Weekly Edition
Return to the Announcements page
| |||||||||||||
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement
Over on his blog, Harald Welte comments on GPL enforcement in light of the Busybox/Toybox controversy. "In any kind of GPL enforcement, you of course not only want the complete corresponding source code to one program, but to all of the GPL/LGPL/AGPL or otherwise copyleft licensed programs contained in the product. We at gpl-violations.org have always been requesting the complete corresponding source code to all GPL licensed software during our communication with the infringing companies. This request was typically honored by everyone, without the need to apply any pressure onto it. After all, releasing only one bit of code causes the risk to get sued by somebody else who owns the other not-yet-compliant part of the code. [...] Now there have been rumors that SFC was not only requesting non-Busybox source code, but also making it a condition for the explicit re-instatement of the license on Busybox. Whether or not there was such a hard condition is subject to debate and there are different opinions on it. For those in the field of FOSS licensing, it has always known that there are different lines of thought with regard to the requirement to explicit reinstatement. We in Germany generally think that it is not required at all, and the existing preliminary injunctions at least implicitly acknowledge that as they enjoin companies from distributing a product as long as it is not in compliance with the license. In other (particularly the U.S.), it is generally assumed that explicit reinstatement is required."
(Log in to post comments)
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 9:15 UTC (Sat) by BrucePerens (guest, #2510) [Link] It's pretty clear that reinstatement has required that you comply with all Free Software licenses for software in the system. This is my personal experience in representing companies that deal with SFC.I don't see why Harald thinks this isn't a good idea. Why should SFC or anyone else condone or abet a clear violation of the license on other Free Software software while at the same time issuing explicit forgiveness for the infringement of Free Software that is only different in that they directly represent the copyright holders? Which one of the companies that are infringing in that way would grant such lenient terms to any party infringing their code?
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 9:54 UTC (Sat) by neilm (subscriber, #28422) [Link]
Bruce,
I don't think that's what Harald said at all. He explicitly stated that gpl-violations.org ask for full source for all components. What I read the article as saying was about the requirement for an explicit relicencing, and the need for that or not.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 14:20 UTC (Sat) by jejb (subscriber, #6654) [Link]
> I don't see why Harald thinks this isn't a good idea. Why should SFC or anyone else condone or abet a clear violation of the license on other Free Software software while at the same time issuing explicit forgiveness for the infringement of Free Software that is only different in that they directly represent the copyright holders?
It's not about condoning violations, it's about scope of enforcement. You have the moral right to dictate compliance terms as far as your copyrights run. If I, as a kernel copyright holder, start trying to dictate what compliance means in, say, busybox, without consulting the busybox copyright holders, I've exceeded my moral authority.
Consider the situation where busybox has already settled privately with a vendor; say they lost source code for shipping product A and you agreed not to insist on a recall in exchange for discontinuation and shipping a newly compliant product B. I find a simple kernel violation in product B and enforce for it; the company goes "oops" and fixes the compliance issue in product B but I insist I won't reinstate their GPL rights until they bring busybox into compliance on product A, which will necessitate a recall. My actions have just destroyed your credibility to negotiate on behalf of busybox because no future vendor will believe you when you try to negotiate terms for a settlement because someone somewhere with a copyright in any GPL component they ship can upend the terms you agreed.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 14:44 UTC (Sat) by lutchann (subscriber, #8872) [Link]
Obviously the busybox copyright holders have no authority to negotiate on behalf of all of the copyright holders of all Free Software that may be used on the vendor's device, so it would be naive for a vendor to feel "betrayed" if a settlement negotiated over busybox failed to satisfy the copyright holders of the other Free components.
I don't see how this is any different than if, say, Autodesk pursued a customer for license violations and their settlement terms included regular license audits by the BSA of all proprietary software used by the customer.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 15:11 UTC (Sat) by jejb (subscriber, #6654) [Link]
> Obviously the busybox copyright holders have no authority to negotiate on behalf of all of the copyright holders of all Free Software that may be used on the vendor's device
Well, no, that's the point ... if broad interpretation of the GPL termination clause holds, legally they have every right.
My argument is that even if the legal right exists, copyright holders should adhere to the moral principle of only dictating compliance terms for code over which they hold copyright to avoid conflicting with each other over compliance.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 15:18 UTC (Sat) by mjg59 (subscriber, #23239) [Link]
Things become more complicated once we raise morality. Are the kernel copyright holders guilty of a moral failing by talking about the benefits of the GPL without doing anything to make sure people who buy Linux devices actually gain those benefits? We could even make an argument that there's a moral imperative to make sure users get all the rights they should have, and therefore that SFC-style enforcement is the right thing to do. Does bringing philosophy into the argument really bring any clarity?
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 13, 2012 2:40 UTC (Mon) by jzbiciak (✭ supporter ✭, #5246) [Link] Ok, there's morality and then there's the legal copyright concept of moral rights. I think they're getting confused in this thread. The right of an author to dictate the terms of how their work gets published or distributed (regardless of whether the author has signed those rights over to someone else, in many cases), is that authors moral rights in the work. If we go upthread to jejb's comment: You have the moral right to dictate compliance terms as far as your copyrights run. If I, as a kernel copyright holder, start trying to dictate what compliance means in, say, busybox, without consulting the busybox copyright holders, I've exceeded my moral authority. I'm pretty sure jejb was referring to the legal concept I linked. What I'm not certain is whether jejb's entirely correct. If I, as an author, feel that the integrity of my work is violated when someone aggregates it with another work but violates that other work's license, am I exceeding my moral authority when I demand they stop distributing mine? Here in the US, the concept of moral rights is rather foreign. We don't really have a strong tradition of protecting moral rights here. But other countries do. In any case, the term "moral rights" is unfortunately confusing, since it's not exactly about morality in the more general case. It's a specific legal construct.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 15:06 UTC (Sat) by mjg59 (subscriber, #23239) [Link]
The situation you describe seems objectionable, but it's not an argument against SFC-style enforcement. If you (as a kernel copyright holder) settle privately in a way that results in a vendor not coming into absolute compliance, there's nothing stopping me (as a kernel copyright holder) objecting to that settlement and enforcing myself. The objectionable act here is enforcing against existing settlements, which applies just as much whether you're engaging in SFC-style enforcement or directly within the scope of the project you hold copyright on.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 15:31 UTC (Sat) by jejb (subscriber, #6654) [Link]
> The situation you describe seems objectionable, but it's not an argument against SFC-style enforcement.
I haven't said anything about the SFC. I've just illustrated problems with blanket termination right enforcement if you dictate terms for compliance in projects over which you don't hold the copyright.
The solution seems fairly simple: Get copyright holders in the other project to join your enforcement action and refuse to restore the licence until both projects agree compliance has been met in their respective parts.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 15:35 UTC (Sat) by mjg59 (subscriber, #23239) [Link]
But the same problem applies even if you only dictate terms for projects over which you do hold the copyright. You have no way of knowing whether your settlement terms are going to satisfy all kernel copyright holders.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 15:58 UTC (Sat) by armijn (subscriber, #3653) [Link]
Which would not make sense anyway, since a binary is not likely to have source code of all kernel developers included anyway :-)
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 15, 2012 9:08 UTC (Wed) by branden (subscriber, #7029) [Link]
Your response puzzles me.
"All"? No.
"More than one"? Almost certainly.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 12, 2012 15:57 UTC (Sun) by dwmw2 (subscriber, #2063) [Link] "I haven't said anything about the SFC. I've just illustrated problems with blanket termination right enforcement if you dictate terms for compliance in projects over which you don't hold the copyright."Wow, where do I start? Firstly, you're conflating the issues. Do not confuse the copyright enforcement on the enforcer's project (e.g. Busybox) with the licence of another project like the kernel. If a company has violated the licence, and has thus lost their permission to use a certain piece of software, then the copyright-holder can make arbitrary requests of the criminal in return for granting them a new licence. It's just the same as the arbitrary requests that you can make in the first place, in return for a licence. Don't get confused about copyright on other things involved in that request. If I release something as postcardware, and grant a licence in return for receiving a postcard, it has nothing to do with the copyright licensing of that postcard. If I grant a licence in return for money, it has nothing to do with the copyright of the banknotes. If I make you promise never to assert patent rights against any of my software, that also has nothing to do with your "moral rights" to those patents. You're being offered a trade; you can take it or leave it. I could give you permission to use my software iff you name your first-born child "Shrek" and agree to release every piece of software you ever write in future under the GPL. You get to choose whether to accept those requirements, or not. Complaining that I have no moral right to assert copyright over your future work is just a complete non-sequitur. And if I demand a postcard, it's also bizarre to complain that I have "destroyed the credibility" of the person who sold you that postcard and told you that you could keep it. It's a completely unrelated transaction. You are making no sense with that line of argument, or the claim that it is somehow "inconsistent". Going back to the licensing... in the case of licensing something that is already being used illegally, the violator is over a barrel to a certain extent. That's true of proprietary software too; if you're found with bootlegged copies of Windows, don't expect to be able to negotiate a discount when they catch up with you!
But still, the violator has the option of agreeing to the conditions, or walking away and paying the standard rate that the court imposes for their crime. If the out-of-court settlement conditions are too onerous, the criminal always has the option of not accepting them. Besides, the conditions of the SFC aren't onerous. They only ask you to come into compliance with the licence of other software that you're shipping, which is hardly an immoral request. You raise the spectre of "inconsistent" enforcement. But your example is extremely far-fetched; I can't see the SFC demanding a recall of an EOL'd product, if the company has already superseded it and come into compliance for the new version. Especially if there was a previous settlement with that company. That suggestion seems somewhat disingenuous and I am disappointed that it has been raised in this context; it seems designed to show the SFC in a bad light by raising concerns that are completely unrealistic in the context of what we were actually talking about. In any more realistic situation where we're actually likely to see inconsistent enforcement (e.g. perhaps if the first generation of the product was still being shipped), the real fault would be with the party who was being overly lax, and unilaterally deciding to let the violator get away with their crime. They should have known that they'd be overruled. As mentioned above, if one of the parties is actually making demands as part of an unrelated copyright settlement (or a sexual assault settlement, or if they're simply paying the company for what they want), that's not inconsistent at all. But still, inconsistency can happen if both parties are enforcing their own copyright on the kernel, but to a different standard. In that case it's a valid concern, albeit somewhat out of place in this discussion thread. And the simple answer is that you should not settle for something less than you should reasonably be demanding. Which, by the way, is one of the reasons that SFC gives for asking violators to come into compliance with the licences of all software they're using: "...if they come into compliance due to my efforts, what stops others from coming to complain, in sequence, and wasting their time?"
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 13, 2012 0:10 UTC (Mon) by jejb (subscriber, #6654) [Link]
> I could give you permission to use my software iff you name your first-born child "Shrek" and agree to release every piece of software you ever write in future under the GPL. You get to choose whether to accept those requirements, or not. Complaining that I have no moral right to assert copyright over your future work is just a complete non-sequitur.
You could *ask* for all of that, but you won't get it. When it gets to court, a court will apply reasonableness standards and unreasonable positions like the above will get completely ignored in the settlement (probably along with all your arguments about what the GPL actually means, so the judge will then rely on the opposition for interpretations of the GPL).
The point is that if you go into this requiring something unreasonable, the opposition will go "goody, we get the court's sympathy in settlement" and we end up with a judgment compromising what most people consider to be the fundamentals of the GPL. If you go in with a reasonable and justifiable position (preferably just leaning towards the unreasonable side of reasonableness, giving room for judicious compromise), the opposition, who are as you say criminals and violators, won't get the court's sympathy and we end up with a judgment affirming some of the fundamentals of the GPL.
The moral dimension enters the equation because if you suffer an adverse judgment because you took an unreasonable position, you damage the GPL and the ecosystem it supports for all of us.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 13, 2012 1:42 UTC (Mon) by dwmw2 (subscriber, #2063) [Link] "You could *ask* for all of that, but you won't get it. When it gets to court, a court will apply reasonableness standards and unreasonable positions like the above will get completely ignored in the settlement (probably along with all your arguments about what the GPL actually means, so the judge will then rely on the opposition for interpretations of the GPL)."No, it has nothing to do with "what the GPL means". I think you've completely missed the point of the example I was giving. It was talking about a situation where the software in question is not already licensed to you under the GPL or any other licence. Like the postcard example, it was an example to demonstrate the silliness of the "your licence cannot affect other people's copyright" argument.
"The moral dimension enters the equation because if you suffer an adverse judgment because you took an unreasonable position, you damage the GPL and the ecosystem it supports for all of us."Wow. That's taking the example drastically out of context, isn't it? It was an example; nobody's ever suggested that anyone (especially the SFC) would actually take such a position in a GPL enforcement! So to talk about "damag[ing] the GPL and the ecosystem it supports for all of us" might come across as a fairly clumsy attempt at FUD.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 13, 2012 2:23 UTC (Mon) by jake (editor, #205) [Link]
> You could *ask* for all of that, but you won't get it. When it
> gets to court, a court will apply reasonableness standards and > unreasonable positions like the above will get completely ignored > in the settlement
So, two parties settle a case under terms they both find "reasonable" and the judge will step in and remove terms from the settlement for unreasonableness? It would seem that a copyright infringer, when faced with multiple millions of dollars in potential damages, might be willing to agree to much to avoid paying the freight ... and the judge will just decide to unilaterally remove settlement terms? Why would the side whose terms were removed agree to the new settlement? Or do they get no second chance?
I'm puzzled here ...
jake
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 13, 2012 2:38 UTC (Mon) by dlang (✭ supporter ✭, #313) [Link]
if they settle then the terms probably won't be considered too unreasonable, but if you ask for terms that are too unreasonable, you run the potential for the target not settling and choosing to fight instead.
At that point the terms go in front of a judge, and as we are seeing in the Oracle - Google litigation, the judge may tell you that your terms are completely unreasonable and throw them out (if you are lucky like Oracle is, the judge may tell you to try again)
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 13, 2012 3:27 UTC (Mon) by BrucePerens (guest, #2510) [Link] if they settle then the terms probably won't be considered too unreasonable, but if you ask for terms that are too unreasonable, you run the potential for the target not settling and choosing to fight instead. The only party that really tried to fight was BestBuy, and they ended up accepting a settlement with SFC rather than continuing in court, after complaining about just these terms as "unreasonable" in their open court documents (they're available through Recap, if you care to look them up). What makes the SFC terms reasonable is that they don't come with financial damages, penalties, import prohibitions, and whatever terms the court would routinely impose upon the infringer to cure the infringement.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 19:41 UTC (Sat) by rfontana (subscriber, #52677) [Link]
> You have the moral right to dictate compliance terms as far as your
> copyrights run. If I, as a kernel copyright holder, start trying to > dictate what compliance means in, say, busybox, without consulting the > busybox copyright holders, I've exceeded my moral authority.
I would rephrase this as "You have the moral right to dictate
So, using the example in your second sentence there, I think it is not
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 20:50 UTC (Sat) by BrucePerens (guest, #2510) [Link] Sorry about having problems parsing Harald's statement. I've worked with a bunch of lawyers in the States, not Germany, who feel that explicit reinstatement is necessary and have directed me to get it for their customers.My feeling on the issue is that we all have to hang together or we will each hang separately. Developers should do what they can to make sure that the terms of all Free Software licenses are honored. In my work, I've also made sure that licenses of proprietary software are honored, which is simply performing due diligence for the customer.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 16:48 UTC (Sat) by arjan (subscriber, #36785) [Link]
Read the blog carefully.
He says that he always *ASKS* for all the sources to all GPL components, and that nobody had issues with that and complied.
Before going up in arms about moral issues: *EVERYONE* is allowed to ask for these sources, copyright holder or not! That's one of the things the GPL gives to everyone (who has received copy of the binaries bla bla), not just the copyright holder. He didn't sue for that, he didn't get aggressive about it, he just asked the question anyone is allowed to ask, and that the vendor is supposed to honor as per the license.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 17:01 UTC (Sat) by pboddie (subscriber, #50784) [Link]
Technically, it's everyone who has received the software who is allowed to ask for the sources, but your point still stands.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 17:04 UTC (Sat) by mjg59 (subscriber, #23239) [Link]
That's not what 3(b) says, and if you didn't distribute the source at the same time as the binaries it seems that you're implicitly distributing under 3(b).
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 11, 2012 18:52 UTC (Sat) by pboddie (subscriber, #50784) [Link]
Certainly, in the absence of accompanying source code, one might argue that it's a choice between immediately being in violation of the licence or claiming that the intent was really to offer source code via 3(b), even though there was never actually a written offer. I suppose this gives the distributor the chance to comply with the licence.
Of course you actually need someone who has received the software to confirm that neither accompanying source code nor a written offer was received. At which point, anyone could legitimately leverage the supposed intent - the path of least incrimination, as it were - to request the sources for any GPL-licensed software where the licensing terms have not been observed. But it needs that confirmation for the offer to be good for everyone. Otherwise, it's too easy for the distributor to give everyone the brush-off.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 12, 2012 2:53 UTC (Sun) by slashdot (guest, #22014) [Link]
IMHO a more interesting issue is: what's the point of enforcing the GPL on widely available unmodified software?
I'd expect that most/all copies of Busybox are unmodified, and no sane person would get the busybox source from their router vendor instead of the official Busybox website.
If the software has been modified things are entirely different, but otherwise such enforcement provides no benefit to users, except possibly from additional funding to open source coming from the settlement money.
And intentionally extracting money with frivolous complaints (not having provided unmodified source code that nobody wants) for the purpose of funding yourself isn't necessarily great PR.
Also, if that's really their purpose, maybe they should try to get more money of it, and perhaps help other copyright holders participate in the squeezing.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 12, 2012 3:22 UTC (Sun) by dlang (✭ supporter ✭, #313) [Link]
According to Rob Landley, in the year or so that he was participating in the Busybox lawsuits, what he found was that they took the busybox code, backported some fixed from later versions, hard-coded some things (instead of using config options), and added debug printfs, but nothing more than that.
so the code was different, but not in any interesting way.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 12, 2012 13:22 UTC (Sun) by jelmer (subscriber, #40812) [Link]
It might not necessarily be interesting for the upstream project.
For users who own one of these devices it is still useful to be able to modify the specific version of busybox that their device is running.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 12, 2012 21:04 UTC (Sun) by Kluge (guest, #2881) [Link]
"It might not necessarily be interesting for the upstream project.
For users who own one of these devices it is still useful to be able to modify the specific version of busybox that their device is running."
Indeed. And it might be worthwhile for the user to know the exact version, the compiler options used, etc., to determine what security holes might be present.
Landley and other "open source" folk seem to be primarily "developer" oriented, to the exclusion of concerns important to the end user.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 12, 2012 4:01 UTC (Sun) by thumperward (guest, #34368) [Link]
The GPL is not wholly about distributing sources. Attribution, for instance, is as important in the GPL as in many other licenses (free and proprietary). Would you expect copyright holders licensing their works under other licenses to allow their work to be used without attribution?
Following the GPL is a basic condition of distributing the software. There may be some wider community benefit to releasing source changes in some cases, but it's not simply an optional bonus. You quite simply have no permission under copyright law to distribute the software unless you are prepared to also attribute the authors and release the appropriate sources. This is what distinguishes the GPL from nonsense like "beerware" licenses where compliance is strictly optional and not really expected.
Chris
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 13, 2012 14:55 UTC (Mon) by NAR (subscriber, #1313) [Link]
Actually I wanted to ask this. I was under the (perhaps naive) impression, that most (if not all) of the GPL infractions come from lazy small companies, who download some stuff from the 'Net, put it into their product, but otherwise don't modify it in any meaningful way, so not much is won by suing them. I was also under the (perhaps even more naive) impression that if anybody tried to modify the Linux kernel to some non-trivial extent, then they would at make some noise on kernel-related mailing lists, if nothing else, then for technical questions, and in this case a half-hearted "code dump over the wall" would be not that difficult, but would satisfy the GPL.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 12, 2012 5:19 UTC (Sun) by jensend (guest, #1385) [Link]
While busybox may not see a lot of interesting modifications, SFC's enforcement efforts have resulted in opening plenty of very important modifications, quite often kernel modifications. That effect is both direct- whether it's the "just asking" method Welte is talking about or the "GPL death penalty" method people have debated- and indirect; the more people out there enforcing the GPL, even on projects where the modifications "don't matter," the stronger companies' motivation to release sources for all the GPL software they use.
The released modifications may not revolutionize the whole software world- they're usually of rather narrow interest- but in a lot of cases, the resulting source releases are the only reason consumers have been able to create and use alternate firmware on their devices. This is a very worthwhile effort and I salute SFC and the gpl-violations.org folks for what they've done.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 12, 2012 11:11 UTC (Sun) by misc (subscriber, #73730) [Link]
But how do you know that the modifications are useful if you cannot see them ?
Or better, what if the modifications do add bugs and security issues ?
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 12, 2012 12:58 UTC (Sun) by robert_s (subscriber, #42402) [Link]
"IMHO a more interesting issue is"
Well I don't understand how you could have that opinion, because it's completely missing the point that the relatively dull busybox is used to open up far more interesting changes to (and possibly drivers for) the kernel.
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 13, 2012 13:45 UTC (Mon) by slashdot (guest, #22014) [Link]
Why aren't they directly enforcing the kernel copyrights then?
Surely among the thousands of kernel contributors there must be someone willing to let them represent him...
Welte: Some comments on the heated debate on SFC / Busybox / Linux GPL enforcement Posted Feb 13, 2012 22:33 UTC (Mon) by dmarti (subscriber, #11625) [Link] +1 Insightful.
Widley un,odified software might not exist Posted Feb 13, 2012 12:06 UTC (Mon) by dps (subscriber, #5725) [Link]
I guess that most LWN reader swould think that there is no credible reason to change ieqbalance, which make a *big* difference on some serious hardware. I have had to modify irqbalance so it recognizes interrupts associated with ethernet interfaces named things like admin.
Lets it usffice to say this was done in a commercial context and there was an attempt to get it upstream---maintaining our down private patches rapidly becomes more expensive than being able to use the mainline version. Unfortunately the lack of response means this patch remains private.
|
|||||||||||||