The New Microsoft exFAT File-System Driver Is Set To Land With Linux 5.7

[oiaohm]

We may endlessly exercise in how we all understand OIN and legal situation around exFAT. This will all however be our assumptions and theories, unless they’re taken into a court and judge says what the reality is.

Not exactly. You miss independent legal review. There are independant legal reviews and courts cases how the OIN license members are effected by the OIN contract.

The statements Paragon is making on a link that I have provided here (https://www.paragon-software.com/exfat-license/) may be questioned and treated in various ways and from different perspectives.

Not exactly that document contains a error. The fact is exfat patents are already part of OIN protection due to the contract Microsoft has signed with OIN. So there is no transfer need in future. Also the contract is automately updating by design.

Yes so Paragon can keep on selling Commercial exFat licenses as long as possible they have to put out that FUD. Its not a independent legal review but a company bias write up to keep on selling product.

Of course you have been sent out here to push the incorrect product information around so your company can keep on selling product.

Commercial exFAT deals are usually strictly confidential and cannot be referred to publicly, except for rare exceptions. These deals are the result of careful legal analysis by a lot more professional lawyers that we can find in this forum (at least a lot more professional than I am)

Without falling into a detailed analysis of your statements I can certainly assert, that big commercial players (dozens of them) do not tend to agree with your vision of the situation. They did choose, they still choose and will be choosing to pay exFAT fee for as long as exFAT patents are valid. This is the reality.

Companies are known to be conservative so FUD will generally work as a sales model for 3 to 5 years allowing you to collect money under false pretence from the companies you just have to word what you write careful enough that you can claim plausible deniability to the error that the Linux kernel define has to be updated to include exfat.

Now if your company had done proper legal review what you had their would be written differently.

The Linux system define at OIN will not at this stage cover tools to create/repair exfat partitions. To read/write exfat using the mainline Linux kernel include driver is covered by OIN.

So you do have a product to sell at this stage.

its not august 28 2019 when exfat driver falls under OIN.

https://lkml.org/lkml/2019/11/24/187

Its 24 Nov 2019 when it released as part of the mainline Linux kernel. So the complete paragon software write up is wrong after that date keep that document not updated with that change allows you to FUD sell more product.

Yes Microsoft release specification Aug 28 2019 but the OIN trigger is when Microsoft formally agreed to accept the merge into the Linux kernel mainline

https://lore.kernel.org/lkml/[email protected]/

That is August 28 but there is a contract delay to allow party to change mind. Same day the release the specifications free of change Sasha Levin Signs off on the merge into Linux kernel. This is OIN contract effecting action is at least 20 days latter to allow a party to give sorry I revoke approval. 24 Nov 2019 exfat mainline version falls under OIN protection as this is now a updated version of the Linux kernel covered by the OIN agreement and there as been no revoke in that time frame so it done.

Yes august 28 2019 going out and using exfat was wrong. After 24 Nov 2019 using mainline kernel with exfat is covered by OIN as the contract requirement are triggered and Microsoft changing mind past that date will require them breaching the OIN agreement they have signed and they need for patents they need.

Samsung is merging their exfat driver mainline to get OIN coverage so they don't have to pay exfat licenses on devices that don't create exfat partitions that only read/write them.

So there are some really big players that agree with my point of view of the OIN license and understand its triggers.

Starting see where paragon write up is FUD its badly over-claiming. Yes OIN platform define could be updated in future to cover the userspace tools for making and repairing exfat partitions and that would leave no uncovered area for a closed source product.

So the parties buying you drivers are they releasing devices that are formatting/repairing exfat media if so their legal department will be telling them to buy your product at this stage because that functionally not under OIN. General read/write for those using mainline kernels is under OIN and due to the OIN platform define wording back-porting is covered.

Final thing that you have not consider is the first exfat driver merged into the Linux kernel a quality driver. No its not it was placed in staging because it not quality. The samsung driver merging is way better quality.

Alex Zu there is something critical to remember you get caught out selling a product under FUD companies in future if you don't correct it first and admit your opps and give refunds where correct those companies will not buy product from your company in future instead choose competitor or find some other way.

That exfat-license page Alex Zu you linked to is very careful not to have a date of last update either. This is a stunt you see commonly done to use FUD legal information so they don't have to admit they altered it or made it after something else changed making the document incorrect.

[Alex Zu]

Not exactly

The reason, why dozens of commercial players are working with Paragon is exactly the «exactness».

This is why big guys are coming to Paragon to know the exact situation. Which is published on Paragon Web Site. And you’re right - it changes in time. Again - I am not going to dive into detailed discussion on where your long logical conclusions are wrong. You have already agreed that at least some of the Paragon statements are correct.

One other thing that you’re totally missing here - is the internal requirements that OEM boys are expecting from the drivers. I bet you do not even imagine what is in those specs.

The reason of the confidence behind the statement, that commercial players will be using commercial solutions - is the very clear understanding that current exFAT implementations in Linux Kernel will fail on a major portion of requirements test. I leave it up to you to believe that open-source community will get to a commercial level in 5.x (and even 6.x) - good luck.

There is an article here which is written by your allies - https://arstechnica.com/information-...appy-about-it/ - where they object the nature of inability of open source community to replace proprietary solutions. I respect your opinion, I respect their opinion - let a hundred flowers blossom (c) Mao Zedong.

One other thing that you do not seem to understand - is that Paragon totally embrace and support Microsoft’s move into OIN and Opensource. This is the way to go. And you will soon see, that Paragon totally follows this trend. This, however does not mean that exFAT patent is available to anyone for free.

[oiaohm]

One other thing that you’re totally missing here - is the internal requirements that OEM boys are expecting from the drivers. I bet you do not even imagine what is in those specs.

https://www.phoronix.com/scan.php?pa...tter-exFAT-v11

Don't really have to. You have a very big OEM company here Samsung who will be providing the Linux kernel 5.7 version of the exfat driver.

The reason of the confidence behind the statement, that commercial players will be using commercial solutions - is the very clear understanding that current exFAT implementations in Linux Kernel will fail on a major portion of requirements test. I leave it up to you to believe that open-source community will get to a commercial level in 5.x (and even 6.x) - good luck.

Yes it does fail a stack of exfat conformance things in the current version of the Linux kernel 5.6. In fact it has been deleted from staging in the current cycle and samsung version of exfat imported for the next version of the Linux kernel. Yes the samsung version is not a staging driver but full mainline driver that does pass the conformance tests.

Linux kernel 5.7 sees samsung production exfat driver mainline. Would have been mainline linux kernel a decade ago.

https://git.kernel.org/pub/scm/linux...2380edb92b0661

Yes Samsung is the future maintainer of the exFAT driver in the Linux kernel mainline.

This, however does not mean that exFAT patent is available to anyone for free.

As long at the party is using the Linux kernel mainline drivers for exFAT the patents issue is block by the OIN contract turtles all the way down clauses. Of course Samsung is going to bring the userspace tools into the OIN platform define as well.

I leave it up to you to believe that open-source community will get to a commercial level in 5.x (and even 6.x) - good luck.

This is LOL. This is not open-source community work. The open-source community work was getting Microsoft to sign off on the first driver. The work going forwards is commercial work from Samsung that they want merged mainline Linux kernel so they have less maintenance costs on their code base caused by Linux kernel changes. A lot of Linux kernel structure changes for mainline drivers are fixed up by the Coccinelle SmPL patches basic zero effort.

Yes the paragon write up is wrong. March 2020 you could release a product using the first driver in the Linux kernel for exfat and be legal. Ok buggy driver not highly recommend but you could. When 5.7 kernel releases you can back-port samsung exfat driver that is a lot better quality driver.

Stating that the first exfat driver was buggy and you were still selling you exfat driver until it improved would have been fine. Saying there was no legal way todo it was wrong.

No its not future support for exfat its Linux kernel 5.4 support November 24th, 2019 so by March 2020 there was a legal way if you were willing to tollerate a staging grade driver bugs and all.

Version 5.7 of the Linux kernel is due to land later this spring(2020). <<this is when the game changes again. Yes the samsung exfat driver that is going mainline was maintained out of tree so there are all the bits with it to take it back to older versions of kernels. That date there will be a quality exfat driver on offer. Also by that data Samsung may have the make and repair tools in OIN define. So for Linux at that point exfat support will be covered.

Yes it was Microsoft patents that were blocking Samsung from merging that driver over a decade go. Samsung has been developing exfat driver longer than Paragon for commercial usage.

Please note you are dealing with Samsung who are known for sueing companies under deformation laws for false statements the clock is more than ticking on Paragon you need to get your site correct yesterday.

[Space Heater]

The reason of the confidence behind the statement, that commercial players will be using commercial solutions - is the very clear understanding that current exFAT implementations in Linux Kernel will fail on a major portion of requirements test. I leave it up to you to believe that open-source community will get to a commercial level in 5.x (and even 6.x) - good luck.

Care to explain why Samsung's exFat driver, shipped in easily dozens of millions of phones, will magically cease being "commercial level" once it hits Linus' tree?

Good luck.

[Alex Zu]

Care to explain why Samsung's exFat driver, shipped in easily dozens of millions of phones, will magically cease being "commercial level" once it hits Linus' tree?

Good luck.

Just had a customer in Q1/2020 who did a comparison of OSS driver and ours - big guys are not stupid and do have resources to test not just their own scenarios, but also run multiple days load tests. Cannot say the name here for obvious reasons*. But I do care to say that the customer have chosen to PAY for a proprietary driver after running internal tests.

The announcement of Microsoft triggered a lot of big guys to finally put their hands on OSS driver in a hope to enjoy the «free nature» of OSS. This was a fantastic step from Microsoft side in terms of marketing proprietary exFAT drivers. Did you ever wonder why most of printer manufacturers are using proprietary implementation of SMB protocol ? Do you think they’re all stupid ? I’d be very interested to hear your theory explaining that.

You guys are living in your own virtual world where «free» means «perfect». I got bad news for you - «free» is usually far far far away from even «not bad». Reality sucks. Sorry.

And one more question - why Samsung is unable for 10+ years to craft a stable NTFS implementation ? My theory is - because OSS does not work. We did. We’re not Samsung - not even a tiny fraction of it. What made you thinking that their exFAT implementation is stable and is good for doing anything beyond a limited set of simple operations under a low load on a cell phone external memory ? I’d be very interested to hear your theory explaining that.

We got a retail version of both exFAT driver for Android- https://play.google.com/store/apps/d...lugins_ntfs_ro - this is a retail product for end-users, not for big guys.

I will disclose you a great secret - lots of endusers are buying PAID exFAT support on Samsung phones for last several years. Are they all stupid ? I’d be very interested to hear your theory explaining that.

* - most of big customer prohibit to disclose their names for a simple reason - delivering a small screw to Rolls Royce does not allow to call yourself «Rolls Royce supplier» - this is too much benefit to a supplier of a small screw.

[Alex Zu]

Samsung has been developing exfat driver longer than Paragon for commercial usage.

You are generating a lot of false statements. I won’t dive in objections on others - but these 2 are evident LOL.

1. How do you know when Paragon have delivered first commercial exFAT driver and how do you know that it was later than Samsung did their own driver ? Paragon is in the business of Operating systems and file system since 1994. We have had file system products like NTFS for DOS released back in 1997. We were doing filesystem implementations 10 years before exFAT was even introduced and we were among the first to develop first exFAT implementations on the market.

I am not even touching here the situation with an access to exFAT specs (how can you develop something without specs ?) and leave it up to you to speculate who was the first to get detailed exFAT specs and when it happened. I’ve got a popcorn and Cola ready to enjoy your version of that. Make sure to go back in history before throwing out new theories not to produce next LOL.

* - just for a next discussion that we will have with you in 10 years about APFS - please note, that Paragon was the first to deliver full APFS implementation on a market - Samsung was surely not the first one here - https://www.paragon-software.com/par...ensic-experts/

2. Samsung was not developing exFAT driver for commercial usage - it has always been for their own needs. Commercial means you can sell it. Selling an OSS which is under GPL ... well. Or you’re saying that Samsung was sublicensing exFAT patent rights ?

Fact checking helps. Exactness rules

[Space Heater]

Just had a customer in Q1/2020 who did a comparison of OSS driver and ours - big guys are not stupid and do have resources to test not just their own scenarios, but also run multiple days load tests. Cannot say the name here for obvious reasons*. But I do care to say that the customer have chosen to PAY for a proprietary driver after running internal tests.

You can't give actual reasons why Samsung's driver is bad, just more FUD from a person who is openly a paid shill from Paragon.

Did you ever wonder why most of printer manufacturers are using proprietary implementation of SMB protocol ? Do you think they’re all stupid ? I’d be very interested to hear your theory explaining that.

No one is talking about SMB implementations this is about Samsung's upstream exFat driver. Could you try staying on topic? Also a ton of printers use (old and crusty) versions of samba that they make their own modifications to.

You guys are living in your own virtual world where «free» means «perfect». I got bad news for you - «free» is usually far far far away from even «not bad». Reality sucks. Sorry.

No one is saying that open source is magically better, the problem is that you're unable to make a technical argument as to why Samsung's exFat driver isn't good enough for commercial use despite being used *commercially* for years.

Your strawman sucks, sorry.

And yes, being deployed in commercial products means it's been used commercially. Your definition of it being commercial only if it's sold as a single package is laughable.

And one more question - why Samsung is unable for 10+ years to craft a stable NTFS implementation ? My theory is - because OSS does not work. We did. We’re not Samsung - not even a tiny fraction of it.

No one is talking about NTFS, your personal theory lacks evidence.

If open source doesn't work, why is Linux the dominant server and handheld OS? How do you explain postgres, nginx, apache, or llvm?
Let me also stop you before you post another screed, no one said proprietary software doesn't work and no one is saying open source software is always the highest quality.

It's very cute that you cherry-pick some dubious story about Samsung and conflate that with the entirety of open source software.

What made you thinking that their exFAT implementation is stable and is good for doing anything beyond a limited set of simple operations under a low load on a cell phone external memory ? I’d be very interested to hear your theory explaining that.

This is what we call FUD ladies and gentleman, unsubstantiated claims of it being unreliable and unproven. I'd be very interested to hear an actual technical reason why it's not fit for production. By the way, exFAT explicitly targets portable devices.

Also, please spare us any more of your dubious anecdotes about how great your company is at selling software.

[oiaohm]

1. How do you know when Paragon have delivered first commercial exFAT driver and how do you know that it was later than Samsung did their own driver ?

You asked the right question latter to answer this.

Paragon is in the business of Operating systems and file system since 1994. We have had file system products like NTFS for DOS released back in 1997. We were doing filesystem implementations 10 years before exFAT was even introduced and we were among the first to develop first exFAT implementations on the market.

This is pointless garbage means nothing about exFAT.

I am not even touching here the situation with an access to exFAT specs (how can you develop something without specs ?) and leave it up to you to speculate who was the first to get detailed exFAT specs and when it happened.

The reason why Samsung had first access to exFAT is they are silicon manufacture company. They make what is inside sd cards. Samsung had access to exFAT specifications the Microsoft internal ones in 2005 before Microsoft released exFAT public-ally as part of the SD Card Association (SDA) and other groups like it about making flash memory modules. Please note the early information did not just go to any member of the SDA only went to those who made silicon for review if there was any issues with exFAT and storage media and to produce storage media.

2. Samsung was not developing exFAT driver for commercial usage - it has always been for their own needs. Commercial means you can sell it. Selling an OSS which is under GPL ... well. Or you’re saying that Samsung was sublicensing exFAT patent rights ?

Please note Samsung got hit over the exfat driver GPL violation because it was not under GPL license but was under own Samsung proprietary license demanding payments for the code.

The Samsung code base of exfat shows them selling it to the other SD Card Association members who had Microsoft patent licenses for exFAT to use it for device manufacture. So yes Samsung exfat was sold for commercial usage and Samsung exfat is the one formats the sdxc media that comes exfat formated 99.9 percent of the time.

Microsoft would not allow samsung to license exFAT to other parties so your business got to exist.

The reality is that Samsung is the first commercial exFAT driver vendor for Linux kernel there client base was restricted by Microsoft being a jackass with patents. \

You are not number 2 either.
https://www.tuxera.com/blog/tuxera-r...e-for-android/
Tuxera is number 2 they were first to get the open license to sell into the larger market.

Tuxera code base in fact starts from the Samsung code base good fun right. Number 1 and number 2 come from the same code base. I am not even sure if Paragon is number 3 into the exFAT market there is every chance Paragon is number 4 or number 5.

When they are so many different vendors you can buy exFAT patent license though other than Paragon if you find that Paragon has been doing FUD you really don't have to give Paragon you business going forwards.

[Alex Zu]

When they are so many different vendors you can buy exFAT patent license though other than Paragon if you find that Paragon has been doing FUD you really don't have to give Paragon you business going forwards.

My arguments are of course all FUD - and I am sure you will be fighting to death to prove it.

There is one problem - real world doest not fit into your dreams.

More and more commercial players are licensing exFAT every quarter. As I said - once commercial player tests and evaluates «magic free solution» - they immediately ask for a quote. In your world they all go to Tuxera, HCC Embedded and Embedded Access and I am very happy those guys are doing great business - but some end up with Paragon by a miracle or an accident - https://www.prweb.com/releases/parag...eb17127839.htm

One other thing that you also not quite understand - is that not all OEMs hardware will use Linux as major OS to operate and control the hardware. I will let you go through this quest and find out what they use. Good luck !

[oiaohm]

My arguments are of course all FUD - and I am sure you will be fighting to death to prove it.

Not really. This is another pack of desperate marking because your time is coming to the end.

Please that prweb bull crap from you was written about device that was in production for release before the open source was an option and before the samsung code merge. Yes the benchmark was not against the newer samsung exfat code. Qualcomm Linux kernel that is for that device is not a new enough version to include the open source exfat driver as a option remember this is not going to stay this way.

The reality is both Tuxera and Paragon are going to have a hard time selling exfat going forwards to Linux kernel users. Remember the horrible saying "Good enough is Near enough". There is going to come the point fairly soon that the reference SoC vendor Linux kernels include exfat support out box that is in the mainline Linux kernel. So why do more integration work to use a third party exfat.

It is fud that using Linux with the include exfat driver based off samsung code base will require paying Microsoft license fees. Microsoft agreement with OIN does not cover non Linux OS now if you were marketing yourself that way you would not be fud and be safe.