OSNews

You bet this is big. While there is nothing new in a small guy being unable to go all-out-legal with one of the big guys and thus missing out on monetizing their own ideas, this particular thing will get tied into Linux and Free Software by the press. Which is an opening in a game and an invitation to a kick in the balls match. Users' balls, for all of the involuntarily involved.

I want to take a shower now.

I'd say the linux/os-freaks tie it to linux/os

99,99% of the people will think of small electronic devices and not software when they hear tomtom

OMG! They're acting like a ... corporation!


(Maybe this will encourage SD group to abandon FAT once and for all.)

Edited 2009-03-30 22:23 UTC

I am glad you said that. But I hope they do it.

Doesn't Windows have a mechanism built into its architecture where you could program support for other filesystems?

Yes. A few companies in fact have installable file systems for Windows. It's a daunting task to make a complete and correct filesystem, however, since there are a lot of intricate locking and pageability rules that must be obeyed by filesystem code.

Yep we have ext2 for windows: http://www.fs-driver.org/

Also some versions of Partition Magic came with drivers for various file systems (yep you could mount them in windows).

Which is probably exactly what they will do now. Either that, or simply use 8.3 file names on FAT.

Microsoft threatened TomTom with patents, and TomTom appear to have got out of it scott free. I would imagine that one of the few conditions on the deal is that Microsoft alone get to put the media spin on it.

Like Microsoft, TomTom is a business. If there was a way on the table to settle this dispute with no harm done to TomTom financially or TomTom's products functionality, then why would TomTom not go for it?

Edited 2009-03-30 23:16 UTC

I agree entirely with you.

I think that many people will answer that windows users are lazy and/or dumb to only use a device if it is not required to install any drivers to use it.

It is not true. Many devices only work in windows after installing their device drivers suplied by the manufacturer.

Why not use one of the many flash-optimized filesystems avalilable in linux and give to windows users a simple IFS driver to permit windows see the files ?

The manufacturers should develop a single, non-patented, free, documented and common standard for filesystems used in flash memories and send Microsoft to hell.

Edited 2009-03-31 15:55 UTC

That is what I am surprised about; specifically why we haven't seen FAT along with many other file system drivers being moved to FUSE instead of sitting in kernel space; not only would it solve some legal problems but also provide a more stable operating system over all.

What I question is after all these issues why FAT hasn't been replaced; what it appears is that those companies want to have a free lunch and avoid actually having to come up with an alternative which would involve them putting aside their petty differences in favour of coming up with something they could all agree on.

Lord knows that FAT is a horrible file system and limited due to the expanding size of flash devices. I have feeling, though, we'll have these very same companies embrace exFAT even with all the short comings that go with it because they would sooner embrace short term savings instead of focusing on the long term consequences of relying on technology which is critical to ones product.

Firstly, Microsoft's patents do not cover FAT. IBM has the patents for FAT, and IBM permits it to be used by all and sundry royalty-free.

Microsoft's (very dubious) patents cover LFN support within FAT.

Thank you for the correction - but there is no need to be aggressive in the tone of your post. Learn some humility when posting unless you want people to make nasty replies. It still doesn't change, however, that there needs to be a replacement for FAT. "

You are reading something into my post that was not there (or at least, not meant to be there).

There is no way to adequately express "aggressive tone" in plain text. There most certainly was not meant to be any aggressiveness or nastiness. I used "Firstly" and "Secondly" only because I had two points to make, and those words identified them. I'm not at all sure what it was that I typed that you took to be aggressive or lacking humility.

I apologise if I appear to be a touch sensitive, it is that my interpretation of aggression is based on the abrupt writing style which you employed in your post. The sense of abruptness is based on a lack of preamble and closure of the post.

I tend to probably read into text a lot more than most people because I tend to do a lot of reading and writing thus I pay careful attention to the language, pose, tone and style employed as to convey the nature of the topic.

Yes, he was not aggressive. but don't worry this kind of misunderstandings happen all the the in online communications.

And LFN is now an essential part of FAT. So this is a bad situation.

Firstly, Microsoft's patent is not for FAT32 per se. The patents are about a obvious method (that Microsoft just chose amongst several possibilities) to store long file names (which are an assumed feature of just about every other filesystem) in the FAT directory structure, while maintaining compatibility on disk with a computer without LFN support.

Secondly, the LFN patents are still very much in dispute.

Here is some non-Microsoft spin on it all:

http://www.groklaw.net/article.php?story=20090330181547642

I like this bit:


Edited 2009-03-30 23:58 UTC

This is from wikipedia, which is slightly more impartial then a pj quote

http://en.wikipedia.org/wiki/File_Allocation_Table#Appeal

so as of 2k6, Microsoft has two patents that have stood up.

http://en.wikipedia.org/wiki/File_Allocation_Table#Appeal so as of 2k6, Microsoft has two patents that have stood up. "

They stood up in respect of a review by the Patent Office. That is not particularly impressive, considering the things that the US Patent office allows to be registered as patents. AFAIK they haven't yet stood up in court in terms of suing anybody ... AFAIK such a case has never got to court.

FAT32 support is in the kernel because linux is used outside of the USA - we do not care (too much) about idiotic US software patents in Germany.

Also, the highest German court has ruled the FAT32 patent invalid for lack of a meaningful invention - see e.g. http://www.golem.de/0703/50844.html or http://www.heise.de/english/newsticker/news/86141

Here is another take on this from Groklaw readers:

http://www.groklaw.net/comment.php?mode=display&sid=200903301306553...

http://www.groklaw.net/comment.php?mode=display&sid=200903301306553...









Well, it is an interesting theory anyway that happens to go well with the facts.

Here is a story of IBM doing the same thing to sun http://www.forbes.com/asap/2002/0624/044.html. The goal is a kick in the face to the new company. The insane thing is that the whole point of patents is to protect the small time inventor from the big companies.

Practicality, I would guess, for all those flash devices that are formatted with it, and it's the only filesystem you're pretty much guaranteed to be able to read on just about every major os being used. They could move it to Fuse, but (and perhaps I'm missing something) I don't see how that would solve these legal issues surrounding the lfn patents in FAT32.
My question is why the hell are we still using that awful excuse for a filesystem anyway, aside from compatibility?

Nobody.

However compatibility and "interoperability" are very important. Until two years ago FAT32 was the only option for dual booters to share their data among Windows and Linux. (Now there are good NTFS drivers for Linux, and ext2 drivers for Windows).

Edited 2009-03-31 05:22 UTC

The answer is very simple: because the world isn't limited to the US and there are place which doesn't support software patents: Europe for example.

The world isn't limited to europe either, and there are plenty of countries outside of the US that don't support software patents. Anything code that is invalid under the gpl in large parts of the world has no business in the mainline kernel, how hard would it be to push it into another project like FUSE? If it was an opt-in thing, TomTom would have known the legal risks going in. As things stands now, this whole thing lends credence to FUD against linux; watch out if you use it as a platform because you have no indemnification against third parties. Sticking your fingers in your ears because it doesn't apply to you in your country does nothing but hurt the reputation of the platform as a whole.

But loading the vfat.ko module is an opt-in thing. For a casual desktop user, maybe not. But for a manufacturer of embedded devices? FUSE would simply have moved their legal problem from kernel space to user space.

Edited 2009-04-01 00:09 UTC

It is a matter of perception. When SCO went around saying that their IP was in linux, and were unable to prove it, it only made Linux's position stronger. Now when Microsoft says their IP is in Linux, they can follow it up with "... like what we sued TomTom over". That makes FUD about linux a hell of alot more compelling. If it were in a third part package, it would be clear to everyone that TomTom went into this knowing full well that they were assuming liability over code that was known to be in a gray area.

It is indeed a matter of perception. For example, my take would be that Microsoft sued TomTom with a lot of talk about "protecting IP", were surprised when TomTom said back "what about our IP?" and were so dismayed a little while later when TomTom joined the OIN that Microsoft pulled the pin on the whole schemozzle right away, and made a face-saving deal with TomTom.

http://www.out-law.com/page-9903

TomTom reports (via the royal Bank of Scotland) that there is almost zero finacial impact on TomTom out of the whole thing.

http://www.bolsamania.fr/analyses/analisis.php?origen=bolsamania_fr...

TomTom continues to use Linux in its products, in compliance with the GPL.

The net result is that the OIN has gained another member, and some more patents in its pool.

So how does all that, in any way, "makes FUD about linux a hell of alot more compelling"? It just doesn't. Microsoft fired their first warning shot across the bow, so to speak, and then ended up running away from the battle the moment the opposition snarled back at them.

Edited 2009-04-01 01:59 UTC

That is pure analyst speculation, so please don't assume it as being factual. It definitely can not be construed as something that "TomTom reports".

TomTom's own website
(http://investors.tomtom.com/analysts.cfm) states that RBS have a 'buy' rating on TomTom's shares, as such they have a vested interest in trying to 'spin' news on TomTom in as positive a light as possible.

The truth is no-one will know further details until TomTom officially reports their figures, and the granularity of the details that they do provide depends a lot on the regulations surrounding the individual stock exchange that the company is listed on (which in TomTom's case is Euronext Amsterdam).

That is pure analyst speculation, so please don't assume it as being factual. It definitely can not be construed as something that "TomTom reports". TomTom's own website (http://investors.tomtom.com/analysts.cfm) states that RBS have a 'buy' rating on TomTom's shares, as such they have a vested interest in trying to 'spin' news on TomTom in as positive a light as possible. The truth is no-one will know further details until TomTom officially reports their figures, and the granularity of the details that they do provide depends a lot on the regulations surrounding the individual stock exchange that the company is listed on (which in TomTom's case is Euronext Amsterdam). "

Agreed, no-one will know the details.

But I wasn't talking about specific details, but rather about the perception of the whole thing.

We had a tiny company (in comparison to Microsoft) threatened with a $10 million plus (in costs) lawsuit against a huge monopoly ... the tiny company joined the OIN and within a few days the lawsuit was dropped. With every appearance that the settlement amount was very nominal.

http://itmanagement.earthweb.com/osrc/article.php/12068_3812891_1/B...



So, what impressions does one come away with from this incident (given that we don't know the true details)?

One feels that Microsoft is trying desperately to spin it PR-wise as a victory, but in reality it looks a lot more like a very hasty retreat.

The other very obvious conclusion to reach is also fairly well stated in the article linked above:

Doesn't strike me that an IFS driver would ruin that. Aside from installing the driver, there's no other reason it wouldn't operate as the same mass storage device and couldn't expose the same underlying files.

Except that this would only work on Windows, and, assuming they'll choose something that _generally available_ under Linux, that would still leave MacOS out in the cold (for which even the TomTom Desktop software is supported currently).

Not to mention platforms like Haiku, Syllable, and many other OSes that have a different file-system driver architecture then Windows....

Bottom line: FAT(32) seems to be the most generally available file-system on the planet.....

I used IFS as an example, they could also write either a native filesystem driver for OS X or employ MacFuse towards the same end.
There's always UDF 1.02, which I actually have working quite well on my thumb drives once they're formatted properly.

Yes, *if* you really invest on research, you should be entitled rights to the outcome.

And MS spends in order of billions of dollars each year on research (like speech recognition, GUI usability, kernel development, etc). However I don't think LFN was a big part of it.

And there is the big problem of "interoperability". Nobody is trying to copy Windows 95 (the source of LFN), but they are trying to make their "own data" to be accessible to the "operating system of their choice".

So for example, it would be OK, if Microsoft sues somebody for ribbon UI, but NOT OK for making an OOXML importer for OpenOffice.

Basically agree with everything you have said above ... except the part about the ribbon UI. A GUI is just as much an interoperability protocol as LFN ... it is just in a different form. People learn GUIs, they develop skills in how to use them. Therefore, the GUI is (as implied by its name) an "interoperability" function between the program and the user.

As an interoperability function, there should be no problem with other parties making their own implementations mimicking that functionality. It must be their own work ... they should not get a "free copy" of Microsoft's GUI code or anything.

By way of an example, it is a bit like this ... you should not be able to copy a Harry Potter novel and sell it for your profit, nor should you be able to paraphrase a Harry potter novel and sell it for your profit ... but it should be perfectly OK for you to write your own fantasy novel about a magical school for young student wizards. There is a line in there somewhere that you should not cross ... it is the job of the courts to determine if a similar work is, or is not, plagiarism.

"
So for example, it would be OK, if Microsoft sues somebody for ribbon UI"

Isn't the Riboon UI just a tab interface with buttons on it? I'll admit I haven't used it much but that is what it seems like to me. Don't really see how one should be allowed to sue someone else if they use tabs with buttons on them.

Fair enough. The moral answer to this is: "do your own work". That does not mean that people should not be able to adopt the good ideas of others when they see them ... but each person should still have to do their own work to produce products for sale.




The GPL covers copyright only. You cannot just copy the work and sell it for your own profit ... you must do your own work.

With the GPL, however ... if you see a good idea implemented in GPL code ... such as, for example, the ODF file formats ... then you are actively ENCOURAGED to create your own programs implementing them. Fill your boots. Just don't copy someones code, then sell it to someone else.

A good rule of thumb is this: Imitation is a form of flattery, but plagiarism is just plain stealing.

Edited 2009-03-31 09:16 UTC