Announcement

Collapse
No announcement yet.

Native Linux Kernel Module Is Out For Microsoft exFAT

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #71
    Originally posted by birdie View Post
    For interoperability you are allowed to write code (from scratch, without using original specs) to operate with closed technologies/standards - there's a provision for that in the US law - that's exactly how in-kernel vfat/ntfs modules were created in the first place - it doesn't matter if they infringe or not on MS patents - they are made using clean room reverse engineering.

    exfat on the other hand was written after MS IP - thus this whole driver is MS's IP.

    Only clean room reverse engineered exfat can be legally merged upstream.

    End of story, end of thread.
    Originally posted by RealNC View Post
    I'm sure there's a law about interoperability which applies here and pretty much disarms patent claims.
    You are both confusing patents and copyright. You can't work around patents with clean room design: Wikipedia

    Originally posted by Wikipedia Clean Room Design
    Clean room design is useful as a defense against copyright and trade secret infringement because it relies on independent invention. However, because independent invention is not a defense against patents, clean room designs typically cannot be used to circumvent patent restrictions.

    Comment


    • #72
      A lot of countries in the world don't have software patents so I'm not mistaken. ;-)

      Comment


      • #73
        Just ONE end-user lawsuit would sink exFAT just like GIF

        Let's get real: no individual user is going to buy Windows because some patent says their camera's files are "illegal" to read from a Linux box. Hell, Mint goes so far as to say that only corporate or institutional users need to use the "no-codecs" versions in software patent countries. If the Sumsung module mixes GPL and MS code (copyrights), we still have the FUSE method.

        I see ExFAT becoming another MP3 or H264-a standard institutional users must avoid, but nobody else. US-based distros will simply keep it in foreign repos only. There will always be countries that don't sign NAFTA-style trade deals to host such repos. Some don't even attempt to enter the WTO world.

        This means that people will always be able to get either GPL or warez software to read the filesystem on their linux boxes. For individual users there is no deterrent, no reason to follow un unenforceable and unjust law. At worst, it would be like finding libdvdcss somewhere so you can use those old DVD's you bought before the MPAA boycott.

        OK, so MS cannot enforce their ExFAT patent on end users, anymore than the RIAA can stop music sharing. What can they do about it? Years ago, the owners of the .gif image format sued websites for hosting images in that format without paying for a patent license. The resulting fear did not get them a wave of compliance. Instead, the .gif format sank as web hosts rejected it. MPEG-LA, fearing the same, as declined to sue individuals over H264 "piracy." Not only that, they now went so far as to grant license for non-monetized use to essentially all individual users and non-monetized websites.

        If Microsoft sued even a single Linux user for reading files out of their own camera, milions of people would fear to ever buy another device using the ExFAT filesystem, and it would sink. Any individual fearing such lawsuits can buy cameras with cash so MS doesn't know who they are, and encrypt their filesystem so nobody can even tell what software they have installed. Might want to boycott Hotmail and 127.0.0.1 out microsoft.com and bing.com if this ever does get going, until ExFAT sinks. In short, the survivial on ExFAT is dependant on MS not suing individuals, probably on not suing corporate newsrooms either.

        Lastly, remember this: a Linux box bought with cash and no ID/discount card on which Windows was never activated, or which was bought used and off the record,is undetectable to Microsoft if their websites are not used and they are firewalled out in /etc/hosts. They would not even know who to sue with the exception of maybe some famous corporate photographer who did not appear in a list of Windows buyers. That lawsuit would guarantee them maximum adverse press, making people afraid to buy not only ExFAT cameras, but maybe all computer equipment. PC sales are already plunging due to Windows 8, a lawsuit about someone's home computer could sink Microsoft outright!

        Comment


        • #74
          I don't care what these entities do... but if it's available in the kernel I will build and use that module. Microsoft can go and ride the big baloney pony... they should have no right to patent things like that to hamper interoperability. It pisses me off because of all the good (open) filesystems we could have had for flash media (e.g. jffs2), we get that rubbish that has even worse fault tolerance than FAT32.

          Right now I wipe devices that come with exFAT and "mkdosfs -F 32 -s 128" for a FAT32 filesystem with 64 kb clusters. Works fine, even for 1Tb external drives (Windows won't let you create a FAT32 filesystem larger than 32 Gb, but it supports them if created with third party utilities), but then you can't have files 4 Gb or larger and it wastes storage space when there's "cluster overhang" on small files. (but that's not such a big deal). I need to be able to take that to any Windows computer and any Linux computer so FAT32 it is for now.

          I don't prefer that FUSE stuff, I would rather just "modprobe filename". exFAT is a simple enough filesystem.

          Take ntfs.ko vs. NTFS 3G in userspace for a relevant example for me. Different situation of course, but I would much rather have a good, fast kernel driver with good read support that lets me mount even damaged filesystems than the newer userspace driver that supposedly is safe for write support. That's rather useless to me, because it disconnects on i/o errors (if it mounts in the first place). At least the old ntfs driver lets me mount it, and I can "cp -R" and let it just fail to copy files stored on bad blocks that can't be re-read, and continue copying on errors.

          Comment


          • #75
            Very much sounds like this code was developed by Samsung, which I assume have a patent license from Microsoft. That means it's Samsung's proprietary code, unless Samsung puts a license on it which says otherwise.

            And sounds like the code was a leak somehow, so at the moment the copyright holders of the source code can just report to github for it to be taken down and it can't be included in the mainline kernel.

            On the topic of the patent, Microsoft is changing their position in the near future:

            "We also support the eventual inclusion of a Linux kernel with exFAT support in a future revision of the Open Invention Network’s Linux System Definition, where, once accepted, the code will benefit from the defensive patent commitments of OIN’s 3040+ members and licensees."



            After Microsoft's patent is part of OIN then possibly this repository could be used to get this code mainlined: IF Samsung says: we'll re-license our code under the GPLv2 or other compatible license for the kernel and the others who posted their changes on Github agree (if they are large enough to have their own copyright, one-liners are not considered a new copyright work I believe).

            On a side note: I see no reason DMCA needs to be involved directly if Microsoft wants it to be taken down, Github is owned by Microsoft these days.

            Edit: for it to be included in the mainline kernel, the code also needs to be good enough quality wise for the kernel.org developers to accept it.

            Also: INAL, but I did see some people post things I know a very likely incorrect.
            Last edited by Lennie; 05 September 2019, 06:34 AM.

            Comment


            • #76
              Originally posted by Grogan View Post
              I don't care what these entities do... but if it's available in the kernel I will build and use that module. Microsoft can go and ride the big baloney pony... they should have no right to patent things like that to hamper interoperability. It pisses me off because of all the good (open) filesystems we could have had for flash media (e.g. jffs2), we get that rubbish that has even worse fault tolerance than FAT32.

              Right now I wipe devices that come with exFAT and "mkdosfs -F 32 -s 128" for a FAT32 filesystem with 64 kb clusters. Works fine, even for 1Tb external drives (Windows won't let you create a FAT32 filesystem larger than 32 Gb, but it supports them if created with third party utilities), but then you can't have files 4 Gb or larger and it wastes storage space when there's "cluster overhang" on small files. (but that's not such a big deal). I need to be able to take that to any Windows computer and any Linux computer so FAT32 it is for now.

              I don't prefer that FUSE stuff, I would rather just "modprobe filename". exFAT is a simple enough filesystem.

              Take ntfs.ko vs. NTFS 3G in userspace for a relevant example for me. Different situation of course, but I would much rather have a good, fast kernel driver with good read support that lets me mount even damaged filesystems than the newer userspace driver that supposedly is safe for write support. That's rather useless to me, because it disconnects on i/o errors (if it mounts in the first place). At least the old ntfs driver lets me mount it, and I can "cp -R" and let it just fail to copy files stored on bad blocks that can't be re-read, and continue copying on errors.
              I agree!

              I consider FUSE is still a pathetically crappy bad joke even by these days, it's a lot slower and worse than native Linux filesystems. Even worse, there's lots of abandoned FUSE projects and are too buggy. FUSE needs to get extremely improved or develop a proper replacement. Despite the total discaster it is I like the transparency and easiness of using an userspace filesystem for stuff like networking filesystems: SSHFS, Samba, NFS, IPFS, etc.

              What about Microsoft allowing and actively supporting NTFS and Refs for Linux kernel? Any news on that?

              Comment

              Working...
              X