objectively so).
From an ideological perspective, I actually disagree with the OP’s initial premise, which is that it’s the “author’s right to license their work.” I don’t think anyone has some innate right to appeal to monopoly protection over creations of the intellect (of which, code absolutely falls under).
With that said, I’m unlikely to be the type of person the OP had in mind when writing that first section. In particular, despite my ideological position, I absolutely agree with the OP that I would be a gigantic jerk if I ran around telling others that they didn’t have a right to choose the license they did. It’s just not a productive expression of my views on intellectual property.
And on the scale of damage that intellectual property laws do, the GPL is barely a blip.
Totally agree. This is why I choose BSD/MIT. The users of my code in my opinion are programmers. The idea that my choice could make people have to reimplement things due to a ideology just strikes me as being a jerk.
I know people say “but what if a company makes a ton of money off your code”. So? Who cares.
I honestly purposefully avoid GPL software for BSD/MIT, mostly because its really annoying to have to avoid using software due to ideology that I fundamentally disagree with. I’ve always been a BSD user at heart (and home), and I just don’t think GPL is worth the loss of freedoms for others. Basically I’m empathizing with other programmers that are just trying to get things done. I’m in the same boat, and think a more free license for programmers trumps a more free license for users.
Probably not an opinion a lot of people share but honestly, I think more people need to say no to GPL publically. LLVM is in my opinion the best example of why GPL is not the best license choice. It seemed to finally light a fire under gcc to improve.
Choose what license you want, but if there is a BSD/MIT/etc.. alternative, I’ll always pick that and work to fix it to beat the GPL version.
I don’t think anyone has some innate right to appeal to monopoly protection over creations of the intellect
I didn’t see anywhere in the article where it claimed that this was an innate human right. It’s a right that we as a society have chosen to grant to each other, theoretically based on consensus that it’s better than the alternative.
I don’t subscribe to the belief that “rights are given theoretically.” And I don’t know how you inferred how the OP was using the word. (I disagree with it in every sense of the word that I’m aware of, so you’re barking up the wrong tree with me personally.)
I didn’t mean that the rights are given theoretically; I meant that the consensus behind copyright is theoretical. Due to the way it’s been abused by media empires, many people consider it to have outlived its usefulness and to be more trouble than it’s worth. But the fact is it remains the law.
Just to be clear, do you disagree that society has any right to enforce this monopoly to begin with? As in, the Statute of Anne was a mistake and never should have been passed because it overstepped the authority of the crown? It’s one thing to say that copyright legislation has gone too far, and quite another to say that it never had any legitimacy to begin with.
Anyway, I agree with you that copyright is not an innate human right, but I think that it’s a bit of a straw-man position. I’ve never come across anyone holding that belief.
This is exactly my view. I think the entire enterprise of “intellectual property” is completely unethical.
Also, I object to your use of the term “society.” It is governments—and only governments—that enforce monopoly privileges to creations of the intellect. Maybe you want to make the argument that governments and society are one in the same, but I certainly don’t see it that way, and I think it muddies the thinking on this too much. (Using the term “society” is also a convenient phrasing, because it’s essentially impossible to disagree with. “What, you mean disagree with society itself? You’re fucking bonkers.”)
Anyway, I agree with you that copyright is not an innate human right, but I think that it’s a bit of a straw-man position. I’ve never come across anyone holding that belief.
I used the words straight from the OP. I grant that “right” has many different interpretations, but I don’t know of any interpretation that I agree with. Is the OP stating a fact about the world? Fine, and I disagree with how our legal system is setup that way. Is the OP being prescriptive? Fine, and I think it’s unethical.
I think the entire enterprise of “intellectual property” is completely unethical.
Thanks for clearing that up; I disagree but respect your view.
For the record, I don’t believe that society and governments are one and the same, only that governments that act against the interest of the people (as they are doing now with copyright law and many other things) lack legitimacy. But I haven’t done enough research on the situation in the 1700s to say whether copyright was justified from the start; I am inclined to think that it was the right choice back then when publishers fulfilled a crucial role in society, but I could be oversimplifying things.
I more or less agree with the thesis, as I understand it: everyone has to pay, whether with code or with money. And that’s how I view the GPL too. It’s fair to everyone. Give and take back. That’s how Linus views the GPL too, as a way to ensure contributions keep flowing back.
I don’t consider free software a charity, but a bargain with society. I am nice to you, and in return, you must be nice to everyone else (usually, including me).
However, there is a marked difference between the GPL and the AGPL–I think that the former is a contract basically to the effect of “Hey, this is awesome and provided for free, and if you find it similarly awesome you need to contribute your fixes back to it”, in effect ensuring maintenance of the project. AGPL is not quite as focused on maintaining a single thing.
It’s difficult to write an exact analog, but for a webapp, I find something like the AGPL necessary if you really want a GPL-like effect, because nobody is distributing binaries that would trigger the normal GPL mechanisms. If Microsoft includes your code in the next Office, the GPL pretty well covers what they need to do in return. But if they include your code in the next Office365, the GPL no longer serves its intended purpose: they can continue to make private improvements to this “in-house” version, even though it’s a public app, because the backend to the app isn’t considered to have been “distributed” in a sense that would trigger the GPL. The AGPL defines making it available as a public webapp to count as distribution, just like shipping a native-code installer would.
But it’s effectively unenforcable, because the usage of such code can be made undetectable.
The GPL made sense for protecting rights back when users cared about what ran on their own machines and had a notion of sovereignty: by using centralized web apps instead of decentralized federated services, they have abdicated such rights.
In short, the AGPL is trying to protect the freedoms of people who have clamored for subjugation.
Now, now. We weren’t the ones who clamoured for web apps anymore than we were the ones who clamoured for an open letter to hobbyists. The GPL and AGPL are addressing the concerns of their times. Times have changed, so have our legal defenses.
While I like the idea of the copyleft licenses, I am unsure whether the GPL on average encourages contributions to a project or not.
It might be a funny dynamic, that the backcontribution rate to non-GPL projects is indeed lower that the one to GPL projects, yet the increased adoption probability for non-GPL projects is so much higher that BSD/MIT/Apache projects still get more backcontribution.
Now, aside from the GPL-debate, I see that this author has a huge ego. He wants to bring those people who develop software purely for the common cause to the GPL by nagging them that companies can “steal” their code without credit.
I license my works under the ISC in the hope that it may be useful to others. In general, the best software really comes out when you solve your own problems with it, so how much would you care that you made someone’s day who was able to put your software to good use? The author is a typical case of MOP.
I don’t think the reason why he doesn’t get a job is because some boss is not enlightened about what software his business runs. In most cases, bosses don’t care, they just want things to work.
The real reason is that startups don’t really look for specialists like Zed Shaw. They need people that are web developers, web designers, graphic designers and, depending on the needs, an engineer for the product itself.
Also, startups won’t care about the license anyway, as they offer mostly cloud services. Nobody can stop them from doing that, the GPL is not powerful enough.
On the other hand, people are forbidden from using GPL software when they want to publish their work under a more permissive license. They are forced to license their project under the GPL as well.
With GPLv3 it is even worse: Then you even have to think about the libraries you are linking to. If any library is licensed under the GPLv3, then your package maintainer can theoretically get in trouble.
The most fundamental problem of the GPLv3 is its complexity and negativity. The GPLv3 is all about restrictions and the culture of freedom it is supposed to promote is more like a monoculture and bad interaction between certain kinds of free software. In which way is an ISC-licensed project “inferior” to the GPL-licensed version? In my opinion, the former is much more what free software is about, whereas the GPL theoretically solves problems while interacting with the corporate sector.
On the flip side of this, one should consider that it’s entirely possible that Mongrel (and I assume in his article he’s actually talking about Mongrel2 which is BSD) may not have been anywhere near as successful had it been GPL instead of Ruby (version 1) or BSD (version 2).
Quite possible. People are always giving free software advocates a hard time for making decisions from ideology rather than technical superiority, so it would have been a delicious irony if the tables were turned and we had all these startups using sub-par HTTP servers due to their fear to engage with the terms of free software.
Note: it looks like this was written in 2009.
As another multi-decade OSS developer, I agree 100%. All of my stuff written in the last 5 years is L/GPL. Want a non-GNU license? Pay for it.
Did using AGPL benefit Zed with job offers and fame and preventing companies from being sold without disclosing they were using Zed’s software, or practically anything else he wrote about?
I doubt it’s correlated, but I’d say Zed’s fame has definitely waned since the switch.
Zed’s “waning fade” is pretty much self-chosen in my reading. He doesn’t give any talks anymore and seems pretty happy with the current way things are.
https://github.com/zedshaw/lamson/issues/34
http://imgur.com/8DYkXQP
Lamson project is kill, I think.
Mongrel 2 still seems okay, but Zed’s the only member of the Mongrel 2 org.
This is Zed at his best. Recent experiences have led me to a similar position as Zed’s, albeit for different reasons. BSD/MIT is typical in the Haskell community, but I’m not sure I’ll keep licensing my future OSS work like that.
I will keep existing projects on their current licenses.
this post and the ensuing discussion across the blogosphere have made me a proponent of the gpl for anything that could have commercial value downstream.
I arrived at this position because of people demanding our Haskell book be free & creative commons despite:
My coauthor being a not-wealthy homeschooling mother of two who works on the book in her spare time
My spending about half of the book’s development only taking light contract work or no work at all, burning my own savings so that I could focus on it
People’s attitude that they are entitled to our labor despite most of them making more money than myself or Julie has made us very hesitant to release the book as creative commons.
This and other things has me leaning towards keeping future OSS projects I start GPL. I won’t harass anyone else I work with into doing it, because quite frankly I think every programmer has had that experience and it’s infuriating. (cf. what I wrote above about creative commons)
inb4randos: I am not open to, nor am I interested in, debating the merits/demerits of our book being made CC. We have discussed the issue at length with serious & committed proponents of the commons and they admitted All Rights Reserved probably makes more sense for our needs.
agreed, cc books are awesome, but that should be each author’s individual and unpressured choice. it’s a shame some people don’t feel any cognitive dissonance in demanding someone else give their work away just because they would prefer that.
I’m really glad that you’re charging for the book, and I really hope it’ll be worthwhile for you both financially. It means that it’s fair for the reasons you mentioned, but also the readers end up with a better product (unlike a lot of the half-arsed free books on many topics out there).
People just have a weird sense of entitlement, and zero understanding of how much effort goes into a book.
Interestingly, for my book, I’ve only had one silly price demand that I recall (“it shouldn’t be any more than $5” – umm, why?).
This and other things has me leaning towards keeping future OSS projects I start GPL.
The problem is that a lot of software is cloud-hosted these days. The GPL does not bring you much there, because the software is not distributed. The AGPL addresses this problem, but is unfortunately incompatible with some other FLOSS licenses.
I used to use the GPL for the reasons stated, but the interesting thing is that a company actually contacted me to purchase (and paid for) a license for one of my Apache 2 Licensed projects. This project had a NOTICE file and the Apache License requires that its contents are displayed in the application or documentation. They basically paid me for not having to put this notice in their software.
Here’s my counterpoint to his argument: I don’t use GPL licensing on my open source software because it imposes restrictions on the use of my software that I don’t want to impose on my users. I write software because I want people to use it, not because I want to make it difficult to use.
In the end it comes down to your motivations I guess. The author laments, “I Don’t Want To Be Ignored Again”. If your main motivation in writing open source software is fame I guess that’s an important consideration for you, but I don’t think I’m alone in not giving a toss about fame and not feeling that writing software is all about having your ego stroked.
I write open source software because I like writing it and because I want to use it myself. Either way I’m happy to share with others. If I GPL my work it sets the legal bar too high for most commercial organizations so they just won’t use it. It’s no skin off my nose for them to use it so why not just let them by going BSD instead?
Absolutely. I’ve made the same argument you have; I open-source stuff in the hope that it will help people, and also because I find it freeing to give up ownership.
I’ve previously seen a lot of argument for the GPL on the basis that it helps people more than BSD does, which… is a legitimate difference of opinion, I suppose.
What is new to me about this author’s argument is that it’s a choice to use the GPL, with the motivation of wanting recognition! I disagree with it, but I still find it new and interesting. I’m highly sympathetic to concerns about exploitation, which were also discussed here at Throwing in the towel. Seeing those as aligned with the GPL is fascinating.
I agree with the theoretical arguments that GPL would help more people. In practice it hasn’t worked out that way - when I worked on GPL code no-one used it, when I work on BSD[1] code companies use it and some of them contribute back. If there were other organizations producing code then maybe the GPL model could work, but even if for-profit companies are only contributing 1/10 or 1/100 of what they write, that’s still more code than the rest of the world produces at all.
I don’t think GPL will give you fame either, at least in the long term. All it means is that your code will be replaced. libedit displaces readline, LLVM displaces GCC - when you release a good BSD library there’s no reason for anyone to ever replace it, whereas even if you release an excellent GPL library, if it does something useful then in the long term there will be enough people who need a BSD version that eventually one will be written and people will switch to that.
(I also think Shaw’s ego is vastly out of proportion with his ability, and the fact that he felt he was getting less fame than he deserved doesn’t at all convince me that this was objectively so).
[1] Actually Apache license but same difference
The point of the fame isn’t the fame; it’s the work that the fame gets you. Or at least, that’s how it’s positioned in the article.
From an ideological perspective, I actually disagree with the OP’s initial premise, which is that it’s the “author’s right to license their work.” I don’t think anyone has some innate right to appeal to monopoly protection over creations of the intellect (of which, code absolutely falls under).
With that said, I’m unlikely to be the type of person the OP had in mind when writing that first section. In particular, despite my ideological position, I absolutely agree with the OP that I would be a gigantic jerk if I ran around telling others that they didn’t have a right to choose the license they did. It’s just not a productive expression of my views on intellectual property.
And on the scale of damage that intellectual property laws do, the GPL is barely a blip.
Totally agree. This is why I choose BSD/MIT. The users of my code in my opinion are programmers. The idea that my choice could make people have to reimplement things due to a ideology just strikes me as being a jerk.
I know people say “but what if a company makes a ton of money off your code”. So? Who cares.
I honestly purposefully avoid GPL software for BSD/MIT, mostly because its really annoying to have to avoid using software due to ideology that I fundamentally disagree with. I’ve always been a BSD user at heart (and home), and I just don’t think GPL is worth the loss of freedoms for others. Basically I’m empathizing with other programmers that are just trying to get things done. I’m in the same boat, and think a more free license for programmers trumps a more free license for users.
Probably not an opinion a lot of people share but honestly, I think more people need to say no to GPL publically. LLVM is in my opinion the best example of why GPL is not the best license choice. It seemed to finally light a fire under gcc to improve.
Choose what license you want, but if there is a BSD/MIT/etc.. alternative, I’ll always pick that and work to fix it to beat the GPL version.
I didn’t see anywhere in the article where it claimed that this was an innate human right. It’s a right that we as a society have chosen to grant to each other, theoretically based on consensus that it’s better than the alternative.
I don’t subscribe to the belief that “rights are given theoretically.” And I don’t know how you inferred how the OP was using the word. (I disagree with it in every sense of the word that I’m aware of, so you’re barking up the wrong tree with me personally.)
I didn’t mean that the rights are given theoretically; I meant that the consensus behind copyright is theoretical. Due to the way it’s been abused by media empires, many people consider it to have outlived its usefulness and to be more trouble than it’s worth. But the fact is it remains the law.
Just to be clear, do you disagree that society has any right to enforce this monopoly to begin with? As in, the Statute of Anne was a mistake and never should have been passed because it overstepped the authority of the crown? It’s one thing to say that copyright legislation has gone too far, and quite another to say that it never had any legitimacy to begin with.
Anyway, I agree with you that copyright is not an innate human right, but I think that it’s a bit of a straw-man position. I’ve never come across anyone holding that belief.
This is exactly my view. I think the entire enterprise of “intellectual property” is completely unethical.
Also, I object to your use of the term “society.” It is governments—and only governments—that enforce monopoly privileges to creations of the intellect. Maybe you want to make the argument that governments and society are one in the same, but I certainly don’t see it that way, and I think it muddies the thinking on this too much. (Using the term “society” is also a convenient phrasing, because it’s essentially impossible to disagree with. “What, you mean disagree with society itself? You’re fucking bonkers.”)
I used the words straight from the OP. I grant that “right” has many different interpretations, but I don’t know of any interpretation that I agree with. Is the OP stating a fact about the world? Fine, and I disagree with how our legal system is setup that way. Is the OP being prescriptive? Fine, and I think it’s unethical.
Thanks for clearing that up; I disagree but respect your view.
For the record, I don’t believe that society and governments are one and the same, only that governments that act against the interest of the people (as they are doing now with copyright law and many other things) lack legitimacy. But I haven’t done enough research on the situation in the 1700s to say whether copyright was justified from the start; I am inclined to think that it was the right choice back then when publishers fulfilled a crucial role in society, but I could be oversimplifying things.
I more or less agree with the thesis, as I understand it: everyone has to pay, whether with code or with money. And that’s how I view the GPL too. It’s fair to everyone. Give and take back. That’s how Linus views the GPL too, as a way to ensure contributions keep flowing back.
I don’t consider free software a charity, but a bargain with society. I am nice to you, and in return, you must be nice to everyone else (usually, including me).
I rather agree with that social contract.
However, there is a marked difference between the GPL and the AGPL–I think that the former is a contract basically to the effect of “Hey, this is awesome and provided for free, and if you find it similarly awesome you need to contribute your fixes back to it”, in effect ensuring maintenance of the project. AGPL is not quite as focused on maintaining a single thing.
It’s difficult to write an exact analog, but for a webapp, I find something like the AGPL necessary if you really want a GPL-like effect, because nobody is distributing binaries that would trigger the normal GPL mechanisms. If Microsoft includes your code in the next Office, the GPL pretty well covers what they need to do in return. But if they include your code in the next Office365, the GPL no longer serves its intended purpose: they can continue to make private improvements to this “in-house” version, even though it’s a public app, because the backend to the app isn’t considered to have been “distributed” in a sense that would trigger the GPL. The AGPL defines making it available as a public webapp to count as distribution, just like shipping a native-code installer would.
But it’s effectively unenforcable, because the usage of such code can be made undetectable.
The GPL made sense for protecting rights back when users cared about what ran on their own machines and had a notion of sovereignty: by using centralized web apps instead of decentralized federated services, they have abdicated such rights.
In short, the AGPL is trying to protect the freedoms of people who have clamored for subjugation.
Now, now. We weren’t the ones who clamoured for web apps anymore than we were the ones who clamoured for an open letter to hobbyists. The GPL and AGPL are addressing the concerns of their times. Times have changed, so have our legal defenses.
While I like the idea of the copyleft licenses, I am unsure whether the GPL on average encourages contributions to a project or not.
It might be a funny dynamic, that the backcontribution rate to non-GPL projects is indeed lower that the one to GPL projects, yet the increased adoption probability for non-GPL projects is so much higher that BSD/MIT/Apache projects still get more backcontribution.
For a side-by-side read I recommend Armin Ronacher’s thoughts on this http://lucumr.pocoo.org/2009/2/12/are-you-sure-you-want-to-use-gpl/
Now, aside from the GPL-debate, I see that this author has a huge ego. He wants to bring those people who develop software purely for the common cause to the GPL by nagging them that companies can “steal” their code without credit.
I license my works under the ISC in the hope that it may be useful to others. In general, the best software really comes out when you solve your own problems with it, so how much would you care that you made someone’s day who was able to put your software to good use? The author is a typical case of MOP.
I don’t think the reason why he doesn’t get a job is because some boss is not enlightened about what software his business runs. In most cases, bosses don’t care, they just want things to work. The real reason is that startups don’t really look for specialists like Zed Shaw. They need people that are web developers, web designers, graphic designers and, depending on the needs, an engineer for the product itself. Also, startups won’t care about the license anyway, as they offer mostly cloud services. Nobody can stop them from doing that, the GPL is not powerful enough.
On the other hand, people are forbidden from using GPL software when they want to publish their work under a more permissive license. They are forced to license their project under the GPL as well. With GPLv3 it is even worse: Then you even have to think about the libraries you are linking to. If any library is licensed under the GPLv3, then your package maintainer can theoretically get in trouble.
The most fundamental problem of the GPLv3 is its complexity and negativity. The GPLv3 is all about restrictions and the culture of freedom it is supposed to promote is more like a monoculture and bad interaction between certain kinds of free software. In which way is an ISC-licensed project “inferior” to the GPL-licensed version? In my opinion, the former is much more what free software is about, whereas the GPL theoretically solves problems while interacting with the corporate sector.
On the flip side of this, one should consider that it’s entirely possible that Mongrel (and I assume in his article he’s actually talking about Mongrel2 which is BSD) may not have been anywhere near as successful had it been GPL instead of Ruby (version 1) or BSD (version 2).
Quite possible. People are always giving free software advocates a hard time for making decisions from ideology rather than technical superiority, so it would have been a delicious irony if the tables were turned and we had all these startups using sub-par HTTP servers due to their fear to engage with the terms of free software.
Thoughtful piece.