Not logged in
Log in now
Create an account
Subscribe to LWN
Deadline scheduling: coming soon?
LWN.net Weekly Edition for November 27, 2013
ACPI for ARM?
LWN.net Weekly Edition for November 21, 2013
GNU virtual private Ethernet
Samsung releases exFAT filesystem source
Posted Aug 16, 2013 14:56 UTC (Fri) by ssam (subscriber, #46587)
If this was in the kernel it would be up to the distros to decide whether to build and ship it. They make similar decisions about lots of code with compile time options.
Posted Aug 16, 2013 15:05 UTC (Fri) by armijn (subscriber, #3653)
Regarding the patent situation for exFAT: I would not touch this code right now, it looks like a patent bomb to me.
Posted Aug 16, 2013 15:26 UTC (Fri) by ssam (subscriber, #46587)
Posted Aug 16, 2013 15:31 UTC (Fri) by armijn (subscriber, #3653)
Posted Aug 17, 2013 1:15 UTC (Sat) by tterribe (✭ supporter ✭, #66972)
Posted Aug 16, 2013 21:59 UTC (Fri) by kreijack (guest, #43513)
What I mean is that even if the source code is released under GPL, that doesn't affect any Microsoft rights of the exFAT patents.
So if you want to ship the Samsung *compiled* code you have to respect the GPL requirements, and you have to ship the source too; moreover you must have a license from Microsoft for selling/distributing the *compiled* code.
The GPL is *another* requirements to the Samsung exFAT code other than the Microsoft patents obligations.
May be that shipping the code, is legal; but for sure you need a Microsoft license for shipping the compiled code or you are in risk to be sued.
That happened in TomTom vs Microsoft court case, related to the FAT32 patents. (http://en.wikipedia.org/wiki/Microsoft_v._TomTom). It must to be noted that Microsoft never prohibit Linus to ship the source code of the FAT32 in the linux kernel. But sued TomTom when it tried to ship the compiled code.
But I don't know what happens if you ship (like Samsung does) the source code and an user
a) download the exFAT source code
b) compile it and execute it on its own computer
Because nobdy ships nor sells the *compiled* code.
To me it seems a way too simple to circumvent the patents obligations.
Posted Aug 17, 2013 19:29 UTC (Sat) by SLi (subscriber, #53131)
Posted Aug 18, 2013 4:12 UTC (Sun) by drag (subscriber, #31333)
It is true that if ExFAT is a GPL derived code then Samsung has a legal obligation to offer source code versions to all receivers of the program.
It is also true that if ExFAT is covered by Microsoft patents then Samsung or the receiver of the program is legally responsible to get a patent license from Microsoft.
The only thing to call into question is at what point does one infringe on the Microsoft if Exfat requires a patent license. It may happen during shipping and duplication of the code, but the patent may only kick in when it's running on hardware. It just depends on what exactly the patent covers.
For example; if the patent has in the claims that one of the steps indicates the software interaction with the hardware.. then as long as you don't have the software installed on a system then you are not liable for patent infringement. So it may be legal to ship the source code, but once you compile it and ship it on hardware then you are infringing Microsoft's patent.
It's all very terrible and illogical, but don't expect Intellectual Property stuff to make any sense at all. The whole thing on what patents and copyright covers is completely arbitrary.
Posted Aug 18, 2013 8:56 UTC (Sun) by SLi (subscriber, #53131)
Posted Aug 18, 2013 9:29 UTC (Sun) by kreijack (guest, #43513)
Anyway I found this page
which highlights a sentence of the point 7 of GPLv2:
"7 [...] For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program."
If I understood correctly, Samsung could not be in position to ship the code (both the source and the Program) at all.
Posted Aug 18, 2013 9:39 UTC (Sun) by SLi (subscriber, #53131)
Nevertheless, copyrights and patents are two very different things (and trademarks are the third category of IPR most often talked about). Copyrights cover the expression, patents cover the idea.
Posted Aug 18, 2013 9:05 UTC (Sun) by Seegras (subscriber, #20463)
You can't just get catatonic for perhaps some illegally granted patents. The only course of action is to ignore them in the first place. Because not only are they illegal, but it's also impossible to clear them in the first place http://www.groklaw.net/article.php?story=20130715054823358
Posted Aug 19, 2013 15:47 UTC (Mon) by drag (subscriber, #31333)
There is really no such 'patents cannot cover math'. All patents do is cover algorithms. That is it. A series of steps to accomplish some goal. Since all algorithms can be expressed in math then all patents cover nothing but math, right?
Instead, what can be covered by patents is very arbitrary. Some times of scientific discoveries can be patented, but not other types, for example.
Don't try to make sense of it. There is no sense. The only thing you can do is get a lawyer who has the ability to to site court history and knows the lingo properly to tell you how likely you are infringing on something.
Posted Aug 22, 2013 11:34 UTC (Thu) by Seegras (subscriber, #20463)
Not at all. Of course everything can be expressed mathematically, but that does not mean it _is_ mathematics. You can patent a screw, and describe it completely mathematically, as a series of functions. The patent is on the screw, and not on the mathematics. You can use those formulae to calculate as many screws as you want, as long as you don't produce them as tangible items.
See also: http://www.groklaw.net/article.php?story=20110908075658894
Copyright © 2013, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds