FAT legal status?

[Sik]

Who cares about supporting exFAT? Microsoft's other FAT patents have expired. I am happy enough about that. People have been trying to explain, bzt.

Pretty much all current SD cards use exFAT, and reformatting them is an extremely bad idea as their firmware is designed to optimize wear levelling assuming you're using the pre-formatted filesystem. Also other devices may refuse to accept a SD card that is large enough but not formatted as exFAT (depending how pedantic they want to be), and even ignoring that most users aren't really in the mood of reformatting a drive that already works (if they even know what reformatting is).

So if you want to support SD cards you want to support exFAT.

[Kazinsal]

Christ. This is a solved problem and reading more than 10 words past "Microsoft holds patents on exFAT" would show you the solution. It's a ridiculously simple one.

Members of the Open Invention Network are granted a license to use the exFAT patents, among many others.

[zaval]

Sik, you are right, with one note, that exFAT is a requirement to only SDXC cards. that's those pricey cards, whose size is >32GB. I am not sure, about "pretty much all current cards" being them.

Part2, page 10. and don't ask where I saw that.

The type of the file system to be used shall be uniquely decided with Card Capacity as follows. Here, Card Capacity means the total size of Data Area and Protected Area size.

Code: Select all

Card Capacity                  |  Card Type                                      |    File System Type for Data Area
---------------------------------------------------------------------------------------------------------------
~2,048MB                       |  Standard Capacity SD Memory Card  |   FAT12 / FAT16
2,088.5MB(*)~32,768MB |  High Capacity SD Memory Card         |  FAT32
32,896MB(*)~2,048GB    |  Extended Capacity SD Memory Card  |  exFAT

Table 1-1 : File System Type
That is, all Standard Capacity SD Memory Cards whose capacity is 2048MB or less shall use FAT12 /
FAT16 file system, and never use the other file system. Similarly, all High Capacity SD Memory Cards
shall use FAT32 file system, and never use the other file system. And all Extended Capacity SD Memory
Cards shall use exFAT file system, and never use the other file system. This includes the prohibition of
partial format of SD Memory Card. For example, 8GB High Capacity SD Memory Card should not be
formatted as 2GB card with FAT12 / FAT16 file system. In this case, whole area of 8GB should be
formatted with FAT32 file system.

I made this quote, as it might be helpful for implementers over here, since there are "some" members, repeatedly rejecting the fact, SDA demands using FAT FSs as the strict requirement. these highlighted words, repeatedly formulated, for every card capacity type, must have convinced even the most thick skinned ones, correct?

[PeterX]

Christ. This is a solved problem and reading more than 10 words past "Microsoft holds patents on exFAT" would show you the solution. It's a ridiculously simple one.

Members of the Open Invention Network are granted a license to use the exFAT patents, among many others.

So if you are so much smarter than us, can you please share your wisdom with us?
What is the answer to bzt's questions?

Greetings
Peter

[Kazinsal]

Christ. This is a solved problem and reading more than 10 words past "Microsoft holds patents on exFAT" would show you the solution. It's a ridiculously simple one.

Members of the Open Invention Network are granted a license to use the exFAT patents, among many others.

So if you are so much smarter than us, can you please share your wisdom with us?
What is the answer to bzt's questions?

Greetings
Peter

The answers are:
- The most recent FAT LFN patents claim US4149793A as prior art, meaning they inherit US4149793A's expiry. As it was filed on 1997-01-28 (after the 1995 increase of US patent lifespans from seventeen to twenty years), it has a twenty year expiration from the date of priority, despite the prior art (US4149793A) being filed on 1993-04-01, when patent terms were seventeen years. Thus, the FAT LFN patents in the United States are expired as of 2013-04-01.
- The GNU General Public License does not supersede patent law.

[bzt]

There's no such thing, "your Samsung license", since the driver was licensed under GPL

How do you come to this conclusion?

Pretty simple, take a look at the comment in the source file, it does not say "Samsung specific licensing terms", rather it reads as "licensed under GPL". This means only GPL terms and conditions apply (which forbids Samsung to add further terms or restrictions).

But software patents and even copyright of third parties (read: MS) is NOT "turned off" by the GPL.

This is the part where most of you are mistaken. GPLv2 section 7 explicitly states that only royalty-free patents can be licensed under GPLv2. Plus GPL explicitly forbids to add any further restrictions on a derivative work (like a patent royalty for example). So while it is true that GPL does not "turn off" patent rights, but it mandates that only code that is patent royalty-free can be licensed under GPL.

If Samsung violates or ignores patents or copyright, it does NOT give the end user the right to use the software. (Maybe we are simply cross-talking...)

Yeah, I think too that we might cross-talking. MS agreed for the Samsung code to be licensed under GPL, and GPL is what gives the right to the end user to use the software without a patent fee. If this wasn't the case, then GPL would not allow distribution of the code.

So if you want to support SD cards you want to support exFAT.

Yes. It's also a requirement if you want to store larger than 2G files on a USB stick for example.

This is a solved problem and reading more than 10 words past "Microsoft holds patents on exFAT" would show you the solution. It's a ridiculously simple one.

If it were that simple, I wouldn't be asking The problem here is, MS gave up on their right to ask royalty for their patents, and according to the GPL, this royalty-freeness must be forwarded to derivative works too (as long as those derivative works adhere to the GPL terms of course).

The GNU General Public License does not supersede patent law.

This seems to be the most common misunderstanding. Of course GPL does not supersede patent law, but it does mandate that only royalty-free patented code can be licensed under GPL.


Long story short: if MS ever wants to ask for a royalty for a Linux derivative exFAT driver, then (in accordance to GPL section 7) the exFAT driver must be removed from the Linux kernel immediately.

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.

Cheers,
bzt

[Sik]

So if you want to support SD cards you want to support exFAT.

Yes. It's also a requirement if you want to store larger than 2G files on a USB stick for example.

Nitpick, but it's 4GB (I know because I had to deal with that a couple of weeks ago or so, ended up running into a file that was just a few KB more than 4GB and made everything a pain). But yeah, that's a good point too.

[bzt]

Nitpick, but it's 4GB (I know because I had to deal with that a couple of weeks ago or so, ended up running into a file that was just a few KB more than 4GB and made everything a pain). But yeah, that's a good point too.

In theory you're right, but in practice there are many implementations that use signed int for filesize, so if your goal is interoperability, then 2G is the max. Sadly not all mainstream FAT drivers can read 4GB files (looks like Android and MacOS both affected, or at least they were. I don't know if they have this fixed or not).

Cheers,
bzt

[eekee]

Does anyone know who currently holds copyright to the Linux kernel? Linus Torvalds didn't seem to want to go against companies who breached it. He even characterized himself as a fat and lazy bishop to Stallman's messiah. He only criticized nVidia's closed source drivers for being a nuisance, not legally wrong. However, since he... angrily retired... I have no idea who holds the copyright or what their attitude is.

[nexos]

AFAIK each portion is copyright by it contributor. i.e., the header comment in one file states:

Code: Select all

/*
 * linux/ipc/sem.c
 * Copyright (C) 1992 Krishna Balasubramanian
 * Copyright (C) 1995 Eric Schenk, Bruno Haible
 *
 * /proc/sysvipc/sem support (c) 1999 Dragos Acostachioaie <[email protected]>
 *
 * SMP-threaded, sysctl's added
 * (c) 1999 Manfred Spraul <[email protected]>
 * Enforced range limit on SEM_UNDO
 * (c) 2001 Red Hat Inc
 * Lockless wakeup
 * (c) 2003 Manfred Spraul <[email protected]>
 * (c) 2016 Davidlohr Bueso <[email protected]>
 * Further wakeup optimizations, documentation
 * (c) 2010 Manfred Spraul <[email protected]>
 *

Making me believe it is copyright of many. Linus, Intel, and Red Hat probably have the most amount. The exFAT driver would be copyright of Samsung.

[bzt]

I have no idea who holds the copyright or what their attitude is.

This "copyright-holder" is an obsolete concept which does not apply to Free and Open Source Software. From wikipedia:

the rights-holder (usually the author) of a piece of software can remove these restrictions by accompanying the software with a software license which grants the recipient these rights. Software using such a license is free software (or free and open-source software) as conferred by the copyright holder.

See also Free Software definition by GNU and Stallman's Public Awareness of Copyright paper.

The Linux kernel is licensed under GPL-2.0, and all contributors (Linus, Intel as well as Samsung) must comply. This means they must gave up their rights as the only copyright holder, and agree to use the terms and conditions of the GPL. Simply put, doesn't matter who the original copyright-holder is, because the rights are transferred to and shared with the users, that's what makes it a free software.

Cheers,
bzt

[PeterX]

This "copyright-holder" is an obsolete concept which does not apply to Free and Open Source Software.

I just want to say publicly that I disagree on this, because it's important. But I don't go into details because I hope this won't end up in an endless flame war like so often here in this forum.

Greetings
Peter

[bzt]

I just want to say publicly that I disagree on this, because it's important. But I don't go into details because I hope this won't end up in an endless flame war like so often here in this forum.

I would kindly ask you to elaborate. Stating something without saying why just makes no sense. I don't want a frame war, I'm only asking about your opinion, because I don't understand your position. Are you saying that a free software license does not grant all the rights of the copyright-holder to all of the software users?

Do you question the sentence on the wikipedia, which reads as

the rights-holder (usually the author) of a piece of software can remove these restrictions by accompanying the software with a software license which grants the recipient these rights.

?

Or do you question GPL, in this:

Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions.

So what does it matter who the original licensor was, when all users automatically receive the same license with exactly the same permissions, terms and conditions?

Please explain what you meant.

Cheers,
bzt

[nexos]

People overlooked the source itself...

AFAIK each portion is copyright by it contributor. i.e., the header comment in one file states:

Code: Select all

/*
 * linux/ipc/sem.c
 * Copyright (C) 1992 Krishna Balasubramanian
 * Copyright (C) 1995 Eric Schenk, Bruno Haible
 *
 * /proc/sysvipc/sem support (c) 1999 Dragos Acostachioaie <[email protected]>
 *
 * SMP-threaded, sysctl's added
 * (c) 1999 Manfred Spraul <[email protected]>
 * Enforced range limit on SEM_UNDO
 * (c) 2001 Red Hat Inc
 * Lockless wakeup
 * (c) 2003 Manfred Spraul <[email protected]>
 * (c) 2016 Davidlohr Bueso <[email protected]>
 * Further wakeup optimizations, documentation
 * (c) 2010 Manfred Spraul <[email protected]>
 *

Making me believe it is copyright of many. Linus, Intel, and Red Hat probably have the most amount. The exFAT driver would be copyright of Samsung.

The (c) symbol isn't there for no reason. It means that Red Hat has copyright of there portion, Manfred Spraul has copyright of their portion, and so on. The GPL means we don't have to be that concerned about it, as we can redistribute it just the same, All of these contributors agreed to the GPL. They didn't give up their copyright. That's called public domain.
If anyone argues and flames, they can see my opinion on the matter here.

[bzt]

The GPL means we don't have to be that concerned about it, as we can redistribute it just the same, All of these contributors agreed to the GPL.

So why does it matter who the original copyright-holder was then? As soon as the agreed to GPL, they have exactly the same rights as everybody else.

But back to the topic: Is a derivative of a GPL'd code receive the same patent conditions as the original GPL'd code? According to my interpretation of GPL, they do. Opinions?

In the GPL preambule, they wrote

Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone's free use or not licensed at all.

Cheers,
bzt