Members of the Linux community have complained that the hot new sub-notebook from Asus, the eeePC, may have violated the spirit of the Linux General Public License. Some Linux advocates claim the eeePC has not included required source code with the installed Xandros Linux distribution and does not easily enable users to install another distro. However, there are indications that eeePC fans probably don’t care.
The ASUS EeePC has been a great triumph for Linux. OK, so one of the drivers hasn’t had it’s source released. Before the community starts burning piles of EeePCs, could someone try asking nicely and see if that works? If, a couple of months down the line, it looks like they aren’t going to play, then maybe it’s time to talk to a lawyer. Until then, I think ASUS are still with the good guys.
Yeah, this is simply Asus being incompetent. They included a printed copy of the GPL in the manual and even cited the GPL somewhere on the retail box. They posted a source tarball on their website, but it doesn’t contain the sources for all of the packages on the eeePC. Maybe the tarball was reused from some other Linux-based product that Asus distributes.
It’s quite possible that getting the proper source code on their website got bungled in their rush to get the machine out it time for the holidays. It’s hard to imagine that Asus would jeopardize a product for which they have very high hopes by distributing modifications to GPL software for which they won’t or can’t provide the corresponding source. I’d be very surprised if they don’t comply and avoid a lawsuit.
As Warren Woodford from MEPIS learned the hard way, the GPL requires the distributor to provide all corresponding source–at least upon request and payment of a reasonable fee–even for packages redistributed unmodified from an upstream distributor. So Asus is obligated under the GPL to provide source for GPLed Xandros packages it distributes unmodified.
However, Asus is not required to provide source for its modified madwifi ath_pci driver, since that is dual-licensed under GPL/BSD. As it seems that the version from madwifi’s SVN currently lacks support for the eeePC’s wireless chipset, it may be that the only solution for those who want to install a different Linux distribution is the binary from Asus.
I’m definitely giving Asus the benefit of the doubt and waiting for some official response. In the computer industry, if a sensational story is Slashdotted, Dugg, and OSAlerted, the company has no choice but to respond. The eeePC is a Good Thing(TM) for Linux, and Asus has done some other interesting Linux work like their SplashTop BIOS that runs Firefox and Skype. As OEMs go, Asus is among the more Linux-friendly, and I’m expecting an amicable resolution to this story.
Edited 2007-11-26 01:50
“So Asus is obligated under the GPL to provide source for GPLed Xandros packages it distributes unmodified.”
Why can’t they just point you to where Xandros has them?
Because, under section 3c of the GPLv2, that option would only be permitted if 1) Asus received the binary packages from Xandros along with a written promise to provide source code upon request for three years (3b) as opposed to Xandros providing public access to the source on their website (3a), and if 2) Asus were distributing these packages as a part of a non-commercial product.
Since they fail both of these criteria, Asus must comply under their choice of 3a or 3b. Since their tarball only partially complies under 3a, they must either provide the rest of the code on their website or distribute to all current and future eeePC owners a written promise to provide the rest of the source code upon request as per 3b.
The rationale is that Xandros is making it very easy for Asus to acquire the source code, and Asus is using the software for profit, so there’s no compelling reason why offering the source code as Xandros does for them would be an undue burden on Asus. Furthermore, since Xandros hasn’t promised to provide the source code for three years, they could very well go out of business tomorrow and take down their ftp server. So Asus has to provide source code so long as they are in the business of distributing the software.
More generally, the idea of a distributed software ecosystem is strained by the implicit pressure on upstream distributors from their downstream redistributors. This is a delicate relationship, and it makes practical sense to require that the downstream takes responsibility for its own operations. An upstream distributor should only benefit from gaining downstream projects. They should not dread their consumption of project resources, including bandwidth demands on ftp servers.
Yeah, this is simply Asus being incompetent.
…
It’s quite possible that getting the proper source code on their website got bungled in their rush to get the machine out it time for the holidays.
As an ASUS customer for many years, I can attest that their support and website quality has degraded a lot since I first started purchasing their products in the mid-90s.
These days, finding stuff on their website is painful, or simply impossible. I have found newer BIOS updates, manuals, and drivers for their motherboards in places other than their downloads section. That is ridiculous. Sometimes you have to go to their non-US website to find the right stuff. Clearly they don’t put enough effort into their website and support any longer.
Fist, I have not yet read the linked article, but the statement “has not included required source code with the installed Xandros Linux distribution” indicates a misunderstanding of the GPL terms.
They are not required to distribute the source code with the binaries at all, they are only required to make it available upon request and even then they can charge a nominal fee (ie. $10 to mail source CD’s).
It just seems like we have been a little quick with the pitch forks in the past.
Somewhat stupid conclusion, since no one asked for the code!
The GPL clearly says that you must provide access to the source, not that you must make it available through download channels!
Edited 2007-11-25 23:36 UTC
I’m glad folks like you and Priest are out there with the voice of reason. There were hundreds of zealots on Slashdot who jumped on the “boycott ASUS” bandwagon with this. Only a relative few were insightful enough to say the obvious, which is that ASUS hasn’t yet violated the GPL. Just because nearly every OSS developer out there automatically ships or overtly offers for download the source to their project doesn’t mean they explicitly have to. As you both said, “make available upon request” is the only requirement.
I wonder if any of the accusers have bothered to email, call or snail mail ASUS and request the source? I doubt it, but unless that happened and ASUS flat out said “no”, there’s really no reason to get upset at the present time.
EDIT: According to bremac’s post below, it was requested and not provided. I’m still going to wait on the sidelines and see what happens here. There’s no need in my mind to jump to any conclusions until more is known. I’m sincerely hoping ASUS does the right thing though; it would be a shame for such a nice piece of hardware to go away because of a stupid mistake.
Edited 2007-11-26 00:09
Not provided, or the person emailing unwilling to tell us what was in the email reply or simply because the company didn’t email back a reply within 5 minutes, they’re obviously rebuffing him.
Dear god, not all of us sit running shotgun on the email send/receive button, and reply within 5 minutes. If there is a reason for not instantly providing the source, it would have been within the email. Too bad there is more focus on the one solitary whiner than what was actually in the email that came form ASUS in regards to the source code request.
As for the person who requested it, anyone find it rather confusing that he is silent in the whole matter? why doesn’t he have the backbone to step forward and tell us the WHOLE story.
Edit: Interesting, someone has provided a link to the source code – and the forum seems to go quiet – isn’t it annoying when sensationalism is destroyed in one act of fact providing?
Edited 2007-11-26 01:21
Dear god, not all of us sit running shotgun on the email send/receive button, and reply within 5 minutes. If there is a reason for not instantly providing the source, it would have been within the email.
I’m not sure why you’re being so defensive towards my post, as it seems you and I are otherwise on the same wavelength here. As I said, I’m not drawing any conclusions until the whole matter is brought out into the open, and for what it’s worth I am actually rooting for ASUS on this. From what I’ve seen so far, they are simply not going above and beyond the call of the GPL and while that’s uncommon in the FOSS world, it’s not against the terms explicitly unless they flat out refuse or intentionally misdirect a request.
Why do people *instantly* assume that a reply to their post is exclusively directed at them?
My post was a ‘reply in general’ – a ‘build upon’ what you said in your post.
Okay my bad, I’m just used to people on this website arguing for the sake of the argument, and it seemed that you might have been confused about my position on this issue. No offense intended, mate.
All good. I tend to be on defensive around here at times because of the way in which people abuse the moderation system sorry if I came off as rather abrasive.
The posted tarball doesn’t contain the source code in question. You would know that if you had read the article…
Thank you for the correction – but it still doesn’t negate from the fact that the person who wrote the article has NO proof what so ever that ASUS has refused to give the source code. Simply because there is no ‘link on the site’ doesn’t mean they’re unwilling to give it out.
If they don’t give it out instantly, there might be a reason – I swear people here have become unreasonable bastards who assume that because something doesn’t occur at the speed of light, the company is obviously ‘teh evil’.
i believe they are also required to provide a written offer, which i believe they did, which provided an url for download..
and if this is really the case, i think he needs not contact them, as the paper states where to download – and should this be the case, asus had better deliver the code now. its simply not acceptable that they push the product like this before they are ready to comply with the licenses, just as they wouldnt want others to do, and because its free software it doesent give them this right.
Three points are key from the article:
1. A modified module (originally under the GPL) has not had the source provided, and
2. When the sources were requested, a large tarball was posted which _didn’t_ contain the relevant source code, though it apparently claimed to.
3. Attribution/copyright strings from the original module were stripped out in the released module.
There is a marked distinction between ‘not making the source available because it must only be distributed on request’ and ‘posting an unrelated tarball which claims to those sources, but actually isn’t’. Of course, one must assume this is an accident, and hope for a speedy resolution in the near future.
Edited 2007-11-26 00:03
AFAIK, ASUS outsourced software development to Xandros.
So who’s to blame?
Isn’t this the source code? http://dlsvr02.asus.com/pub/ASUS/EeePC/701/EeePC701_071012.zip
It’s on the Asus website provided here http://support.asus.com/download/download.aspx?SLanguage=en-us
Haven’t checked the code out but it says source code.
i saw in the article that this isn’t the source code, or more or less the changes asus have made arent in it.
Edited 2007-11-26 00:45
The code in question is NOT in the ZIP file. It has been checked and verified. I hope Asus updates this file soon so this issue can die. If they do not do the right thing it could get ugly. I am sure Microsoft would be happy to bankroll anyone willing to sue Asus over this issue.
It’s going to be a cold cold day somewhere warm the day Microsoft decides to bankroll the FSF (the place that makes the majority of said lawsuits).
Considering the amount of software* released by Microsoft under the GPL it would be unwise for Microsoft (and its religious followers) not to keep their mind open
* parts of Microsoft SFU (Services for Unix) are under the GPL.
“I recommend that folks reading this article read over my blog post before commenting. The article takes it out of context and tries really hard to sensationalize it.”
Today’s sensationalism without further analysis is brought to you by OSAlert. Still, it’s worth following this story, I suppose.
“Members of the Linux community have complained that the hot new sub-notebook from Asus, the eeePC, may have violated the spirit of the Linux General Public License. Some Linux advocates claim the eeePC has not included required source code with the installed Xandros Linux distribution and does not easily enable users to install another distro. However, there are indications that eeePC fans probably don’t care.”
How much familiarity with the license in question and topic at hand do those that mis-statement imply. Its sad when you take things out of context due to a lack a familiarity and speak what you don’t know.(I’ve been guilty of that in the past, hence my apreciation for the Common Usability Terms)
How do reporters strike a balance between Informing and Inflaming cituations and events?
OSAlert staff do not necessarily write all the submissions (although the staff do some quality control on user submissions). When you see the line “submitted by” under the story title, the descriptive paragraph is mostly what the user has written.
WRT journalism, the only way to strike a proper balance between informing and inciting is to do the research necessary to fully understand the situation, but this is not always feasible.
For example, if a person claims that Microsoft includes code in Vista to randomly rearrange a few bytes in Linux kernels on other partitions, then the story is published as a claim (not a fact). If we later learn that the person has made other such claims in the past that were exposed as lies used to get attention, then the news story appears deceptive. Anyone who comes in late and reads just the story will know only the unsubstantiated claim and not the poor character of the person making the claim.
Good journalists will do at least some background checking on any new item they publish. The whole point of being a journalist is to provide *quality* information, gleaned from the noise and FUD.
[Sure, there are scum out there who will call themselves “journalists” while deliberately publishing lies so that the rest of the media (and the general public, too) will get swept up in some grand, quixotic fantasy of saving the world from the foul machinations of an evil company or political machine. (Who does not wish to be a hero?) But such fantasies distract from the task of a journalist, which is much more basic, and humble (like an earthworm): journalists exist to verify information and to pass on the truth.]
The fact is, one cannot tell how much research there is to be done before one can be certain that the claims in a story are reasonable and well-supported. Statistics being cited may have come from a report funded by someone (or some company) who frequently funded other questionable (or rigged) studies on the same subject. Or the person submitting the news item may have taken a quotation out of context or misattributed a quotation. Or at times even a major news site like the NYT will publish articles that are mere fabrications. While some news items can be easily verified by the staff, it is entirely unreasonable to expect *every* news item published here to be *completely* accurate.
Yet this unreasonable expectation is held by the general public. Why? In my opinion, the root of the problem is the general human tendency to assume that a statement is true just because it has been published on a high-traffic website, a well-regarded newspaper, or even in a scientific journal. A few people see a false news item and immediately begin to realize the general public will just assume the news is true. That may happen (even routinely), but just assuming the news is reliable is still a very bad idea. IMO, the general public is also at fault: they should be a little more skeptical of the news as a general rule.
What anyone lets into his or her mind should be carefully tested and filtered, whether that person is in the media or not.
So I would say that some journalists betray the trust of their readers, and some readers overreact to honest errors of legitimate journalists. Journalists are expected to exercise some self-control so as not to get caught up in sensationalism, but readers must also exercise a bit more control over their own emotions so as to not call FUD unnecessarily.
A quotation from an interesting character:
He who fights with monsters might take care lest he thereby become a monster. And if you gaze for long into an abyss, the abyss gazes also into you.
–Friedrich Nietzsche, Beyond Good and Evil, Aphorism 146
Sometimes, the only difference between the people posting FUD and the people calling FUD is which side they happen to be on…
So did they violate the GPL or did they “violate” the spirit of the GPL? The first is a breach of contract, the second is just unethical or stupid.
From what I can gather, neither so far. They haven’t flat-out refused a request for the source, but they aren’t being nice enough to preemptively give it away on their website either. They haven’t violated the letter at all yet, and as for the spirit, well that is subjective in my opinion. If the hardline supporters of the GPL want to rant and rave about a company not instantly publishing source changes without an initial request, well whatever gets them off I guess. Granted, now that the issue is out in the open, I certainly hope ASUS will step up and publish their changes so this will become a non-issue.
Either way, someone who knows what to look for should email or call ASUS and request this code already so we can get on with our lives.
Mine’s in the mail now. It bothers me a bit that they have possibly violated the GPL. I’m hoping (and expecting) that they will comply soon. Going to court doesn’t help them OR the community, so I hope this is resolved in the best possible way.
On another note: With the Eee plus the recent success of the Walmart $200 PC could this be the mythical “Year of Linux on the Desktop”?
I’ve been running Linux mostly exclusively for almost 5 years, but…I’m always pulling for projects like the Eee to succeed.
I get the impression that the site is not quite finished, and that ‘Source Code’ is something new to their support download structure. I think, as many others, that folks are jumping the gun. Would seem that all will be appropriately sorted out before very long.
They have provided source code before for other of their products that use Linux, such as the WL700ge NAS/router. Because they provided source code, and hobbyists have taken that, you can now even get fully open versions running on their hardware:
http://wl700g.homelinux.net/portal/
The open versions actually end up considerably extending the device functionality over what is originally offered for sale.
This is a reason to support ASUS, and to actually buy their hardware, not to hassle them.
Edited 2007-11-26 05:02
Hope you guys will not push ASUS too hard, and please do it in a friendly way and on a win-win basis, otherwise they might be forced to release only the version pre-installed with XP, at a higher price of course, and we suffer …
Just reading the comments to this article…
A company (Asus) is caught in a flat-out violation of the GPL, with clear proof of said violation, and people call for calm and tempered reactions. They’re good guys — they can do no wrong — oh, please don’t overreact.
When other companies have been merely suspected of a violation (take your pick) — or even just accused, people call for exhaustive searches for evidence, for preemptive lawsuits and legal sanctions against the companies… and there are ill feelings towards the companies for a long time afterwards…
What’s the difference? Marketing, I suggest…
Yeah, it’s called “learning”. Even if you overreacted a gazillion times, you can always *grasp* think about it and try to behave properly, specially when:
1) There’s no flat-out violation, ASUSTek is not obliged to provide download links for the modified source.
2) Their product can bring a lot of benefit to Linux and FOSS in general.
Attacking ASUS right now is just plain dumb, just ask them politely to rectify the matter and that’s all.
Agree completely. While I understand they’re in violation, suing them over this to me is biting the hand that feeds.
Laptop is a breakout sensation and causes more attention to be focused on Linux and FOSS in general. So let’s go ahead and sue the hell out of them without giving them a chance to rectify what they’re doing wrong.
Excuse me? Tell me again exactly who produced the code and who consumed it?
Let’s not over exaggerate. Asus is not passing this work off as their own, they just didn’t follow guidelines for posting source code.
My God, it would have been easier for Asus to just have put a $90 OEM XP Home license on there and tack it on to the selling price.
At least then it wouldn’t have the FOSS babies crying. That way we can continue to have the Windows market domination everyone here whines about, instead of getting Linux into more hands.
Finally someone who gets it!
I’m also sick of all this butt licking just because “ooh the eeePC is going to skyrocket Linux usage”. This is ladies and gentlemen a typical indirect power misuse.
ASUS didn’t even have to move a thumb, but just because they are now in the position of “power” (eg: the one guy who is about to distribute a linux-PC) they get a different meter than others.
If a small group of enthusiasts was getting rammed by a huge company because they managed to violate some stupid license in one file no-one would give a f*.
Great justice guys..
BTW: I’m not saying “burn them”, of all the people I think that all these things (no matter who does what wrong) should first be handled on the nice level. But I also know that if you go nice, it invalidates a lot for you in the court if it gets to that.
I’m just sick of this double-standard. There were other violations and the reactions there were completely different, just because a company is “doing good” for Linux doesn’t mean we’ll handle them differently now…
Hang on a minute. They were asked for source code, they provided source code. They have provided correct source code in the past for other products.
This time, they seem to have made an error in their provision of the source code. Lets just give them a chance to correct that first, hey?
All that stuff’s been happening over at Slashdot, so don’t worry – it’s covered. Over here, the crowd tends to be a bit less dug-in. I don’t seem to recall too many people condemning Mepis either when they got into trouble.
You say there are ill feelings towards towards companies afterwards, but off the top of my head I can’t name a single company that’s been in trouble over the GPL. I only remembered Mepis because someone mentioned it earlier.
I don’t remember reading in the GPL 2 and now looking at the GPL 3 any place that mentions a time line between when someone asks you for the source and you deliver it.
If it was there I believe that such a section would talk in weeks or months since not everyone can immediately distribute the source. But since ASUS has said here is the source but it is not there. It just looks like a typical large corporation mistake.
However the stripping of some of the attributions is a serious crime. If there is a serious crime here that is it.
They should declare a fifth freedom: the freedom to run an excellent product without getting bashed by GPL fanatics. Seriously… I do love free software, but for me freedom is not restrictions and limitations.
They should declare a fifth freedom: the freedom to run an excellent product without getting bashed by GPL fanatics. Seriously… I do love free software, but for me freedom is not restrictions and limitations.
Then, don’t use software that abides by the GPL or software with more restrictive licenses, like proprietary software. There is a huge body of BSD/MIT/Apache licensed software. Use that, and start whining about the licensing choices of others, who give their code away, with the restriction that others should be able to get the code too.
Oh please – as I’ve stated many times in the past, if they don’t like the terms of the GPL, then they should get their greedy mitts off the software. It’s THAT simple. No one is making you, or them, use GPL’d software. If you dislike the GPL that much, then please get rid of any GPL’d software that you might be using on your system.
Dave
Asus is delivering an AAA+ product that brings hundreds and hundreds of people to Linux, then hordes of GPL zealots are ready to take them down for not offering download for sourcecode of several driver files that maybe 10-20 people could do something with and 3-4 people are actually interested about.
Only thing that is seriously holding Linux back is the GPL-zealots.
Asus is delivering an AAA+ product that brings hundreds and hundreds of people to Linux, then hordes of GPL zealots are ready to take them down for not offering download for sourcecode of several driver
This has nothing to do with zealotism. They chose to use GPLed software, then the have to abide by the terms of the license. Try to include a piece of unlicensed Windows or OS X code in your product, and see how fast Microsoft or Apple will come after you.
If you are not happy with the requirements of the GPL, write your own code, or pick up a BSD system, which one can use with virtually no restrictions.
10-20 people could do something with and 3-4 people are actually interested about.
When the sourcecode is available, the changes can be integrated in the upstream kernel, meaning that *all* eeepc users can choose to install another distribution than Xandros, with proper hardware support.
And remember that many of the initial buyers are likely to be more experienced computer users and techies, who might prefer another distribution.
…files that maybe 10-20 people could do something with and 3-4 people are actually interested about.
The fact that Asus haven’t released their modified ACPI driver makes it difficult to use any other Linux distribution on the EEEPC, since it won’t support it’s power management. They should of course have submited the modifications upstream. That’s how open source works.
Edited 2007-11-26 12:06
Hi,
I’m curious: can a company provide the source code 50 years after people ask for it, and still comply with the GPL?
The GPL says that source code must be provided, but doesn’t seem to say when or how quickly the source code must be provided…
Isn’t this the same riot we had when apple didn’t release the source code to one of there tiger / leopard releases straight away. They released it a few months later and everyone was happy.
I say give Asus a bit of time, yes it is a linux machine but perhaps the EeePC division is pretty small and they are only interested in providing quick fixes and other updates as the machine has only just been released.
Isn’t this the same riot we had when apple didn’t release the source code to one of there tiger / leopard releases straight away. They released it a few months later and everyone was happy.
The OSX sources isn’t GPL so Apple wasn’t doing anything wrong by not releasing them.
I believe that the parent poster was talking about Webkit, which is indeed based on the GPL (LGPL?) KHTML engine and thus he is right in this regard, although he could have worded it a bit better.
Edited 2007-11-27 17:37
That wasn’t Apple, as tiger / leopard aren’t GPL.
It was Parallels, if I recall correctly, and they had used Grub. Yes, they did release the code a couple of weeks after a fuss was raised.
because people are being hardasses about this. Yes, they should release the updated source soon, and preferably free via a link on their site.
I don’t have an EeePC because I already have a MacBook, but I’m considering getting my wife one. She won’t give a damn about source code, but I agree Asus needs to follow the GPL.
Man listen to you guys. Maybe they should have just used windows? They would have a lot less grief. Probably would have sold more.
How so? Windows costs a fortune. There are per-machine costs that represent and significant proportion of the machine’s total, and yet the system has a bare OS and no applications when delivered.
The only cost to ASUS for their use of GNU/Linux/FOSS will be to release the correct source code. EeePC users will get a full set of applications to enjoy the use of, without having to pay extra for application software after buying the machine.
Whatever…
Try openly violating a Microsoft License. See how far it gets you.
… of the FOSS community eating its young. Man, freedom tastes so gooooooooooood, doesn’t it?
You all can beat this topic (and ASUS) to death while I have fun with my new Linux-based eee pc. And you can count on me showing all of my friends how versatile and fun Linux is…