If the author is required to publish the source code, guess what his second to market competitors will have to do as well? Free software is not an ideological exercise. It has its benefits. The author missed out on a huge opportunity.
The author is greedy. GPL v3 successfully cut off the parasite. Good riddance.
Now, he does make a good point though. We have to think strategically. Read gnu.org for details but here is the cinch: if there are commercial alternatives to what you are making, make your licensing liberal (arguably android because iOS existed and WinPoo 7 was apparently in the works). This encourages adoption of foss. However, when foss has the market cornered, don't let the big guys take your ball and run home. Use GPL v3.
The author is greedy. GPL v3 successfully cut off the parasite. Good riddance.
Now, this was a real soap box. Let me hop off it real quick.
"Because that one little "worm" - used simply to support a startup script, could have created a situation where we would have been required to open source our entire product."
Someone assist in my enlightenment: Under what sort of enforcement circumstances would that requirement to open-source a lot of code actually be likely? As opposed to some enforcement action where you remove the offending code.
Yeah. This company was about to distribute GNU grep, meaning that they would be obligated to provide source for that version of grep, and likewise for any modifications to grep they've made. Yawn.
When a clever person is so completely misinformed about a subject that's the very core of their business, I would rather call it FUD.
And I'd also like to point out they aren't in the business of creating value, as an open source product offers more value than a black box: their want to create costs.
I always find it suspicious when someone clever falls victim of - and further spreads - FUD that is beneficial to his/her business. Remember they sell proprietary (actually, re-licensed) software that's not under one of those "dangerous" and "viral" licenses.
It can't happen. However, the plaintiff may offer to forgo monetary damages in exchange for you open-sourcing. So it becomes pay money, or open-source.
> A compilation of a covered work with other separate and independent works, which are not by their nature extensions of the covered work, and which are not combined with it such as to form a larger program, in or on a volume of a storage or distribution medium, is called an “aggregate” if the compilation and its resulting copyright are not used to limit the access or legal rights of the compilation's users beyond what the individual works permit. Inclusion of a covered work in an aggregate does not cause this License to apply to the other parts of the aggregate. (GPLv3)
Using GNU grep, as GNU grep, and merely aggregating it with the whole, would seem, from the quote above, to not cause the whole to be GPLv3 covered.
The only slightly questionable part is the "combined with it such as to form a larger program" portion of GPLv3. Is using grep, to "grep", sufficient to say that grep is "combined ... to form a larger program"? If the answer is no, then yes, the author is simply misinformed.
Why should distributing GNU grep necessitate releasing the source of other things that you distribute? It only seems obvious to me that they would then have to distribute the source of GNU grep.
Are they linking to grep but grep doesn't have the linking exception?
You're correct. There is no circumstance under which he's compelled to open up his source code.
If the hardware system he's selling has some kind of crypto (like login passwords that you don't know!) that disallows modifications to the base system, GPLv3 can require you to give up enough control to let the user install their modified grep too. (That's still not giving away your source code, but it is more than just giving out the source code to grep.)
Good point, you couldn't include GPLv3'd code into something like a Tivo, but that is hardly an unintended side-effect of the license. Whining about that would be rather silly. ;)
I became an AGPL3 fan after reading many papers from the 50s->70s, before the rise of open source & the Unix sharing habits. I believe its a crying shame that all those wonderful tools and systems that I read about disappeared into the void and got hidden by closed-source approaches to development and licensing. I am fully aware that this limits certain avenues of commercialization, and I am fully OK with that. I believe that an AGPL3 world is a better world for computer science and technology.
From my perspective as an engineer in the enterprise, we value functionality very highly; if you sell us a GPL3-based hardware solution, we care that it works. We're not going to copy and resell it. We have our own products to care about and do excellent things with.
IMO, they disappeared exactly because they embraced the ideals you love. Of course people aren't going to work on extending and polishing these systems if they can't get paid to (at least not at the same magnitude as for the closed solutions). At least, that is what history seems to indicate so far, except in a relatively limited space where companies have both the technical expertise to modify the source AND the economic incentive to avoid paying for closed licenses.
Although that is not the situation this guy finds himself in. He seems to just be confused.
> By contrast, pipes, sockets and command-line arguments are communication mechanisms normally used between two separate programs. So when they are used for communication, the modules normally are separate programs.
- http://www.gnu.org/licenses/gpl-faq.html
For even greater emphasis: the FSF has a vested interest in promoting a maximally broad interpretation of the GPL. It's not legal advice, but when even their own FAQ says you're in the clear, you're likely to be pretty damn "safe."
The FSF wants the GPL to be a license that people will actually use, and within that constraint, they want it to be interpreted in a way that most strictly protects software freedom, as they see it. If they just wanted it to be as broad as possible, they would be shooting themselves in the feet; it's easy for ticked-off software creators to switch to a less restrictive license if they own the copyright to software under the GPL.
This, for example, is why the FSF always mentions explicitly that they do not consider merely using a tool like grep as part of a shell script to be a derived work. They have to be reasonable, or they lose their power.
Agree, but I think the GPL examples about communication mechanisms are pretty weird and sometimes absurd. Consider this scenario:
Someone takes a GIMP plug-in licensed under GPL and converts it to a Photoshop plug-in (which is a DLL and shares memory space with Photoshop).
According to GPL, Photoshop would now have to be open-sourced, but the person, who converted the plug-in does not have access to Photoshop source code.
What now? Goes GPL actually forbids the existence of this? What if the converted plug-in with the source code is published? Is that actually illegal? What if they only publish the adapted source code and not the binary? What if a third person compiles it and publishes the binary? Who would be the guilty party?
I've followed GPL discussions from the sidelines for sometime, and whenever someone comes up with a 'trick' scenario like this, the standard advice is "if you'd like to do that, call your lawyer".
(The legal definition of 'derived works' is based in case law and is more complex than what you'd find in a FSF FAQ.)
IANAL, but: Distributing a photoshop plugin containing GPL code is a violation of the GPL as it is worded. If that part of the GPL is enforcable (no precedent that I am aware of) then you would essentially be pirating the GPL code, and can be sued for monetary damages by the owner of the copyright for that code.
> Distributing a photoshop plugin containing GPL code is a violation of the GPL as it is worded.
That would only be true if the distributor failed to provide the source code to any interested party who asked for it. If he met that requirement, and if he included all the original copyright notices in the source along with his own, there's no violation.
There is no way on God's green Earth that you can say his product is a derived work of GNU grep because it is used in the way he describes. At the very least, they just have to provide the source for grep.
Yeah, that is my thought too. Maybe he thinks including a call to grep that the users shell would read/fork/exec is a type of linking? Seems like a stretch.
If you are building propriety products around lots of open-source components then being mindful of the GPL is just basic common sense. Acting like it is some hidden minefield is being disingenuous.
Seems most people didn't really read his story from beginning to end. Specifically the part at the beginning of the post: "The product itself is not delivered as a separate executable, but as a complete product. We don't permit our customers to crack it open..."
If he distributes his product as a monolithic and atomic item, wouldn't he be forced to release all parts of it if requested?
37 comments
[ 4.3 ms ] story [ 79.1 ms ] threadThe author is greedy. GPL v3 successfully cut off the parasite. Good riddance.
Now, he does make a good point though. We have to think strategically. Read gnu.org for details but here is the cinch: if there are commercial alternatives to what you are making, make your licensing liberal (arguably android because iOS existed and WinPoo 7 was apparently in the works). This encourages adoption of foss. However, when foss has the market cornered, don't let the big guys take your ball and run home. Use GPL v3.
The author is greedy. GPL v3 successfully cut off the parasite. Good riddance.
Now, this was a real soap box. Let me hop off it real quick.
Someone assist in my enlightenment: Under what sort of enforcement circumstances would that requirement to open-source a lot of code actually be likely? As opposed to some enforcement action where you remove the offending code.
And I'd also like to point out they aren't in the business of creating value, as an open source product offers more value than a black box: their want to create costs.
Using GNU grep, as GNU grep, and merely aggregating it with the whole, would seem, from the quote above, to not cause the whole to be GPLv3 covered.
The only slightly questionable part is the "combined with it such as to form a larger program" portion of GPLv3. Is using grep, to "grep", sufficient to say that grep is "combined ... to form a larger program"? If the answer is no, then yes, the author is simply misinformed.
Are they linking to grep but grep doesn't have the linking exception?
If the hardware system he's selling has some kind of crypto (like login passwords that you don't know!) that disallows modifications to the base system, GPLv3 can require you to give up enough control to let the user install their modified grep too. (That's still not giving away your source code, but it is more than just giving out the source code to grep.)
From my perspective as an engineer in the enterprise, we value functionality very highly; if you sell us a GPL3-based hardware solution, we care that it works. We're not going to copy and resell it. We have our own products to care about and do excellent things with.
Although that is not the situation this guy finds himself in. He seems to just be confused.
edit: As a matter of fact, part of why Unix was able to spread was the AT&T restrictions forcing it to be shared.
I've met many during my career.
> By contrast, pipes, sockets and command-line arguments are communication mechanisms normally used between two separate programs. So when they are used for communication, the modules normally are separate programs. - http://www.gnu.org/licenses/gpl-faq.html
This, for example, is why the FSF always mentions explicitly that they do not consider merely using a tool like grep as part of a shell script to be a derived work. They have to be reasonable, or they lose their power.
Someone takes a GIMP plug-in licensed under GPL and converts it to a Photoshop plug-in (which is a DLL and shares memory space with Photoshop).
According to GPL, Photoshop would now have to be open-sourced, but the person, who converted the plug-in does not have access to Photoshop source code.
What now? Goes GPL actually forbids the existence of this? What if the converted plug-in with the source code is published? Is that actually illegal? What if they only publish the adapted source code and not the binary? What if a third person compiles it and publishes the binary? Who would be the guilty party?
(The legal definition of 'derived works' is based in case law and is more complex than what you'd find in a FSF FAQ.)
That would only be true if the distributor failed to provide the source code to any interested party who asked for it. If he met that requirement, and if he included all the original copyright notices in the source along with his own, there's no violation.
Your sense of entitlement is mind boggling.
/Someone/ wrote the code you are complaining about and /their/ idea of /value/ is that it should be open sourced under a "viral" license.
If he distributes his product as a monolithic and atomic item, wouldn't he be forced to release all parts of it if requested?