Linux 6.9 Set To Drop The Old NTFS File-System Driver

[avis]

Not how I would see it, since the corruption issue I had (https://bugzilla.opensuse.org/show_bug.cgi?id=1198669) is still unaddressed almost two years later. So I'm staying away from ntfs3.

It's best not to use/enable compression for writing files with ntfs3. I've had troubles with that as well (last time two weeks ago under Linux 6.7.4) but I've not been able to come up with a reproducible test case, so I've never filed a bug report. And there was nothing in dmesg.

But I've had troubles with ntfs-3g as well. I once tried to download a ~12GB file using qBitTorrent and before the operation even completed, the file system became corrupted. chkdsk took quite some time to fix it and luckily only this file was damaged which was no big deal. From then on I stopped downloading torrents to my NTFS partitions. Simply copying files to them have always been completely reliable.

[avis]

Being against software patents does not automatically make you be against say drug patents. What makes software patents different (and why we don't need them) is first that we already have copyright as protection and secondly that there are extremely few cases of "we did this method in software" that is non-obvious to the average programmer when presented with the same problem, which is why all the software patents are so extremely silly as they are.

The third problem is that software is a too fast evolving platform to have road blocks like patents and their 25 year expiry, for drugs and industry 25 years is basically nothing but for computers it's the difference between the era of the 80286 and the Core2 Quad which is an insane range.

You advocate against software patents just because you believe software algorithms are "trivial"/"unworthy" which has never been true.

If everything was trivial, software would be worthless and everyone would be creating e.g. ChatGPT4 in their spare time. It's not happening.

Have you ever heard of NanoZip? Probably the best closed-source compressor/archiver ever and despite having been abandoned for over a decade (it's speculated its author, Sami Runsas, died - I personally chased him for over a year but I wasn't able to find any clues, i.e. for all intents and purposes he indeed died), no one has come up with anything close to it.

And I know a ton more such examples. Software ingenuity is staggering.

I'm not advocating for software patents though, I'm just trying to say that's it's not so clear cut to me.

I'm 100% percent against APIs being patented or copyrighted though. That's a plague which doesn't belong to this world.

[pong]

... on the NTFS partition I use to move data across Windows and Linux.

You've probably picked your preferred solution, but just in case it helps as a possible set of alternatives:

1: AFAIK some / many windows OSs support NFS (7/10 I guess anyway, maybe things before / after). So just sharing an isolated NIC and running a proper NFS server on the LINUX interface backed by a real LINUX file system might be nice.

2: Obviously you could also share an isolated (or whatever) vNIC between the guest / host and run a samba CIFS server on the linux box or somewhere else (NAS, ...) which could be backed with a decent (linux/unix) file system.

3: Then there's other stuff YMMV like using a SSHFS (is this a thing on windows?) exchange, or GIT server, or some other file system andrew AFS or whatever else someone has ported to both win and linux, etc.

4: I have no idea what the state of virtio-fs or other kinds of "share this directory" stuff is on windows at the moment, though IIRC some such things were said to be works in progress / possible to some extent at some point.

[L_A_G]

You've probably picked your preferred solution, but just in case it helps as a possible set of alternatives:

Technically I could just use my NAS (a 4 bay Synology job) for that, but keeping it running 24/7 would waste a bunch of power (manufacturer advertises 5W standby but from the wall in real life conditions its probably more like 15W) and its much slower than a PCIe SSD. Tried using an EXT4 driver for Windows, but it seems as thou Microsoft just doesn't want third parties to make file system drivers so they've always been kinda ropey.

Tripe.

What a truly insightful and well constructed argument /s

[F.Ultra]

You advocate against software patents just because you believe software algorithms are "trivial"/"unworthy" which has never been true.

If everything was trivial, software would be worthless and everyone would be creating e.g. ChatGPT4 in their spare time. It's not happening.

Have you ever heard of NanoZip? Probably the best closed-source compressor/archiver ever and despite having been abandoned for over a decade (it's speculated its author, Sami Runsas, died - I personally chased him for over a year but I wasn't able to find any clues, i.e. for all intents and purposes he indeed died), no one has come up with anything close to it.

And I know a ton more such examples. Software ingenuity is staggering.

I'm not advocating for software patents though, I'm just trying to say that's it's not so clear cut to me.

I'm 100% percent against APIs being patented or copyrighted though. That's a plague which doesn't belong to this world.

You are conflating complete software solutions, that are perfectly covered by copyright, from software patents that cover certain methods. I don't just believe that software patents are trivial, I have to date not seen a single software patents that isn't trivial.

NanoZip is a great example of an application that clearly didn't need patent protection, and it's own author never filed for one. Now I don't know how the underlying algorithm of it works but I still make the guess that had the lzw patent not expired in 2003 then NanoZip would have never seen the light of day.

[L_A_G]

And food doesn't put itself on the table. I don't care about your "software ideals" (most ideals are irrational), the Open Source movement has consusively proven that financially it does not work, period.

Considering Linux, BSD and their derivatives has basically taken over everything that isn't desktop or tiny embedded systems that's absolute hogwash. Even Microsoft has become more and more of a Linux shop with things like WSL and how the vast majority instances on their big money maker right now, Azure, are Linux VMs.

Lets just go trough market-by-market:

Servers: Over 90% of servers run Linux. Select for top 1 million and this goes up to over 96%. Linux dominates the server market.

Videogame Consoles: The Sony Playstation 4, 5 and Nintendo Switch all run FreeBSD-derived OSs and are so chock full of open source in their runtime (GLIBC etc.) and development tools (GCC). Outside of their proprietary graphics APIs/libraries there's barely any entirely proprietary code in OrbisOS (PS4&5) or NX (Switch). Together they control about 90% of the console videogame market.

Mobile devices: Android, being a Linux derivative with most of it open source, controls about 70% of the mobile market. The other 30% is iOS which in turn is a FreeBSD derivative so depending on how you look at it, open source has either 70 or 100% of this market.

Supercomputers and mainframes: Looking at the supercomputer top 500 there basically hasn't been anything that doesn't run Linux. Even in mainframes that's increasingly been moving to Linux. Even IBM, after buying Redhat (for 34 billion), is in the process of sunsetting AIX as they move its last users to Linux.

Embedded: Outside of a few Google products on Fucia and its almost almost entirely sown up between Linux, BSD, Minix and FreeRTOS. Windows CE has lost out so bad it was discontinued in 2018 and extended support ended last year. The only part where open source OSs don't dominate is embedded systems so small they don't even run an OS.

In other words; Outside of desktop and bare-metal usecases, open source either dominates or has a majority of the market. Saying that open source doesn't "put food on the table" is just so wrong its not even funny. Only way that makes any sense is if you're so stuck in the last millennia that you can't wrap your head around business models that don't revolve around up-front license fees.

Disagreeing means having a valid opinion but guess what this opinion is invalid.

There's a certain irony to this sentence considering what you wrote just before this... Being so wrong, yet so confident in what you're saying...

The world has gone mad in allowing people to opine about everything even if their opinion is asinine or even dangerous.

Really? My "opinion" that open source basically runs the world is so "dangerous" that you think it should be banned? Really?

Your "moral compass" (whoa, that sounded so haughty) is 100% bankrupt when it comes to people earning their living. I don't like people lying to themselves and others. There's been too many lies recently.

Now you're just rambling... People absolutely make money on open source. It may not have the same one that proprietary software has long had, being more software-as-a-service oriented, but if you can't compete with it, that doesn't mean there's anything wrong with it. That only means your business model is either outdated or was never any good to begin with. Large companies making and relying on open source along with major buyouts like IBM paying 34 billion for RedHat just shows that open source isn't just the future of software, its the already the present, the here and now.

It's not the first topic where you're disagreeing with people putting food on the table. I wonder if you've ever followed your own advice and done something for free. Something which is akin to a full time job for months/years. How did it work for you? Oh, wait, I'm sure you never even attempted to work for free. So full of <>.

Building proprietary software on top of open source software and contributing back to its developers is literally what I've been doing for a living for the last few years.

You may love Open Source all you want. That doesn't mean Open Source can work the way you want it to work and it certainly doesn't. In the topic about VVC you were so wound up about how horrible VVC was in principle but you never admitted that AV1 had handsomely been paid for by Google to serve Google's own needs. They didn't give a damn about "Open Source", "freedom" or "patents". They didn't want to license HEVC or VVC and instead chose to invest into something different which would work for them.

Google created VP9 for its own needs and offered it as the base for AV1. Which in turn was then developed as a collective effort and so now everyone can benefit from a free open source video codec. Being open source and widely spread allowed it to be supported in hardware much faster than it would have otherwise. It requires some pretty twisted logic to turn AV1 into anything other than a roaring success for open source.

What about filesystems? Oh, wait, not a single modern full-featured FS has been created this way either.

If you're going to be anal and limit that to file systems developed in the open ext2-4 springs to mind right off the bat. XFS and ZFS were both open sourced over 20 years ago.

Show me a proof that anything complex can be created as Open Source/being patent free for free.

The way most larger open source efforts work is that its a collective effort by often large companies. Most of the Linux kernel developers do it as part of their jobs at big companies like IBM, Google, Intel, AMD, Nvidia and Qualcom. By doing this they're also contributing their patens so that anyone who uses it, functionally gets a free license for said patents. Another example of this is the webkit browser engine, which even Microsoft switched to, which is also developed by employees at companies like Google, Apple and Microsoft as part of their jobs.

A lot of projects like the Linux kernel may have started as hobbyist projects, but major complex software tends to be developed by salaried employees at businesses who themselves or their customers use that software. Its basically become about pooling resources in a way that everyone ultimately wins. The old MIT hacker ethos at work.

Lastly you, like many people on Phoronix, are obsessed with me personally which shows not only that you don't actually have a moral compass, it shows how morally bankrupt you are if you cannot argue without ad hominem. What a disgrace. FFS. I feel like I'm about to vomit from your overt and inane rightnessness.

Obsessed with you personally? Wow! Somebody really needs to take their psych meds because getting told off by multiple people when you say something so wrong or stupid its downright asinine doesn't mean people are obsessed with you personally. Having that happen regularly just means you're a combination of stubborn as a mule and completely wrong.

You remind me of the average slashdot commenter. A boomer or early genX:er whose minds are firmly stuck in the last century who doesn't understand open source and business models other than just selling closed source software with an up-front payment while thumbing their nose at anyone under 40. I've literally had more than one person try an ad hominem purely on my age and the last time that happened I was 31.

[avis]

L_A_G

100% whataboutism, offtopic, rant and ad hominem which I'm not even going to read. Would have worked with an actual id iot but I'm not. With that, I will not see your tripe any longer. I should have learnt from the previous topic what you are and even back then you definitely didn't represent people who allowed others to be entrepreneurs or run successful businesses. You don't even understand Linux and your Linux victories are inappropriate and shallow. No one here gives a damn that Linux runs on servers. Netflix has more FreeBSD servers than Linux, so what? This is not what this website is about.

And if you reject how this world works, well, I cannot argue with you any longer. Far too many people here exist in an alternative reality. I don't have the mental capacity to live in several world simultaneously.

[avis]

You are conflating complete software solutions, that are perfectly covered by copyright, from software patents that cover certain methods. I don't just believe that software patents are trivial, I have to date not seen a single software patents that isn't trivial.

NanoZip is a great example of an application that clearly didn't need patent protection, and it's own author never filed for one. Now I don't know how the underlying algorithm of it works but I still make the guess that had the lzw patent not expired in 2003 then NanoZip would have never seen the light of day.

If individual software methods become free and available to anyone, that could mean that your complete software product might be undercut by those reimplementing it using the same software methods but slightly altered in a way they are all put together. Sorry, this distinction of yours between a complete "copyrighted" product and individual software methods/patents becomes way too vague to me. I don't understand how this can work legally. It looks like the entire system becomes ripe for abuse. Or maybe I don't understand copyright but then I've never used it for anything other than literature and art. Maybe you're right. I've not seen that many software patents either.

[L_A_G]

L_A_G
100% whataboutism, offtopic, rant and ad hominem which I'm not even going to read. Would have worked with an actual id iot but I'm not. With that, I will not see your tripe any longer. I should have learnt from the previous topic what you are and even back then you definitely didn't represent people who allowed others to be entrepreneurs or run successful businesses. You don't even understand Linux and your Linux victories are inappropriate and shallow. No one here gives a damn that Linux runs on servers. Netflix has more FreeBSD servers than Linux, so what? This is not what this website is about.

Whataboutism? Do you even know what that means? Because there was nothing in that post that could even be described as such. My post also wasn't any more off topic or ad hominem than your post that I was replying to. If you're going to do ad hominem attacks and insult someone you've got absolutely nothing to complain about when you also get them in return. You're just some random schmuck on the internet.

You claimed open source was a failure and I pointed to multiple markets where it either dominates or has a majority of the market. Don't forget you're on a site that, among many things, tests server CPUs, storage and software so servers are absolutely relevant on this site. The fact that you don't like how open source outflanks the old business models software used to be sold is just your own problem. If you just can't compete with something, then your business model sucks. Its just capitalism.

If you can't deal with capitalism, then go set up shop somewhere else. I hear the weather is nice in Cuba and Venezuela.

And if you reject how this world works, well, I cannot argue with you any longer. Far too many people here exist in an alternative reality. I don't have the mental capacity to live in several world simultaneously.

That's pretty rich coming from you. You're stuck in how the world used to work to a truly baffling degree. I live in the present, not the past and I suggest you update your world view to the era after Darude's Sandstorm played on the radio.

[miquels]

Copying to Linux partitions will result in in btime (creation time) being lost but Linux still doesn't have an API call for that either in the kernel (syscall) or glibc.

Ext4 has stored btime for a very long time, it was just not possible to query it for a while.
Linux added a statx() syscall in 4.11 that can return btime. Glibc support was added in 2.28. Coreutils added support in 8.32 (ls and stat) in 2020.