Microsoft, TomTom Settle Patent Infringement Battle (eWeek) [LWN.net]

Microsoft and TomTom have settled their patent dispute as reported by eWeek. TomTom will pay Microsoft an undisclosed amount of money to license the patents, while removing code covered by the FAT patents over the next two years. "According to Microsoft, the agreement includes patent coverage for Microsoft's three file management systems patents provided in a manner that is fully compliant with TomTom's obligations under the General Public License Version 2 (GPLv2). TomTom will remove from its products the functionality related to two file management system patents (the "FAT LFN patents"), which enables efficient naming, organizing, storing and accessing of file data, Microsoft said. TomTom will remove this functionality within two years, and the agreement provides for coverage directly to TomTom's end customers under these patents during that time." While Microsoft and TomTom say this is all GPLv2 compatible, there may be others who disagree.

(Log in to post comments)

Microsoft, TomTom Settle Patent Infringement Battle (eWeek)

Posted Mar 30, 2009 19:40 UTC (Mon) by JoeBuck (subscriber, #2330) [Link]

I suspect that it will. There's no incentive on TomTom's part to fight for functionality that it does not need.

Cute little devices

Posted Mar 30, 2009 23:31 UTC (Mon) by dmarti (subscriber, #11625) [Link]

TomTom devices are nifty, and a handy size to stick on your jacket pocket, or mount on your dashboard or handlebars -- but I would not want to deal with long filenames on one. If the UI is making users deal with long filenames on a tiny device, something is wrong.

Is U.S. the only country in the world?

Posted Mar 31, 2009 7:52 UTC (Tue) by khim (subscriber, #9252) [Link]

"Long filenames" are not the only thing VFAT does. Think about names of failes from just two or three letters. I mean Russian letters or, even better, Japanese ones.

Sorry, but so-called "long file names" have other benefits besides length...

Is U.S. the only country in the world?

Posted Mar 31, 2009 9:45 UTC (Tue) by Cyberax (✭ supporter ✭, #52523) [Link]

Russian characters take only one byte in Win1251 encoding which is usually used in FAT.

However, FAT is _still_ the main source of problems with Russian languages, because in addition to FAT we have also: KOI-8R, KOI8-U and UTF-8.

windows-1251 != IBM866

Posted Mar 31, 2009 12:53 UTC (Tue) by khim (subscriber, #9252) [Link]

Only IBM866 is used with FAT. Not windows-1251. But since it's not written in FS itself it's very unreliable. With VFAT you have full Unicode support and so codepage is irrelevant...

Too bad

Posted Mar 30, 2009 23:28 UTC (Mon) by bojan (subscriber, #14302) [Link]

So, TomTom chickened out. I can't really blame them, given the financial pressures and all.

Two ultimate piece of irony here:

1. Microsoft claimed this was not about Linux and open source at all. Why is GPL being mentioned then?

2. "Our car navigation patents, which are at the heart of the enhanced auto experience enjoyed by millions of drivers today, have been licensed to many companies, including leaders in the car navigation sector." What? The "bolt a notebook into a car" thing? Pathetic...

Too bad

Posted Mar 31, 2009 6:34 UTC (Tue) by man_ls (subscriber, #15091) [Link]

Companies extorted by such bullies as IBM and Microsoft should join a lobby and try to shoot down software patents ASAP. If they each put $1M I bet they could collect a sizeable bit of money, and if successful the inversion would pay itself in a couple of years. I bet a consortium of 200 companies in the industry would make a lot more noise than the FSF and similar organizations alone.

Is it time for a CONFIG_FREE_WORLD option in the kernel?

Posted Mar 31, 2009 9:35 UTC (Tue) by dion (subscriber, #2764) [Link]

Perhaps the best way to solve this problem is to create an option that can be used to only enable the patent encumbered features when in the free world, iow: Outside the US/Japan.

That way we can continue to enjoy a fully featured Linux with long file names on legacy file systems and merging of memory pages.

With an option like that only the citizens of unfree countries who fear a visit from the (thought)police would have to make do without the patented ideas.

I know there is a danger in working around unacceptable restrictions as that makes objections to those restrictions less vocal, but samizdat is often the only workable solution under an oppressive regime.

Is it time for a CONFIG_FREE_WORLD option in the kernel?

Posted Mar 31, 2009 10:02 UTC (Tue) by forthy (guest, #1525) [Link]

Given that any TomTom device has a GPS receiver, it could disable long filenames in the regions where Microsoft patents are valid - without user interaction ;-).

Microsoft, TomTom Settle Patent Infringement Battle (eWeek)

Posted Apr 4, 2009 0:20 UTC (Sat) by giraffedata (subscriber, #1954) [Link]

So now TomTom products will have less features and will cost more. What a fantastic demonstration of the power of the patent system.

Virtually every patent lawsuit is that same demonstration, and no demonstration is necessary because anyone who ever imagined a patent system saw it working this way.

People who support the patent system (in some form) look at a bigger picture.

Microsoft, TomTom Settle Patent Infringement Battle (eWeek)

Posted Mar 31, 2009 15:37 UTC (Tue) by etienne_lorrain@yahoo.fr (guest, #38022) [Link]

Just wanted to check if (what I think the patent contains without reading it) is implemented by Linux, so I did:

[root@fc10 ~]# mkdir /mnt/disk
[root@fc10 ~]# mount -t vfat /dev/fd0 /mnt/disk
[root@fc10 ~]# mkdir /mnt/disk/tmp
[root@fc10 ~]# echo "dummy file with long filename" > /mnt/disk/tmp/alongfilename.txt
[root@fc10 ~]# ls /mnt/disk/tmp
alongfilename.txt
[root@fc10 ~]# umount /mnt/disk
[root@fc10 ~]# mount -t msdos /dev/fd0 /mnt/disk
[root@fc10 ~]# ls /mnt/disk/tmp
alongf~1.txt
[root@fc10 ~]# umount /mnt/disk
[root@fc10 ~]# mount -t vfat /dev/fd0 /mnt/disk
[root@fc10 ~]# cat /mnt/disk/tmp/alongf~1.txt
dummy file with long filename
[root@fc10 ~]# cat /mnt/disk/tmp/alongfilename.txt
dummy file with long filename
[root@fc10 ~]# umount /mnt/disk
[root@fc10 ~]#

So obviously, there is two way to access a file with a long filename on Linux/VFAT: by its long filename (/mnt/disk/tmp/alongfilename.txt) and by its short filename (/mnt/disk/tmp/alongf~1.txt).
It seems that I would not miss the access by its short filename...

Etienne.

LFN fix for Linux

Posted Apr 3, 2009 21:48 UTC (Fri) by spitzak (guest, #4593) [Link]

Here's an idea:

Apparently it is ok to *read* LFN FAT disks. But not to create files with such names.

How about Linux change the FAT driver so when you try to create a non-existent file needing a long name, it writes the name is a whole new way (perhaps in a hidden file). This will avoid the patents. It will read both the new and old ways of doing LFN. Recommended that the implementation be given in BSD code so that everybody who wants to can implement it.

Any FAT disk will look fine in the Linux device. But when you plug it back into Windows they will see MICROS~ style filenames for all the new files and it will look crappy. They put it back on the Linux machine and it looks fine! Suddenly it is Windows that looks like the broken system.

LFN fix for Linux

Posted Apr 4, 2009 0:31 UTC (Sat) by giraffedata (subscriber, #1954) [Link]

Any FAT disk will look fine in the Linux device. But when you plug it back into Windows they will see MICROS~ style filenames for all the new files and it will look crappy. They put it back on the Linux machine and it looks fine! Suddenly it is Windows that looks like the broken system.

Nah, it'll look like the Linux device is broken. Files you generate with every device except the TomTom look fine in Windows, but the TomTom's files look crappy. Must be TomTom that's broken -- it doesn't know how to export files. And it must be because it uses that second-rate Linux thing.