Slashdot Mirror
Is Sveasoft Violating the GPL?
Ron Harwood writes "First, Linksys was violating the GPL by not releasing their source for their Linux implementation on the WRT54G wireless router and WAP54G access point. When this was rectified, third party firmware started showing up. Well, now it looks like Sveasoft (one of the third party developers) has decided to restrict access to their modified source code to subscribers - that also will need to pay $49 for a CD rather than being able to download it." The thread summary at DSLReports only makes it clear that this is all very complicated.
There is no way out of this cycle. The lawyers will be getting rich, the "small" guys will still not have the $$$ to protect their GPL licenses in court, and the circle will continue, albeit viciously.
Everything was fine, up until they tried to prevent peopel from redistributing code. Last I checked, you had to pay to get the binary AND the source code. Nothing in the GPL says you have to give out the source, or binary, for free. It says that if you give out a binary, you have to give out the source. So you are free to charge for your software, but the source had better be on the disk as well (or available on the web).
Where they cross the line is in trying to stop people from distributing modified version. Nope, sorry, can't do that. That's the "viral" part MS complains about. Even if your product costs money, someone is free to make their own version that does not for free, the GPL gaurentees that.
This is really the thing that commercial companies worry the most about. Source distribution, though generally not done, isn't a big threat to most of them. The threat is then that with that source people are allowed to make their own versions of the product for free.
However, as you said, if you can't deal with the GPL's rules, don't use GPL code. I have no sympathy. The redistribution part is real, real clear.
Yeah, but by that logic they could just make the source avaliable only on special solid platinum CDs hand-carved by tibetan monks and charge three billion dollars each. Sure, it's the actual cost of distribution, but it's not reasonable. An FTP server would do the job for at least three orders of magnitude less money and be more convenient for everyone to boot. I don't think a judge would be impressed by that argument, even if it doesn't exactly violate the letter of the GPL.
main(c,r){for(r=32;r;) printf(++c>31?c=!r--,"\n":c<r?" ":~c&r?" `":" #");}
'Standards' in computing only impress those who are impressed by things like 'standards'.
he GPL only kicks in when you DISTRIBUTE your modifications.
My thoughts exactly. Now suppose someone like Sveasoft wants to have beta testers of their product, and to become one, you pay $49. You're now part of the club and maybe could be considered part of the organization.
So, they send you binaries to test. Are you considered internal to the organization now? Or is this an external distribution? If it's the former, they may not have to give you the source at all. And if they do give you the source, they could "kick you out of the club" if you chose to distribute that source that is in beta form. If it's the latter, then what actually constitutes "in-house" vs a public distribution?
This issue will be of interest to many businesses considering working with FLOSS as part of their business model. Many could be uncomfortable with not-quite-finished versions of their software gaining world-wide distribution, especially when their name is attached to it. By forcing them to distribute source to their beta-versions, they really lose the ability to beta test with any more people than are actual employees. Losing beta testing could be a factor to decide against adopting FLOSS.
On the other hand, by allowing beta "in-house" distributions, there is the risk of being perpetually in beta. Here, if someone wants the software at all, they put their name on a web form to become a "member". The software never exits "beta" and they never have to distribute their source code changes.
(Get together with another subscriber and compare source first, to see if any "tags" were placed in the source, too.
It should be much easier to remove the tags from the source code than from the binaries.)
Those who sacrifice security to condemn liberty deserve to repeat history or something. - Benjamin Santayana
Well, we are getting into pretty philosophical territory here. You're right, I could see a judge going either way on this. But don't think the GPL just ignores other contracts that might conflict with it; section 7 of the GPL deals quite nicely with this issue. If you have signed a contract that makes it impossible for you to satisfy the GPL's conditions (such as not restricting redistribution), you cannot distribute the software at all. Just because the restricting clause is in a different contract doesn't excuse you from complying with the GPL.
main(c,r){for(r=32;r;) printf(++c>31?c=!r--,"\n":c<r?" ":~c&r?" `":" #");}
That article you mention has a link that pointed to Satori_v2_2.00.8.7sv-pre1.bin.zip as having QOS. Note that this is version 2.0 so I'm guessing QOS has been in the firmware for quite sometime.
On this page they list links for both binaries and source for Satori v.4.0
http://www.linksysinfo.org/modules.php?name=Downlo ads&d_op=viewdownload&cid=8
This is the PUBLIC version which sveasoft has realeased for FREE.
Are you saying that only the latest, pre-release firmware will satisfy your needs? Are you saying that they removed QOS in laster versions and are only releasing it to subscribers? Otherwise, I just don't understand why you are slamming Sveasoft when they have given you what you wanted for free.