Or switch to using a filesystem other than FAT32?
Microsoft's prosecution of TomTom over alleged violation of patents is looking increasingly like a failure in its long-running policy of tying down Linux users through cross licensing of its IP. Computerworld has dredged up an email exchange with Microsoft's IP and licensing legal chief that explains Microsoft's got a long …
Isn't the tone of this story a little one sided.
So Microsoft are entering into these licence deals to stop these companies from distributing the Linux kernel? But hang on a mo, weren't Microsoft persuing such cases before GPL v2 came along? In which case wasn't the "Liberty or Death" clause actually introduced to try to stop distributors from entering into such deals with Microsoft, or anybody else?
Now I'm no fan of Microshaft and their tactics in fighting open source, or indeed any other competitor. However I do feel that the introduction of clause 7 was just as bad for anybody wanting to use Linux. It will in the long run be just as harmful to the development of Linux as the actions of Microsoft or any other company trying to license their IP. If the clause is enforced then potential commercial distributors will shy away from Linux and go for an alternative, and what alternative will they settle on? Windows of course.
An alternative to the view taken in the article is that the introduction of the "Liberty or Death" clause was short sighted and counter productive. It doesn't prevent Microsoft, or anybody else, from bringing these cases, so all it will do is hinder the development and uptake of Linux.
But surely, if TomTom choose to use software that is published under a licence such as GPL, that's not Microsoft's choice is it? Should they then turn around and say "ok, you've violated our IP, but since YOU'VE chosen to do that as part of YOUR business model, we're going to overlook it and let you use our IP for free?".
Whilst I'm not trying to suggest MS might not be fully aware of the implications of their actions, from a purely legal point of view I for one can't see why they should be obliged to ignore IP violation (if that is genuinely what has occurred here) simply because the other party *chose* to use Linux.
Or at least that needs to be more explicitly stated: If they were to cross-license and not be allowed to distribute their modifications in source form, they wouldn't be allowed to use Linux in their devices since GPLv2 requires that any source modifications be made available to end users should they request it. It's not just a case of them not being allowed to release code back into the ecosystem, since a ban on that would prevent them from modifying and distributing any GPLv2 software.
""It isn't a case of cross-license and everything is ok. If TomTom or any other company cross licenses patents then by section 7 of GPLv2 (for the Linux kernel). they lose the rights to redistribute the kernel *at all*."
In other words, Microsoft is eroding Linux and open source and slowing their development. "
Sounds like it's GPL not MS that is at fault here. It is the GPL agreement that is allowing any part of the code from re-entering the community. I know it's not politically correct to say, but let's call a spade a spade.
Hey, I hate M$ as much as the next guy but come on. Tom Tom is using an M$ patent that they have not licensed. What does this have to do with open source or linux?
And since when was Tom Tom distributing linux anyway? If they didn't want to use a proprietary filesystem, they should have used a non-proprietary filesystem.
"Make no mistake, this is intended to force TomTom to violate the GPL"
Indeed. Patents on software are all about lobbying to get special permissions to infringe copyright, to force others to infringe copyright, and to threaten others whose copyright is being infringed and who dare to complain about it, all to uphold some vague and unjustifiable "intellectual property" monopoly.
So you have the functioning rules governing the distribution of specific, concrete works being overridden by an instrument of monopoly and the mere suggestion that others are on the patent holder's turf because a note which reads "we thought of that kind of thing" was stamped by the patent office and a fee collected.
Microsoft: the Sheriff of Nottingham of computing.
In the US, the threat of software patents was crushed in re Bilski: http://www.fsf.org/news/esp-bilski
GPL V3 was written with software patents in mind. The Linux kernel is mostly licensed GPL V2, which has a clause that anticipated disasters like software patents, but was not explicitly designed for them. Ask your lawyers for opinions - they probably will not all agree with each other.
Assuming GPL V2 is strong enough to deal will patents, it does not prevent returning modified source code to the community. It prevents binary distribution of a kernel that includes material covered by patents that are not licensed for use by everyone.
If somehow the FAT patent is found to be valid TomTom must choose between buying a license for FAT that applies to everyone, removing FAT from any kernel they distribute or stop distributing Linux in their devices.
In reality, this is just nuisance litigation, and it is not well tollerated outside Texas. The fact that other have cross-licensed only indicates that they decided that was easier than disputing the validity of patenting FAT. It does not mean that any of Microsoft's patents are actually enforcible.
IBM lost lawsuits back in the 60s to stop plug compatible peripheral makers.
The law decided you can't block someone from making a clean room implementation.
Microsoft - remember you only have a business because somebody did this to the IBM bios.
What Microsoft are trying to do is to say, anyone can legally copy our plugs - but we have a patent on putting 5 pins at the top and 4 pins on the bottom of a 9Pin D.
If this passes then I patent putting an upper case 'M' in front of a lower case 'a' when spelling Martin.
Maybe Google ought to let Ballmer replace Billy G. as the more evil than Satan himself top knot.
Reg ol buddy, we need an evil Ballmer icon, unless you know something that we don't.
I suppose we could use evil Stevie as he kind of looks like Ballmer without the monkey dance or the sweaty pits.
I think it's clear that Microsoft would prefer to be as difficult to Linux and companies using Linux in commercial products as possible. If this results in TomTom not being able to use Linux, and needing to switch to some Microsoft embedded OS, this would presumably be fine by Microsoft. Linux would still be available as freeware, just not from people like TomTom. (Or TomTom could avoid using FAT if this is the only issue).
I've also not looked at any of the patents and have no comments on the merits of this case at all.
So Jeremy Allison is correct that Microsoft may be trying to put TomTom in a catch-22 situation ("can't distribute without violating the patent, and GPL forbids distribution under license"). But this doesn't mean that Microsoft or anyone else's patents don't apply to free software.
There is a real question here with three possible outcomes:
1. You may think all software patents should be forbidden. No doubt many patented items are obvious, but there are some valid non-obvious things that have been patented (the old Unix setuid invention, for example). Possibly also codecs. It's also not obvious why a hardware patent should not apply to someone's product just because they implemented the codec in software.
2. You may say that the patent law should say "free software should get an exemption from patent law". You'd need to change the law in this case though, and it's not obvious this would be justified, other than because it's not easy for Linux and the current GPL to deal with patents.
3. Or (the current situation) is that one day there is going to be some case where some free software is going to violate a strong patent, and the free software folks will need to find a way to deal with this.
Right now it seems many of the people advocating in favor or option 1 (no SW patents) are really doing so because of the reasons in (2), not necessarily (1).
It is indeed surprising that Microshaft's secret dealings have not been leaked before.
I, for one, do not like Microshaft's duplicity and double dealing and do not want to be 'owned' by them. Neither do I want to live in a constant state of suspicion, thinking or fearing the worst of people, all the time. Unfortunately .........
Separately, we are reading that the EU have decided that they can relax their supervision of said Microshaft when, without this belated revelation, they still seem to be using (abusing) their dominant monopoly to extend dependence on their products.
Correct me if I'm wrong, if this latest revelation is true, then Microshaft may have deal a really damaging blow to the GPL, with the collusion of some significant Linux commercial companies who now appear to have sold out. .
Tom Tom are not well placed financially to defend this action. Notwithstanding that they are not a community player, they are going to need some considerable assistance if they are not going to cave in. Groklaw to the rescue!
By the way, where are Red Hat placed following their recently reported secret deal.
I've said it before, in many contexts, and I'll say it again. I would like to leave a decent world to my children and grand children. Without being too pessimistic, that seems to be rather optimistic.
PS It looks as thought this particular revision to the GPL, section seven of GPLv2, is potentially an own goal.
But why not shun FAT altogether and use an open source file system?
Can someone please enlighten me as to why switching to an open source file system is a problem?
The open source community should not be incorporating MS patented tech in their solutions anyway. MS are, well we all know what they are and they should be shunned/bypassed/outcoded at every opportunity by the open source community.
Microsoft did not get to the position they are in by altruistic practice. Quite the opposite.
Excuse me for being dumb, but why not just stop using FAT32 and use ext3 (or one of the other file systems) - TomTom could include a driver for the chosen FS as part of their 'Home' software package, or even include it on a separate drive built into the TomTom itself.
Paris - 'cause it seems soo simple, yet probably isn't!
Presumably patent 5579517
They should look at the Mac circa 1984, the Mac handled long file names by actually holding only a short filename. The long filenames you saw (255 characters) were actually stored separately and the shorter filenames used to access the file (32 characters from my memory).
"Yes, other companies have signed FAT patent licenses, both in the context of patent cross licensing agreements and other licensing arrangements."
i.e. nobody has specifically licensed that patent and even if they had, herd mentality is no excuse for agreeing to a patent without a 'novel' invention in it, which was filed 5 years after it was first proposed in printed articles and whose intention is to suppress competition and they were not the inventor of this technique.
I tell you, it's very simple. Microsoft saw that Tom Tom had a bad quarter, they wrote down their Tele Atlas buy, they need to trim some staff to bring their profits back. Tom Tom announced that they may breach their lending covenants, which typically would include a clause saying 'extra capital needed if share price falls below X'.
Tomtoms share price is down because people downgrade motor manufacturers (forgetting that aftersales, add-ons, spare parts and repairs of cars go UP not DOWN when new car sales go down because people try to get more out of their existing cars).
Tomtom announce the quarter and Microsoft is ready to attack, in the hope of further driving the price down. Microsofts attack was 3 days after the bad figures from Tom Tom.
But it's a bullshit claim.
So essentially MS is moaning about the usage of FAT in TomTom's devices. Don't know what the compelling reason was to use an antiquated file system with seriously restrictive 'options' in the first place, but suspect it was accessibility for windows users to data on TomTom devices.
Does this mean that if TomTom starts to use ext2/ext3 file systems, then makes sure that a windows ext2/ext3 driver is supplied with their software so windows users can actually access data on TomTom devices from their windows PC, the problem will essentially go away?
I fail to understand how TomTom are in a two way either/or situation here - there are clearly three choices: switch to 100% MS, run into GPL problems or the third way, switch to 0% MS-licensed technology. If Microsoft are trying to blackmail them as the article suggests, TomTom should quite clearly explain to Microsoft exactly where they can shove their big FAT disks and tell them to Foxtrox Oscar.
There's no reason why TomTom can't run their sat-nav's internal storage on a Linux file system and the same goes for SD storage cards - format them with a non-MS file system and provide Windows users with a driver to interface with it rather than relying on mass storage device. It'll be a bit of a pain for a small amount of users who like to tweak their kit but most wouldn't ever notice that their TomTom was using a different driver.
Patents were created to increase the rate of technical progress by providing incentives for inventors to describe there inventions to others. Modern patents can only be read by patent lawyers - partly because they are written in patent jargon - but mostly because of the damages for willful infringement. If an inventor reads a patent (fails to understand it) and infringes it, she is liable triple the damages she would have to pay for infringing without reading it.
The world would be a better place if the entire patent system was abandonned. Patents on CODECs are particularly nasty. Many CODECs have features chosen to infringe as many patents as possible - even at the expense of making the CODEC less efficient. The purpose of this is to create a cartel of companies who have cross licensed that can prevent any new player from entering the business.
For all those newbies thinking clause 7 is an own goal, it was chosen explicitly to prevent a problem. Some people distribute software under variants of the BSD license. These people often choose that license because it gives them evidence that the educational grant (tax payer's) money they received is being turned into software that has commercial value (tax revenue). Good for them, but the rest of us do not have that revenue stream.
Programmers pick GPL so that their software cannot suffer the familiar embrace, extend, extinguish cycle. BSD software can be incorporated into closed source products. Extensions can be added so that file formats and network protocols become incompatible. Victims of the closed source software are then locked in because they cannot move their data back to competitive software products. The GPL requires Microsoft to either make their own complete implementation, or to make the source code for extensions available to all.
The fact that Microsoft cannot get a free ride from GPL software drives them batty. Hence the old GPL is unamerican/unconstitution/comunist/... rants. The new plan is to make loud threats with unenforcible patents in the hope of getting people to switch from GPL to BSD licenses (SCO version 2?). No-one with a clue takes them seriously.
Alternatives to FAT:
The only reason why TomTom, digital cameras and so forth used FAT was to make life easier for Microsoft's users. Some kind person has written an ext2 filesystem driver for Windows (http://www.fs-driver.org/) so Windows users will not be left completely in the cold. If necessary, Linux devices can pretend to include a CDROM and CD containing Ext2fs.sys or a webserver, so Microsoft users can install Ext2fs.sys from a Linux device before Ext2fs.sys is installed.
Then to be fair everyone anywhere who makes flash/ssd devices and sells them preintalled with FAT/LFN would have to pay MS something - or would you be happy to see every one of these devices come unformatted?
OH! And then they'd charge TomTom for some pathetic patent they have on 'how computers might possibly talk to each other for the first time in decades now that MS has patented the Bleeding Obvious again"
First and foremost, this will only affect TomTom in the US. The only other country in the world that supports software patents is Australia.
Software patents will ultimately destroy Microsoft, or indeed any other company that pursues them. It will also cripple the software market for that country, as no-one will in fact be able to innovate new ideas, as they will infringe on some patent or other.
Also, Microsoft fanbois and shills, in their utter ignorance, equate "Software Patents" and "Intellectual Property" as the same thing. IT IS NOT .
Microsoft muddied up these waters by using the terms "software patents" and IP interchangeably in statements. When they were talking about patents, they dropped in IP, when they were talking about IP, they dropped in patents. THEY ARE NOT THE SAME THING.
All software is protected by copyright, yes, even open source. This is the protection for intellectual property, it needs nothing more.
I can buy a patent for putting a question mark at the end of a question.
I cannot buy a patent for inventing the question mark.
If I invent the question mark I can claim copyright on it.
Even if I invent the question mark, and people recognise I invented it, I cannot claim patent on using it, if someone got there and bought the patent before me.
Like I said, muddying all this up only causes confusoin, and also will tie up ALL, (not just open source), software development in the future in the US, and, in effect, the US will be left behind when the rest of the world marches on, with their own FREE and unencumbered software.
How secure is MS's patent in this respect anyway?
I can't for the life of me remember what it's called but there is a principle in IP law that states that you can't enforce a patent that is designed to fit a product (like you can't enforce the notion of a jacket).
This was used against Nokia when it tried to protect it's patent on snap-on covers.
Surely if the codebase for FAT support doesn't come from Microsoft, then it is merely made to fit an MS product and the patent isn't enforceable.
Or am I missing something?
To those questioning the need to support FAT file systems: Since most memory stick/ thumb drives use FAT as far as I know, doesnt this also attack desktop Linux systems?
BTW I would have thought it should be easy to support ext3 or whatever on windows, just insert an auto run exe file like they do on USB 3G dongles, and that capability is added (I've tried it as a sort of semi security for USB drives)
All GPL restrictions are on a per-compiled code segment. All the FAT drivers are kernel modules, which can be discretely compiled. With correct use of initramfs, or even initrd it's never necessary to compile drivers directly into the kernel unless it's convenient. So all this burble about restricting code publishing is, in my opinion, absolute rubbish. They, or any other compay already operating under such deals, won't be able to publish changes only to code covered by the agreement, which is a barely noticeable amount and what MS've admitted to contesting so far has all been shown to be legacy, static, modular code. Nothing to need any changes there, so nothing to publish anyway. As for actually using FAT ... well that's already been well covered. Since all their updating is done via USB anyway, it's totally unnecessary ... but the fact that they *have* done it left them vulnerable to the case. Muppets. It's not as if Microsoft hadn't been attacking the camera manufacturers over FAT for long enough to work out what would happen eventually. Join the queue of licensees or fight; the same choice everyone else has had.
Why are we all still using FAT/VFAT? UDF was designed for just this sort of thing, being read/writable everywhere (or it should be, although my OS doesn't have newfs_udf yet - come on, guys, NetBSD has it), having the additional advantage of being an open standard. I think what confused everyone was the OSTA punted it as a replacement for ISO9660 rather than a good general purpose filesystem ideal for most high-capacity removable media.
Front the article:
"...In other words, Microsoft is eroding Linux and open source and slowing their development..."
'GPLv2 is a stupidly naive and ill thought out millstone around the necks of people living in the real world an needs to be scrapped'
I hate FAT with a passion but ext2/3 or UDF aren't the answer.
I'm running ext2/3 drivers for WinXP and the best I can say is it works. Its slow, unpredictable (XP can't seem to workout how much free space is really available) and Linux needs to fsck it every time - its not fun and not ready for ordinary users and probably not going to get better till Microsoft improve their support for installable drivers. This isn't immature code, its the official source, the problem is the interface to Windows.
And UDF as a file format? On embedded devices with bugger all RAM for the file table let alone buffering? Total non-starter.
"2. You may say that the patent law should say "free software should get an exemption from patent law". You'd need to change the law in this case though, and it's not obvious this would be justified, other than because it's not easy for Linux and the current GPL to deal with patents."
Tomtom do not make free software, they use free software to sell their devices and services. Therefore they should not be exempt from anything and should pay a license for FAT. M$ may be taking advantage of their poor market position, but then Tomtom chose to use FAT even though the patent issue has been around for years.
Tomtom uses FAT becasue the majority of their customers are Windows users, and the majority of the Tomtom devices still use a conection to a WIndows PC to download map updates and upgrades. The majority of those users are roadwarrior types using their Tomoms for business trips, ie for commercial reasons. I don't see Tomtom giving their devices or IP away for free, so why do people expect M$ to? Tomtom need to think of an alternative that still allows simplicity for their Windows users, or pay to use FAT. Seems pretty open and shut to me.
"So Microsoft are entering into these licence deals to stop these companies from distributing the Linux kernel? But hang on a mo, weren't Microsoft persuing such cases before GPL v2 came along", Gareth Jones
They have made claims in the press the Linux violates their patents but I can't recall them actually going in to court on it before. What specific MS v GPL cases are you referring to. And in this case they don't specifically state which Linux kernel or what third party code violates the patents.
"I do feel that the introduction of clause 7 was just as bad for anybody wanting to use Linux .. It doesn't prevent Microsoft, or anybody else, from bringing these cases", Gareth Jones
GPLv3 clause 7 says that no additional restrictions can be imposed on the recipients’ and that any other "further restrictions” may be removed. And the case doesn't apply here as Tom Tom released additional drivers under the GPL 2. Cases like this being of the kind GPL 3 was designed to address.
"I for one can't see why they should be obliged to ignore IP violation (if that is genuinely what has occurred here) simply because the other party *chose* to use Linux", anonymous coward
Tom Tom uses third party and GPL software in its products. Neither of which are owned by Microsoft. Therefore it owes nothing, nada, zilch to Redmond ..
Hey, I hate M$ as much as the next guy but come on.", Anonymous Coward
I don't understand how someone can hate a company. It's just a device for converting base matter into shiny objects .. :)
"Tom Tom is using an M$ patent that they have not licensed. What does this have to do with open source or linux?", Anonymous Coward
Tom Tom isn't using a M$ patent, Tom Tom is using third party and GPL software. The real question is what does an old long-file-names-for-fat patent got to do with Open Source. And since neither Tom Tom, the third party company or the Linux kernel team would have ever read this M$ patent, what has Microsoft got to do with Open Source.
"And since when was Tom Tom distributing linux anyway? If they didn't want to use a proprietary filesystem, they should have used a non-proprietary filesystem", Anonymous Coward
They used a mixed Linux kernel third party solution. Redmond is appariently upset that peopel are selling GPS devices without paying them for the 'innovation'.
And with the number of much older prior art examples now surfacing, it looks like they are invalid in the US too, as they probably were in the first place. Looks like a new anti-trust case on the way - at least in Europe where their tie-in, competition stifling agreements will be very illegal.
" All GPL restrictions are on a per-compiled code segment", Matthew Joyce
I'm sorry but could you elaborate further on where n the GPL text it addresses restrictions on 'a per-compiled code segment', or could you tell what that even means.
"all this burble about restricting code publishing is, in my opinion, absolute rubbish. They, or any other compay already operating under such deals, won't be able to publish changes only to code covered by the agreement", Matthew Joyce
An agreement to sell devices that utilize GPL code without distributing the related source code is a violation of the GPL.
Precisely - it's an attempt to get around the fact that you cannot copyright a fileformat. So they patent the implementation of the format.
What would be interesting is if you could claim that UAC in Vista was preventing a user installing an alternative FS river - then you might have some interesting monopoly infringement.
"So Microsoft are entering into these licence deals to stop these companies from distributing the Linux kernel? But hang on a mo, weren't Microsoft persuing such cases before GPL v2 came along? In which case wasn't the "Liberty or Death" clause actually introduced to try to stop distributors from entering into such deals with Microsoft, or anybody else?"
The GPLv2 dates back to 1991, and I don't think Microsoft was quite so interested in coercing others into patent pacts before then. The GPLv3 does attempt to strengthen the protection against the injustices of patents due to the Microsoft/Novell deal which offered patent indemnification to the recipients of GPLv2-licensed code.
"An alternative to the view taken in the article is that the introduction of the "Liberty or Death" clause was short sighted and counter productive. It doesn't prevent Microsoft, or anybody else, from bringing these cases, so all it will do is hinder the development and uptake of Linux."
But the FSF don't care about popularity: it's about the freedom of the end-user. Obsessing about being the most popular doesn't come into it - otherwise, everything would be permissively licensed or released into the public domain (where possible).
"Sounds like it's GPL not MS that is at fault here. It is the GPL agreement that is allowing any part of the code from re-entering the community. I know it's not politically correct to say, but let's call a spade a spade."
Here's an explanation as to why you have something like section 7 of the GPLv2. Let's say you're distributing code under the GPL and a patent aggressor informs you that their "intellectual property" is being infringed by that code. Let's say you cave in to their demands and buy a patent licence and keep distributing that code. Now consider the position of the recipients of the code: can they distribute the software or even use it without a patent licence? What about the people who obtain the code further down the chain?
If Microsoft had their way, they'd be able to divide the recipients of your code into two camps: those authorised by Microsoft to use the code, and those who haven't paid their tax to Microsoft and who are not allowed to use the code. Clearly, this undermines the licence governing the distribution of the code under the rules of copyright, effectively letting Microsoft veto large tracts of that licence.
So, no: it's not the GPL's fault in this case, because if there was no section 7 then the only "re-entering the community" possible for the code would be via the orifice defined by Microsoft.
Again, Microsoft are just looking for new rules which let them off playing by the existing, functioning rules defined by copyright.
The idea that this action is going to stop TomTom from using Linux is as silly as suggesting that this is Microshaft's intention. TomTom need not enter into a licensing deal with MS even if they loose the case. They can simply switch to another FS in future and pay Microshaft compensation for the devices sold so far.
As such they will not violate GPL. I suspect that rather than being an action intended to damage Linux this is more likely to be a revenue raising exercise, after all MS have been hit by the recession and need new revenue streams. TomTom have been using their file system for some time now, how much money would MS make if TomTom ended up paying just a few dollars per device sold so far?
IMHO the system for such actions needs to be changed. Their needs to be a time limit to such actions being brought. If the plaintiff does not begin their action within, say, 12 months of the device or software going on sale then they should forfeit any rights to licensing or compensation for the IP used. MS could have brought this action years ago, but probably decided it wasn't worth it going after such a small company. TomTom are now much bigger than they were and MS revenues are down, MS have probably been going through old paperwork and thought "That one could make us quite a bit of money now." I doubt they are really after a licensing deal, just a few million in compensation.
"How secure is MS's patent in this respect anyway?"
Microsoft could spend millions of dollars per day forever for lawyers; tying up things in courts for years makes any patent pretty "secure", no matter its actual legitimacy. Look how long SCO kept things going with a fraction of the legal resources.
If they licsence FAT for use on there hardware, then there binding the IP to there hardware legaly and not impossing it to there software. now if they liscence the IP to be used in there software - THEN they fall foul of there GPL, otherwise if they do open source software thats for all effect open, then there good. But Liscensing any IP thats proprietary to be used in that software instead of tieing it to the hardware will make then open to Microsoft abuse.
Now what is funny is all this is realy down to the camera manufacturers etc who were lazy and just thought FAT was as good as open source, Microsoft let them build there houses befroe they pointed out the flaw and only then after they had built many houses. Why point out a flaw early on when you get nothing from it when you can be lazy and point it out later and make money from it. Now its that type of situation that realy does point at evil IP trolling or are they just lazy in there RnD from the aspect that they never checked existing patents out they were using.
What I find realy crazy is that alot of companies tend to go for the one shoe fits all approach and with FAT your very limited at the filename level - hardly the direction for a multi-language one shoe fits all stance companies like to adopt for savings in costs now is it.
BTW this has nothing at all to do with Microsoft trying to block Linux in any way, its more case of common sence, or lack of. Besides technicaly its the tighness and well thought out GPL 2 thats casing the issue here, but thats what you get when you think things thru properly as without that clause they would after all be opening up people to creating a market for some patent troll to cash in on later at your expense, just like microsoft is doing now. So GPL 2 actualy prevents this kind of mess in the first place, shame it wasn;t around when camera companies started embracing digital media along with the likes of mp3 players.
If Microsoft's patents are valid, *no-one* can legally distribute Linux, whether Tom-Tom cross-licenses them or not. If the patents aren't valid, then the cross-licence is based on a falsehood, and probably unenforcable. It depends on the exact terms of the cross-licence, but it's not clear that Tom-Tom's position is different from anyone else's.
I wouldn't be surprised at all if this patent, being dated long after Macintosh first implemented long filenames, includes not only FAT's implementation of LFN, but all LFNs in general.
As such, simply switching to Ext2/3 would not solve the problem, you'd have to switch to a file system without LFN support. Ofcourse since Macintosh was there WAY before Microsoft, this patent is moot anyways. Fight it in court and sue MS for expenses and damages.
This, incidentally, applies to the vast majority of Microsoft patents. The problem with (software) patents is that there are no checks being applied whatsoever at the patent office, and many companies cannot afford to fight a legal battle against a patent troll or large corporation like Microsoft.
The solution is not the abolishment of patents in general, it's becoming a lot more picky on what patents to accept and importantly, implementing a system of actively challenging patents, so that when a patent troll patents something that should be common sense, or has prior art, someone can step up and have the patent kicked out, all expenses in that case should ofcourse go to the patent troll.
It doesn't even need to be decided by a court of law, the vast majority of these false patents are blindingly obvious to those who know what they're talking about and easily demonstratable (Like the case of the Macintosh LFN implementation, just look at the changelogs from way back then)
> Patents were created to increase the rate of technical progress by providing incentives for inventors to describe there inventions to others. Modern patents can only be read by patent lawyers .... mostly because of the damages for willful infringement. ... Patents on CODECs are particularly nasty.
I mentioned the example of a codec because it points to another reason for patents beyond "getting inventors to disclose". Namely the case where you have a pure research lab which spends a lot of time and money on research to create a codec with good compression, in which there's merit in a system that allows them to get paid for their research. In other words, either of the following would be bad:
1) Pure research lab that does not make products itself -> no revenue. Meanwhile product companies get rich without paying them.
2) Company A does R&D and sells product. Company B freeloads off company A's research and therefore sells the product cheaper, undercutting A. (This one is the reason that drug companies are so much in favor of patents.)
The point is that the justification for the patent system is to make sure the inventor can get "fair payment" for their invention and/or R&D. And the "incentives for inventors to describe their inventions to others" are the price that society imposes on inventors as a quid-pro-quo for this, not necessarily the original goal. Microsoft may not necessarily need the money, but some others do.
It's true that lots of people don't read patents though. I did say I hadn't read these specific patents, after all :) Stuff still gets disclosed in technical journals though, which people do read.
> For all those newbies thinking clause 7 is an own goal
You may not have aimed this comment at me, but I'm not suggesting it's an own goal - it's there for a very understandable reason. But the catch-22 problem can still arise, and it's not the patent holder's fault. The FSF just decided they preferred this all-or-nothing situation instead of a lot of GPL software where the freedoms weren't really usable because of patents the original author knew about.
seems to be making the point (at http://blogs.computerworld.com/linux_companies_sign_microsoft_patent_protection_pacts) that MSFT has done licensing deals with 18 companies, under NDAs, that puts these 18 companies in breach of the GPL.
Whilst it may well be that the FAT patent can be evaded by using another fs, there is still the problem of existing breaches of the GPL by persons unknown.
To start... IANAL
Now to continue:
People saying that GPL Section 7 is at fault. Was the intent of that section to stop law this type of law suit by saying If you use patented software you can't legally release all the code as per GPLv2 rules? or am i missing something?
Would it be much of a hassle to set up a firmware upgrade to set the device to use EXT2 file system instead of FAT? That way they could have the "Closed Sourced" current FAT based system or a "clean source" version using GPLv2 compliant code and EXT2 filesystem.... and let the Users decide which they use...
Wasn't there a case involving M$ and specifically FAT in the recent past? Like 2-3 yrs back? That M$ didn't win because FAT was considered to be "too widely available".
Seriously, going after a GPS maker for a FAT issue is lame. That _is_ just nuisance noise and it certainly doesn't improve my opinion of M$. And I say that as someone about 50/50 on Linux and Windows (Linux @ home/ win @ work).
I think M$ is threading a fine line between protecting its commercial interests, which it is entitled to. And, again, coming across as a monopolistic bully, thus ruining its PR. With Office 2007 and Vista under its belt, you would think that its energies would be better occupied fixing its products.
An excellent idea, but only if there is some agreement on who provides the file system driver. If every widget provider abandons FAT and starts shipping their own code for (say) ext2, I foresee misery for any consumer who buys more than one such device. Like PDF, you wouldn't need to agree to a *sole* provider, but the end-user would have to have the final word on which code they used. Such cluefulness on the part of hardware vendors is probably too much to hope for.
Incidentally, *is* there already ext2 or ext3 support in Windows? (3rd parties, obviously)
There are facts and there are fairytales in this discussion.
1. The idea that TomTom can just use Ext3 etc is ignoring the fact that they *currently* use Fat. If an embezzler stops taking money it does not wipe the slate clean and retribution is still expected - with patents this is done either by a license or cash retribution.
2. Whether you agree with software patents or not does not change the fact that patents are currently enforceable by law.
3. Microsoft has been enforcing its IP, including via patent protection, since before GPLv2 (1991) but as M$ had only a hand full of patents then it was other companies that this clause was likely aimed at.
4. Section 7 of GPLv2 states "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." and seems to be directly aimed at making people choose *either* GPL or commercial - no mixture of both. This is a double edged sword - it may be as likely to force people away as embrace OS.
Personally I believe in patents in theory - just not the idiotic granting that has occurred and with maybe a three year protection window. The fact that an algorithm is patentable in hardware but not software would be ridiculous - but the fact that you can patent "sending a bill by email" is a farce.
A good description on the patent explosion is at http://weblog.infoworld.com/openresource/archives/2007/06/microsoft_was_o.html
Microsoft is trying to suffocate Linux, pure and simple. You can go on about the legalities of this, that and the other but this is what it all boils down to.
Linux is superior to Windows, fact. Microsoft knows this, fact. Linux is gaining market traction, fact. Microsoft has spent billions through various methods (buying politicians, ISO etc) suppressing Linux, fact. This business with Tom-Tom is merely the tip of the iceberg, fact.
Microsoft apologists are ten a penny. Most make excuses for their beloved Redmond masters because they have something to lose if Linux takes an even bigger bite out of the Windows apple (LOL!).
Make no mistake, Microsoft can deny all they like that this is about destroying a legitimate competitor. But that is exactly what it is, and no amount of hyperbole can disguise this.
Reading the comments, I still don't get what FAT is in use. Yes, Windows PCs can read FAT as the lowest common denominator. Whoop-dee-doo. But just about any USB device I have tried to use on Windows has wanted some driver or other to be installed, and a GPS unit is not a "dumb memory stick". I don't think users will freak out if they have to plonk in a CD or somesuch to get it going.
And, as others have said, the GPS unit could hold a dummy CD of its own to bootstrap the process. Whilst this may be technically hard to implement, it would solve the legal issues.
The divers (say for ext3) could be made freely available and, in Europe anyway, the EU could compel MS to pre-install them as part of the monopoly investigation.
Perhaps this is a symptom of a fundamental shift. A typical sat nav is starting to become much more than a sat-nav... Time to strike before it's a missed opportunity... I think FAT is used to encourage people to put their mp3s / personal files / funny characters on them as easily as possible... Probably ms didnt care when there was no market overlap.
I wonder - maybe one day TOMTOM will take BMW to court and say "they shouldn't bundle ONLY BMW satnav in a new BMW... It's anticompetitive!"
And hence the floodgates open - fingers in pies, on the double.
Given that the other patents are so crap that frankly they're only good for compost (Patent for a computer in a car that runs an "open platform, multi-tasking operating system"... I mean that's just crap. You might as well claim a patent for putting computers in pink boxes), they certainly don't show any real inventiveness or novel purpose, I think the only conclusion that can be drawn is that MS are going for Linux and trying to kick TomTom in the teeth because they had the balls to stick up against MS rather than just rolling over and being another of Ballmer's lapdogs.
If you're not worried about "tainting" yourself as a GPL developer by reading the piles of shit that these patents are, you really should go and read them, and when you've picked yourself up off the floor and wiped the tears of laugher (or despair) from your eyes just ask yourself what the hell the US Patent Office was doing granting them - it would be like granting King Camp Gillette a patent on how to shave.
The implementation of FAT in the kernel is GPL. TomTom could create a new implementation under a different license and distribute it as a closed source kernel module. There is little point in doing this as Microsoft would still demand royalties for unenforcible invalid patents.
@facts and fairytales
1. Big difference between wilful and accident infringement.
2. Software patents can be purchased almost anywhere. In the US, software patents were seriously defanged in Re Bilski. In the EU, you have to call 'software patents' 'computer implemented inventions' or they will be rejected. Judges have caught on that a 'computer implemented invention' is a software patent, and should never have been granted. Even if there was not prior art, the FAT patent would unenforcible everywhere (except perhaps Australia).
3. IP is not legal term. There are patents, trademarks and copyright. The term IP is used by people who do not know any better and those deliberately trying to misapply the reasoning behind laws for physical property to patents, trademarks and copyright. If you have a point here at all, be very specific: what patent, what date, and against whom?
4. The GPL has always given you a choice: Cathedral or Bazaar. The only new thing here is that Bazaar is now popular enough to hit Microsoft's profits hard.
@Charles Smith, @Remy Redert
The patent is about long file name support that is compatible with a previous file-system that was limited to short names. There is still prior art. Not a problem to ext2/3/4.
SCO only achieved one thing beyond their own destruction: They demonstrated that caving in to nuisance litigation is more expensive than standing up to empty threats. Also: pay of one troll and you will get whole horde of them at your door tomorrow.
> The point is that the justification for the patent system is to make sure the inventor can get "fair payment" for their invention and/or R&D.
The patent system fails to reward inventors. If you invent something, manufacturers can wait twenty years to use your idea for free (popular tactic for the car industry). Manufacturers can use your idea and be confident that you cannot afford to defend your patent - it is really expensive to defend a patent, and all the time you are trying, the manufacturer has a revenue stream from your idea which he can use to delay the litigation.
The most effective way to profit from an idea is to sell your own product. You get at least one and often two years before competitors get something to market. During that time you have a revenue stream for researching a second generation product, so you can repeat the cycle before a patent can even be granted.
There are ways to profit from patents: You can lie about the existence of a patent and get it included in a standard (Remember Rambus and JEDEC?). You can patent other peoples ideas. If any of them turn out to be profitable, you can the sue whoever did all the work to change an idea into a revenue stream. If you are part of a cartel of companies with cross licensed patents for a communications product, you can prevent new entrants into the market by updating the protocols to use new patents whenever old patents expire.
And as for drug companies: the patent system ensures that poor countries that cannot afford patented drugs and become sources of infection for people in rich countries. It would be far more sensible to reward drug research directly than to bump up the global price of medicine.
Out of all the thousands of patents filed in the last year, can you name a single one that should have been granted, earned money for the true inventor, was not submarined into a standard and will not sicken thousands of people in the third world?
"If TomTom or any other company cross licenses patents then by section 7 of GPLv2 (for the Linux kernel). they lose the rights to redistribute the kernel *at all*."
Microsoft lost a law suit with SUN and SUN has a cross-licensing deal with Microsoft.
TomTom can just use Solaris or OpenSolaris FOR FREE and avoid the whole legal issue... since Solaris is not under the same license restrictions.
Ok, I've used ext2 drivers under windows - just for the odd peek at my linux partion or to migrate files between the two systems and it seems to work ok, but that isn't the point.
In-car GPS's also often do other things, like being an MP3 player. You might install a driver when you install the tomtom software, but what about when you try to copy files from another PC - crunch, "tomtom can't do that" is the perception. Then you decide to use your SD card in your camera. Oops, that doesn't work either.
MS doesn't have to stop linux completely, it just has to make things more difficult than using windows. MS doesn't even have to win the patent dispute. If other device manufacturers know MS may show up on their doorstep claiming royalties for one of the other "infringing" aspects of linux, they may as well just license windows and avoid the legal costs.
Of course, whether linux needs tomtom or not is a separate matter. However, I find it offensive that MS or any other organisation is allowed to use these sorts of strategies to strangle their competition.
I hope tomtom fights this all the way and shows MS up for what it is and generates lots of publicity from it.
Google is your friend.
http://www.fs-driver.org/ works and is a simple install (not even a reboot -- isn't that blasphemy for Windows installs?)
For you MacTards (of which I am one, seeing as how I'm posting from a MacBook), I haven't tried this yet but http://sourceforge.net/projects/ext2fsx/ looks promising. I plan to give it a try right away.
JFTR there's a commercial driver for HFS+ under Windows called MacDrive, and Linux seems to have drivers (at least Ubuntu just mounts them without me doing anything special), so there's another possibility. But not a free one for Windows, alas. Ah well.
Isn't there a law that makes an offence of attempting to enforce a patent which you know or suspect to be invalid?
Because there bloody well ought to be. There's simply no way in hell that Microsoft's patent is valid -- not only is it beyond the scope of what is legally patentable, but there is also prior art; and even without a shred of prior art, it would also fail the non-obviety requirement in a heartbeat.
It would be interesting to see Microsoft branded vexatious litigants and barred from the courts ..........
I really wish people would stop thinking that Linux (and any other GPL'd products) are the only F/OSS solutions out there. They're not. And TomTom could equally easily move to one of those instead and not end up running any Microsoft product as not all of them suffer from the millstone around their neck of the GPL.
The USPTO would LOVE to be more selective in its acceptance of patents. However, their hands are tied for two reasons. 1) They're short-staffed. The Department of Commerce (that the USPTO falls under) doesn't exactly get the attention, say, the Department of Defense does in the budgets. They don't have the manpower needed to go through the patent applications intelligently. 2) There is a time limit on which to act. According to federal law, once a patent application is filed, a protester only has a certain amount of time to act and has to it smartly (this was good-intentioned--Congress was trying to prevent applicants from getting harassed).
Microsoft hopes to get TomTom to settle this case. I doubt Microsoft wants to litigate the FAT patents. But, if a settlement is in the works, language will imply that the FAT patents are covered as well.
If you read the case it is not only about FAT patents. Several deal with GPS stuff. And they may or may not be valid. But everyone knows that most litigation is in fact settled. So Microsoft and TomTom get a settlement and Microsoft can claim one more patent license on Linux.
If you go back to VMS V1.5 (or thereabouts if my memory serves me correctly) circa 1977/78 you will find that the Files-11 ACP2 Files System was able to handle long filenames
As well as doing versioning up to 2**16 versions of the same file which is something that the vast majority of other filesystem types can't do even today. File versioning came in tihe RSX11-D/M in the mid 1970's on PDP-11 processors. They used the Files-11 ACP1 format.
Personally I hope that the Microsoft Patent is found to be invalid and that TomTom get a bid wad of $$$ as damages but due to the tricks that lawyers play with the US Legal system in order to line their own pockets, I would doubt that TomTom would ever see any of the money at all.
I've been looking at this since this first appeared on The Register.
I've been thinking that Microsoft must be aiding their own demise - Microsoft must made a large sum from IP as well as selling software (Windows/Office/etc) and hardware (Xbox/mice/keyboards/etc).
If they (Microsoft) win this case surely they will be aiding companies (said companies will no longer need/want to use the FAT without a license or will convert their legacy devices to ext2 or some other free file system). If 'said companies' use FAT, then they are also having to pay a royalty and in these financial times they may not wish to continue to do this.
Also what is to stop TomTom from going to said "other companies" which license or currently use unlicensed FAT usage from switching to the same future file system (ext2 or other) and release it under a license which bans/stops Microsoft from using it.
This would put Microsoft in a position where they are in a 'catch 22' situation (especially if said file system was released under say a GPL 2 or 3 license).
TomTom could do this out of revenge for loosing in court and Microsoft would lose royalties from FAT as nobody would continue to use it).
TomTom is also in a position where they could force a firmware update on older device and build in support on new devices when maps are freely downloaded (they would technically then remove the FAT file system and use ext2 instead). They could do this as a "security update".
Long Document Names entry in WordPerfect 5.1 Help:
"Long Document Names: Enables you to assign long names to documents. A long document name can contain up to 68 characters (including spaces) and is saved with your document with the normal DOS filename. This option determines whether WordPerfect prompts you for the document's long name and type when you save or exit a document. List Files will be set to display WordPerfect documents with their long names and types."
Can you say "Prior Art"? Can you say "Obviousness"?
Or has Microsoft banned the use of those words?
It took only two men to show the world that Everest could be conquered. The next one to the summit was only a matter of time. So it is with inventions, innovations, whathaveyou. Once WordPerfect showed it could be done, it was too obvious for words.
"TomTom can just use Solaris or OpenSolaris FOR FREE and avoid the whole legal issue..."
No they can't. "Open"Solaris is heavily dependent on a number of binary blobs. If you can manage to build it at all on anything besides x86, you'll still find it depends on a missing binary-only component.
As others have mentioned - Tom Tom could quite easily use another filesystem.
But as windows will need extra software for installed for this filesystem to work (expect a service pack update to break it on purpose on a 'security' basis) - this complicates things for end users.
All OSs should have a common open free filesystem with mandatory full support.
Please someone who in Govt who isn't taking a bribe. Please break that p.o.s company up.
Reading post on here you would think Apple and Microsoft were the only two computer companies that have ever been... DEC VMS had long file names long before either of these two crawled out of the primordial cess-pool. Surprising though it may seem the history of computing does not begin in 1975, and there is nothing new under the sun (oh I forgot about them, do the still exist ?)
``BTW I would have thought it should be easy to support ext3 or whatever on windows, just insert an auto run exe file[...]''
What fs type will you use for the partition that contains this exe? You can't make it ext3 (or whatever) because the user would need to have the driver installed already, and you can't make it FAT for the very reasons we're having this discussion.
I guess none of the commenters here have partitioned a drive in the last 10 years. When you fire up fdisk, or whatever your favourite partitioning too is, you partition the drive using the MSDOS partition table. Up to 4 primary partitions and as many secondary (extended) partitions as you like. So every Linux system out there is using MSDOS regardless of what file system they format the drive with. Try to set up your ext3 partition without using the msdos partition table. I for one have never seen an option to use any other.
FAT is being used because it's the only file system that is read/write supported on almost every operating system, something that TomTom require to allow users to update their database. They can't switch to a unix file system, as we all know that microsoft don't support other file systems than their own and no matter what we may think, Microsoft still own the Lion's share of the desktop/laptop PC market.
Clause 7 or not, this is a situation in which Microsoft really has them over a barrel: To meet their market they need to support Windows desktops, and to do that requires FAT. They've got nowhere to turn, with their only hope being that Microsoft's patent on FAT gets overturned.
FAT was developed in late 1970's. It's old, but serviceable. What I don't understand is how the hell Microsoft can still HAVE a patent on technology that is over 30 years old. Patents are supposed to run out after 19 years. This is pure, unadulterated abuse of the patent system by a company noted for its abuse of any process it gets involved with.
However, that's the rules. Unfortunately, Microsoft still refuses to support a filesystem they don't own, and because of Microsofts de-facto monopoly nobody can use anything else either and expect their stuff to work with Microsoft products. Perhaps these anti-trust folks dropped the ball somewhere, and should force Microsoft to YET AGAIN stop using their de-facto monopoly to push competitors out of business. This means force Microsoft to support at least ONE filesystem that they don't own and control software (ie: abusive, monopolistic and completely against the purpose of patents in the first place) patents in. Which is what they are yet again trying to do with this lawsuit against Linux.
Fuck Microsoft. For 30 years they've held back progress by creating users who think crashes are "normal", by forcing innovative companies to die off before they couldn't compete with Microsoft's monopoly, and by using shady and downright illegal methods to push their crap on the world, such as the recent ISO debacle. Someday, when the revolution comes, I'll be all to happy to put the local purveyors of this filth up against the wall and do my bit for a more competitive computer ecosystem.
> > BTW I would have thought it should be easy to support ext3 or whatever on windows, just insert an auto run exe file[...]''
> What fs type will you use for the partition that contains this exe? You can't make it ext3 (or whatever) because the user would need to have the driver installed already, and you can't make it FAT for the very reasons we're having this discussion.
However autorun is a major malware vector so in a locked-down environment you won't be able to run the executable.
TomTom shouldn't be using FAT file systems - the correct one for their product is JFFS. FAT is just used on removable lightweight devices like USB sticks for compatibility with things like Microsoft.
Its a dicey licence to play with anyway. FAT predates Microsoft by a good few years but you wouldn't expect a lawyer to know that.
If Linux is a pain then you can always do what lots of networking companies do -- switch to FreeBSD based product. Its a bit more work to use but its totally open. I wouldn't switch to using CE; its never proved to be much use, its big, slow and clumsy and, of course, relatively expensive.
I have read this in detail.
18 companies including Kenwood, Alpine and Pioneer according to Horacio Gutierrez of MS (Isn't that violation of the NDA they signed) are using GPL
I could not figure out why someone would not want you to know their software could read a now 28 year old file format with a method of supporting long file names which is at least 10 years old.
So what would happen if you approached one of them and asked for a copy of the source for one of their devices?
Logic says the should say "Sorry its closed source." If they don't MS can presumably do them later, and anyone who got source off them and hence is "infected." As others have noted it creates a 2 tier customer base. Those who have rights to MS elements and everyone else.
Tom tom seem to be taking a different route.
Feeding a large hungry troll is no way to get rid of them. It makes them larger. It does not lessen their hunger.
Biting the hand that feeds IT © 1998–2019