Fat File System Licensing?

[Tom`]

Another issue in this case is that "long filename support" is such an incremental and obvious improvement that it should not be patentable for that reason, in addition to those already presented. Theoretically in the US patents have to fulfill a certain standard for innovation and non-obviousness, but this seems to be rarely enforced.

 

[put4558350]

... This id my idea ...

Problem : fat, fat32 AND ntfs is own by microsoft ?
Solution : Not support it by default then have user fix it - it's ntfs solution for linux - and using Ext for os (I prefer ext4 as it's have more performance and pandora is on battery so no power loss and no data loss)

Problem : Dummy user ?
Solution : Make program that let user add fat, fat32 and ntfs support in "ntfs-3g" like interface have browser link to this program and run in it first run. or just pre install (might not be a safe option)

Problem : And how to recovery it ?
Solution : There is ways to do this is w/o fat. samples of my ideas.
- To recovery ask user to have a empty sd-card have a windows-side program make a Ext recovery disk after pandora is recovered let user format it back to fat (using windows)
- Pandora have usb-in so have a windows-side program dump os's image file directly to device.
- older fat (file name in 8.3 format) is not include in the microsoft's license (unsure about this one) so support fat but not support vfat when in recovery mode

 

[Monk]

Wtf shouldn't a company producing mice for microsoft make mice for other companies?

Let me be clear - they did not simply make mice "for other companies". they made MICROSOFT mice for themselves, and sold them as MICROSOFT mice for less money than M$ themselves did.

They used the exact designs - indeed the exact production line - that they used for Microsoft to run off extras for themselves. They not only did not pay to develop the mice designs themselves, they were paid by MS to make the things! LOL. You don't see anything wrong with that, or with the Analog Nub (or case or circuit board) makers for OpenPandora making OpenPandora nubs (or case or circuit board or...) _from the OpenPandora designs_ for their own profit, none to OpenPandora, as on any way "wrong"?

And the FAT-case is something very special, since it's very hard to implement custom filesystems into windows and ntfs isn't an adequate fs for flash-media, Nobody uses vfat for it's superior technology but only for compatibility.with an operating system controlling the market.

Let's be clear here - I'm not defending Micro$oft or any other company - not even OpenPandora - at all. Certainly not on the basis of "technical excellence" or the ability to beat the competition to a pulp more via business acumen than technical ability - I'm just querying where the line is, when or why it's OK to appropriate someone else's work.

Technical excellence is NOT the point here to me AT ALL - "benefit" certainly is. In the FAT instance, Linux (it seems to me) would not even countenance using FAT or vFAT if there was no benefit. The benefit is "only" compatability? Fine - then the benefit IS compatability. To me, once that is established any argument about FAT's technical excellence is moot - compatability is the benefit to be gained. I am curious about why Micro$oft - a notoriously CLOSED company from my point of view, OWES that compatability capability to anyone?

I suppose that patents are some sort of useful, but they seem very childish in nature. To enforce a patent says, "I had that idea first, you are magically not allowed to do it"

Conceeded, but I put it to you that much of civilisation (at least the Western bits that I am familiar with - I'm not so familiar with Eastern philosophies) actually IS based around childish behaviour. You only have to look at the stupidity of the whole financial system to wonder what on Earth we've been thinking of - it's based on greed and open (nudge-nudge-wink-wink) lies and always has been... but our whole economies are built upon that foundation of greed and "looking the other way".

It's also how "the little guy" gets to climb up the ladder - OpenPandora obviously hope they should make some money from the Pandora project. They won't be able to do so (or not as "easily") if someone actually takes all their data and reproduces the product.

It helps the business that holds the IP, but it keeps their customers from getting anything better from someone else, especially for huge companies that can afford to fight for their patents.

I'm not trying to restrict the discussion to patents (a system which is - rather like banking! - widely acknowledged to be fundamentally broken as it is run incompetantly) but I believ ein essence that it is not "patents" that prevent "better" from coming out - for example, Linux has access to better filing systems than FAT, after all - but rather it is anti-competitive business behaviour which achieves that aim.

Take "gun law" as another analogy for what I'm talking about. It is said that guns do not kill people, people kill people. That is fairly close to the heart of why many americans are allowed to have guns (FAR from the only reason, but it IS one rationalisation given). To me it seems that patents don't kill better ideas (they can at best, after all, only kill equivalent ideas surely?) - it is businesses, or people, that kill the competition.

And at some point, whoever is selling licenses is making money off of something for which all the thinking has already been paid for.

Two issues here. One is - who says that licenses only pay profit, not development costs etc? I saw recently someone assuming tha tOpenPandora were making a profit from batch 1 Pandoras. I am not saying OpenPandora aren't - I don't have access to their accounts - but we are told they aren't. Assuming that licences for patened stuff is only paying profit feels like it MUST be wrong on a global scale. Obviously, there are also patent trolls/squatters who have patents for things they have done squat to achieve, but that is specific cases not genaralisations (which I'm trying to use to understand the overall position better).

The second issues is - so? Let's propose a company - call it OpenMonk - puts a lot of hard work into something, and patent it. They've done the work hoping to make a profit - that's they they did the work in the first place. Once their licenses have covered the cost of the staff employed to create the IP, why should they NOT then go on to charge license fees to actually earn that profit they were chasing in the first place? AFAICS that is a big part of "The American Dream" - to go out and make your fortune. If you're not allowed to make a profit from your efforts - beyond daily subsitance for the current days work - what's the point?

I don't know how much Microsoft invested in its employees before they hacked long filenames into FAT, but it's possible that the license has already paid for the development of the technique, and now they're just making money because they happened to use the technique first.

Getting rather more specific than my query, and yes the Long Filename thing I would expect to have already paid for the staff who came up with that EXACT way of doing it - but why does someone else (whoever wrote the vFAT library for Linux) deserve that IP for free? Why do the originators deserve NOTHING?

Again, I am not defending Micro$oft and really would prefer the conversation not steer me in that direction to keep it going, but I am truly curious about why the developer of any random bit of IP does or does not deserve to "own" it, and charge for someone else to use it. it could be the vFAT rights (which I'd hope would be incredibly cheap, worthy of their nature) or a series of books to make "Lord of the Rings" pale into insignificance, or the Pandora designs. And if some of those examples DO deserve to belong to their putative owners, where is the line that says "below this you can own it and charge for it, above the line it should be free for all".

But more importantly, we aren't supposed to be arguing about the nature of patents in this thread, and it's a waste of time for me as I'm hardly in charge of issuing them.

Ah. You possibly shouldn't have posted that then

CHIP - guidance please. I know how much topic drift seems to get hated around here, so can we move these posts to a thread or will this just get locked?

 

[Monk]

It's unfortunate TomTom caved

Agreed

QUOTE

I'm genuinely curious about the degree of upset that SEEMS to exist over the concept of people wanting remuneration for their IP.

Renumeration for what? It's not as though Microsoft made this FAT extension out of some event of sheer unmatched brilliance. Linux kernel developers had done this without any Microsoft code.

Well, I carefully didn't restrict my comments to the FAT extension becaus ethis is FAR from the only IP topic that I've seen this sort of attitude for, but remuneration for the IP - whatever the IP is. The vFAT extension's worth could be determined IMHO honestly and fairly in two different ways, with very different consequences:

A) What is the technical quality of the IP. In this, the "worth" would be judged by how much effort went into designing the IP, how complicated or technically excellent the IP is. If the two parties (IP owner and whoever wants to use the IP) disagree on the value then the one who wants to use the IP would simply go design their own IP from SCRATCH.

B) How much does a third party WANT the IP. In this case the "worth" of the IP would be determined not by technical exellence but by how desireable the features of using the IP would be. If the two parties (IP owner and whoever wants to use the IP) disagree on the value then the one who wants to use the IP would simply go design their own IP from SCRATCH.

There are probably others.

Linux coders didn't use Microsoft code. That's fair enough. Diud they use Microsoft's design? The design of any IP is not necessarily restricted to code alone, if someone wants to use someone else's DESIGN I am after why the actual designers (or owners of the design) SHOULD not own their own work. Why SHOULD it be free for others to use - is there any criteria other than "I want it"?

The only things worth patenting are those things that require enough money and resources that to not allow them to be patented is a travesty and would cause it to never have been possible to develop. That simply does not apply the FAT patents. There have been multiple better file systems developed since then without any use of the FAT code. If it's so easy to just remake it, was it special enough to really deserve many years of protection?

I've read that a few times and it doesn't seem to hold together for me - after all, there HAVE been multiple better file systems developed since then. So why not ignore the FAT system? The answer is that A benefit derrives from using the FAT system. While I do NOT want to get bogged down in defendinf M$ or the FAT system, does it really matter WHAT the benefit is before one can decide whether it should be free or charged for? For that matter, how do you decide whether an IP is "worthy" of patent protection? Surely a lot of effort could go into projects with the developers not sure if the results will be worthy of protection... couldn't that stifle development?

Patents in this day and age are a total farce

Agreed. No argument. I have zero interest in defending the current patent systems, I am genuinely curious about the wider aspect of IP ownership of which patents is but a small (if significant) part.

if it's developed by a corporation, then it entirely losses "little guy wins" appeal.

Ah - I think I've hit on your personal approach/attitude - little guy = good, corporate = bad. Not a bad philosophy. A littel over-generalised for me personally.

QUOTE

if it is easier to reverse-engineer someone else's work than to do your own from scratch

Yes, because FAT Is the ultimate filesystem and people who run Linux actually use it for their main file storage...um.. ? They only had to reverse-engineer it to remain compatible

That tells us only, really, WHAT the benefit is and is not (or how important it is considered to be). The benefit is not technical excellence, it is compatability. If it's "only" compatability then the answer is simple - don't do it. If compatability is important, then it's a big deal.

Microsoft has a habit of using these patents to lock people out.

Do not ask me to defenc Micro$oft or their business practices, please. I could probably attempt it but it might just cause my devil's advocate muscle to explode. Which would leave my wife and child bereft

None of these "technologies" were at their inception more advanced than their peers but they practically bribe others and continue to extend, and to lock-out others to maintain dominance and now they beat people with that same stick using patent lawsuits. You defend them and say they are justified because they deserve renumeration.


I do not beleive I defended Microsoft's business practises.

I'm genuinely curious about the degree of upset that SEEMS to exist over the concept of people wanting remuneration for their IP.

I just want to make clear here (as the above quote hadn't actually occurred to me when I wrote my earlier post ), that there's none from me - I'm fine with that, as long as the systems for doing so aren't misused to prevent fair competition.

I have become disenchanted in recent decades with terms like "business" and even "professional" as they no longer mean what they meant to me in my youth. Which is sad, I think. Too much time and money is spent on "screwing the other guy" IMHO

What makes me uncomfortable is the notion of money (any amount, even if it's just a tiny one) I've paid for a product I want, going to a company whose products and inventions I don't want, don't use, don't want to use, and will not be using, who had no involvement in the creation of the product I chose to buy. It irks me in a way I can't seem to find the words for at this moment. :lol:

I think we'r ein genreal accord, except that of course for the Jurassic Park Principle.

In JP there's a line spoken by the annoying Chaos Theorist character along the lines of "You built on the shoulders of giants and now you're packaging it, you're selling it". A number of consequences or even discussions can come out of whatever the precise phrase was, but the principle here is that a person/group/company... even a corporation... can simply "take the next step". We are endebted to the designers of everything that went before, in some way shape or other, in getting us where we are now. In many cases that "debt" is officially paid, although we continue to benefit from things as diverse as the USA's "Constitution" (which probably owes a little itself to the earlier "The Articles of Confederation and Perpetual Union") to the invention of the transistor and so on and so forth.

Some things we still pay homage to their creators for - although, as with the examples immediately above, this is not always a financial payment but often more one of respect. this whole "FAT" thing has simply built on top of other comments to make me wonder at why so little respect is due to some IP ownership while considerably more is - I hope! - owed to others. I sincerely hoped that my comparisons to OpenPandora would show a solidarity of support for OP, and that it would allow for the dislike of other IP owners into sharper relief. Is it just because of M$'s success (which IMHO is far more to do with business acumen than technical prowes) or is it success in general? Is it more to do with the... agressive... behaviour of some IP owners as opposed to the more genial approach of others?

FWIW, the Pandora would not be the machine it was originally sold as if it cannot use FAT formatted SD cards, in which respect it is built on top of the work of SD card designers and the FAT format designers. Without the existance of those two groups, we would be endebted to two other groups that would have developed something similar, I am sure... but those other groups didn't, or didn't become popular enough or... the point remains that OpenPandora are leveraging the achievements of others to create an utterly wonderful but of kit. They can schew to use licensed IP emtirely, but if they do so their handheld will be a lesser delight - so I wonder, is the license fee worth it? Because if the license fee isn't worth it, don't include the feature - problem solved. Obviously SD cards are worth it, license paid. vFAT access? If it is worth it and above board, why NOT pay for it? If it isn't worth the money, one can simply not include the feature.

Grief. This response (Prometheus) went on too long but I can't see how to trim it. It was as long as it was because you seemed open to chatting, so I felt able to "chat" I appreciated your comments particularly for some reason.


@WizardStan

I have not looked up the history of FAT32, vFAT, and patents, but I think we can agree both that the patent system is FUBAR and that Micro$oft have some ruthless, in fact IMHO despicable, business practices. I don't want to get into saying anything that could be considered libel, but they are not my favourite company by a very great distance. I thank you for your response greatly

@Chip thanks for your response too, and I agree that business practices such as you and WizardStan have outlined are... well, typical Micro$oft to be honest. Not nice at all.

Thank you also for your letting the conversation stray this far - it's been suggested that this isn't the place to hold the discussion. I am more than happy to hold it elsewhere, perhaps in a thread less tightly focussed on the FAT issue?

This should tell us all to stay away from Mono or Silverlight or DirectX or any other "standards" Microsoft pushes.

Microsoft won't start charging for Silverlight unless they have muscled their way into holding a majority over Flash, IMHO.


But that's the point - I don't want to type on any more than I have to (Chip has already been lenient enough with me and this post is turning into War And Peace) but the point is to not trust M$ in the belief that, based on past experience, if they get a large enough hold on ANY standard, that will be to the detriment of the customers in the long run.

@hedwards

I'm feeling that for some it really is the IP, a lack of any respect for IP owndership at all - at least, that's what I was feeling. Hatred against M$ I can more than readily understand. My thanks for your post and POV

IP law in the US may not have gone as far off the deep and as banking did, but it seems to be just one of several systems deep in the core of Western Civilisation which is diving off the high board into the shallow end of the pool

@Tom :nod - again, the patent system is, erm, poorly operated?

Thanks for your input folks. It seems, if these posts are anything to go by, that we may have more similar opinions on IP than I had been getting a feel of. Which, if it does nothing else of value, at least raises my spirits a little

Ta

 

[Chip]

The broken part about IP patenting (specifically software patenting) is that it allows vague concepts to be patented. In the physical world, you can only patent a device designed to perform a task. In the software world, you are allowed to patent the idea of performing a task. It sounds like a negligible distinction but it is not. In the physical world, if you invent a better mousetrap, you can patent it. As long as each design is original, thousands of different people can patent different ways of trapping mice. With software, you can patent the very concept of mouse capture. Once patented, nobody else can trap a mouse by any method without your permission.

As has been pointed out, Linux FAT handling uses entirely original code. None of MS's actual work has been taken. The programmers who added FAT support to Linux did so through entirely original work. It does essentially the same thing as the MS method, but it does it in a novel way. Unfortunately, due to the broken nature of software patents, this is not good enough. In essence, the Linux team has invented a better mousetrap, but they're not allowed to give it away because MS owns the idea of trapping mice!



All these arguments are essentially moot. As morally bankrupt as patent trolling is, it is still legal. Whether or not this particular patent could conceivably be overturned in court, Openpandora could never afford to litigate such a case. Nobody here is a lawyer, and certainly nobody here is in a position to repair the broken US patent, trademark, and copyright systems. Nothing said here will change the system, nor will it affect Openpandora's decision on how to handle the situation.

As any further discussion is essentially irrelevant to the Pandora project, I'm moving this thread to off-topic.

 

[Phawx]

Do you think OP Ltd would even need to worry about this FAT licensing issue. I mean OP Ltd barely sold 4,000 units.

How many units did TomTom sell before MS wanted a piece of the action?

To put my question in perspective: Microsoft had a hardware networking division. They sold wireless routers, brides and other related networking goods.

support.microsoft.com/kb/902461

The home networking division was going well until they missed the mark to deploy 802.11g units.

news.zdnet.co.uk/internet/0,1000000097,2132678,00.htm

Eventually, the competition slowly took more of the market share. And when MS only had 20% market share for the consumer market, they had to close up shop. They had 1/5 of a consumer networking market and decided that it wasn't enough for them.

I am having a difficult time finding this article, but I clearly remember reading it when it happened as they had clearance's on all Microsoft Home Networking products.

So what I am getting at: Should OP Ltd. even care to license FAT from MS right now?

 

[Monk]

So what I am getting at: Should OP Ltd. even care to license FAT from MS right now?

That is probably at least part of what they are looking into now.

@Chip I dunno about the "trap a mouse" description. I'm torn between my love of a good conversation and my dislike of dischord, so I'll be brief.

I think I prefer to think that attempts to reverse-engineer something - like a file format - is not an attempt to produce a BETTER mousetrap, but an attempt to reproduce essentially the SAME mousetrap. Kind of like trying to print your own money, which is just a format for laying ink around a piece of paper (at its simplest). The banks and courts frown on people who try to produce their own money from clone printing presses on clone paper. I beleive that a format for data is as much the result of thinking as any other design (programming code, circuit board layout, case design, paintings, comic books, animations etc.). I'm slightly worried that some intellectual endevours are discounted as worthwhile while others are lauded as worthy of pretection (yes, I have IP outside the realm of programming code - also within that realm).

 

[Immerman]

Licensing FAT Long file name IS NOT A LEGAL OPTION

Okay, in all the discussion I didn't notice anyone bring up the reason that TomTom agreed to eventual removal of FAT-Long File Name support rather than licensing it as part of the settlement deal along with the mapping patents - namely sub-licensing. The GPL prohibits distribution of software that is encumbered in a manner that prohibits redistribution, If TomTom were to have licensed the FAT-LFN patents they could no longer legally distribute the linux kernel because their customers would not have a license to redistribute it, Since there's no way Microsoft was going to grant them perpetually sub-licensable distribution rights they were pretty much out of options.

Check out Groklaw.net for a more in-depth discussion of the legal issues, you'll need to go back a week or so to get to the meaty articles.

As I see it there's three options for Open Pandora here:
- Ignore it and hope MS lets it slide, (or be prepared to duke it out in court if they don't)
- license the FAT-LFN patents and switch to a non-GPL OS that doesn't require sublicensing rights
- Remove LFN support up front, and switch to another filesystem entirely for "internal" use

Rock, meet hard place, meet inconvenience due to (possibly) patent-encumbered standard filesystem
Option 1 could get very expensive, so I do hope the O'Pan team have a back-up plan in place, some good options have been offered in this thread. I'd vote for removing LFN support up front and letting users download an unofficial version of the OS if they really want it.

As far as the unbrickable restoration procedures are concerned, I don't see that LFN support is much of an issue, just make sure that a minimal restoration disk can boot using only 8.3 filenames. Might make for an ugly restore disk, but it's not going to see much use anyway. Just boot up off it, format the internal flash and extract a fresh install onto it from a suitably short-named .tar file, and reboot.

I'm really looking forward to playing with my Pandora, and I'd hate to see the project get a MS funded kick in the teeth when it's finally taking off.

 

[Xanas]

Ah - I think I've hit on your personal approach/attitude - little guy = good, corporate = bad. Not a bad philosophy. A littel over-generalised for me personally.

I was being somewhat facetious on this point. I was saying some people see the idea of easy patents from "flash of brilliance" as a rather novel idea because it helps to build up people who otherwise would have made very little of value. They aren't consistent winners, they are many times regular losers who happened to find a single niche that worked. I personally think these people are valued far too greatly. I know others really disagree with this, and they like it cause of that "little guy wins" mentality

I was saying that when it's a corporation, you no longer give people that feeling and it becomes much more apparent to the public the problem with patent law. But these problems always existed. People make inventions that are just simple little things that many could think up with a little effort, and they get a lot of money when really they didn't even do that much.

I really think patents should only be for those things that cost a lot of money in research and that not paying them back for that is a travesty, like I said before.

People "getting rich quick" is not a desirable ability IMO.

 

[DigitalWings]

Since the Pandora is UK based, I think we may be in the clear because I have heard the the EU does not have software patents. However I have not read this entire thread so it could also be nothing to do with it.

 

[Bronek]

Licensing FAT Long file name IS NOT A LEGAL OPTION

Okay, in all the discussion I didn"t notice anyone bring up the reason that TomTom agreed to eventual removal of FAT-Long File Name support rather than



actually, aq21 noticed this . I was not aware of this clause in GPL when I started this thread, but it makes perfect sense to me now.

As I see it there"s three options for Open Pandora here:
- Ignore it and hope MS lets it slide, (or be prepared to duke it out in court if they don"t)
- license the FAT-LFN patents and switch to a non-GPL OS that doesn"t require sublicensing rights
- Remove LFN support up front, and switch to another filesystem entirely for "internal" use
... I"d vote for removing LFN support up front and letting users download an unofficial version of the OS if they really want it.

+1

I do not need LFN re-flashing (I hope), swapping SD cards with my camera or other devices etc. . As far as "primary" filesystem goes FAT, even with LFN support, is still bad choice, due to lack of security and poor performance on flash (compared to alternatives). In other words - remove vfat.o module because it's more (potential) trouble than (potential) gain, use msdos.o instead and be done with it.

 

[arrrgh]

Could the OP team write their own closed source LFN driver, and then pay MS the licensing fees for distributing that?

 

[Chip]

Could the OP team write their own closed source LFN driver, and then pay MS the licensing fees for distributing that?
I've already asked this question once, but since it was ignored: Even if you sort out the FAT file system, what about the 40 other patents that MS claims the Linux kernel violates? They've never sued anybody before, and this one time they chose to sue over FAT. Who knows what they'll choose to sue over next time? With thousands of consumer electronics products from hundreds of companies running Linux of some sort, it's not like they can sue everybody. It's not like GPH is paying for FAT licensing in the Wiz, and at this point they're a better target than Openpandora anyway. They probably have some actual money to take.

Cost benefit analysis: The chance of MS coming after OP specifically is almost zero. The cost of licensing the FAT file system is very high and would create more problems than it solves with the GPL. Even if FAT was licensed, the risk of suit over the other 40 mystery patents still remains.

Conclusion: Nobody is going to pay MS anything. It doesn't make any sense.

 

[BackAssward]

I'd say it would be easier to pay the licenses than have to deal with legal costs if they do decide to go after OP. Some people don't care about the size of the company, and some don't care if the legal costs force you out of business.

But, we have someone looking into it.

Cost benefit analysis: The chance of MS coming after OP specifically is almost zero. The cost of licensing the FAT file system is very high and would create more problems than it solves with the GPL. Even if FAT was licensed, the risk of suit over the other 40 mystery patents still remains.

Conclusion: Nobody is going to pay MS anything. It doesn't make any sense.

So, is this the final on this then (as the PR voice of OP)? If so, can this thread not be closed?

 

[Squidge]

Sure, since no one knows anything more, it'll only turn into more speculation and then gradually drift off-topic, so best to close it now.