Linus Torvalds: Speaks on Fatigue and the Future of Linux
HTML-код
- Опубликовано: 11 сен 2024
- Linus Torvalds speaks on maintainer fatigue in the Linux Kernel and how it effects the ability to get more people involved in the Linux Open Source Project. Will there be an end to Linux?
My Linux Cheat Sheet and 25 Page Checklist here:
📚 learn.savvynik...
Share this free tool and support Small RUclipsrs
editbulk.com
(I made this tool to help creators)
Useful Links/Commands:
Discord Link - / discord
Summit - • Keynote: Linus Torvald...
#opensource #linus #linux
Will there be more or less maintainers in the future?
I think this is more of the skill is dying out like the dinosaurs.
I hope more. I'm trying to learn C so one day I can help out. It has to be done. Bad actors are salivating just waiting for maintainers to blink so they can do something nefarious.
I think that kernel will be migrated over rust and will attract more maintainers
@@kelvinpina3392 Rust will not replace C. It will be used for smaller things.
a Linux Jesus would soon appear
Linux won't end, but companies will try to move in and take it over. This is my biggest fear. People need to get involved; donate money, learn C or rust. The kernel is the biggest component. It must stay free.
Well said
I totally agree.
People need money. Working for free ain't gonna do it.
@@eckee they meant free as in freedom not free as in beer, are you even paying attention or do you just want to argue?
@@eckeeThe top Linux maintainers are paid. In addition to numerous corporate investments, they also receive yearly donations from the Linux foundations.
Linux kernel developers should sell books on kernel development and write journals. Not just preserve their code but their thoughts, hopes, and plans.
Nice pun there. 🥳
totally agree. The entry into the linux world is extremely difficult, non-responsive and frustrating.
@@lerneninverschiedenenforme7513 Linux is its own ecosystem even though in the past 10 years there's been great effort and success in merging the worlds of Linux and Windows. If you want to get into Linux it requires some effort on your part. More of my point is if the kernel maintainers want their kernel to outlive them they need to draw up idealist plans for future generations to reference.
Y'all might wanna check out this book: The Art of Linux Kernel Design Illustrating the Operating System Design Principle and Implementation (Lixiang Yang)
Shit I am 60 and have made a living with UNIX/Linux for 32 years. I hope to work 15 more years. I enjoy learning news things... I hope to died while learning -God willing
Great
The day we stop learning is the day we die ..
We don't die.We just left the physical body.
It's not about learning. The problem will be younger people unwilling to donate work for the benefit of others.
@@fctoryReset Zombie 🧟♂️🧟♀️🧟 😅
Kernel releases have been quite frequent since 2.6. Especially version 6.0, 6.1,6.2 ... 6.8 in about 12 months. I'm just a simple user but I could imagine that amount of work that contributors put in.
And we all appreciate the devs and maintainers for all their hard work bringing the kernel to where it is today
started linux at 2.6 :)
Tbf, linux kernel versions basically mean nothing, heck linus once bumped a major version just because "number were too big"
But yeah, the fact that contributions are a lot, is indeed true
@@monki_sudo I think in some many cases the contributors don’t have access to proprietary info from vendors to write specific drivers.
the version numbers dont really indicate any major progress, they just switch numbers whenever they feel like it lol
Take a break. The system is damn good and stable now. I'm running 6.6.10 and its better than ever. I stayed on 5.15 for several years before even considering updating. If it ain't broke.... I know it's got to be a ton of work doing what Linus has done but damn its appreciated. I feel so much better using Linux over windows. I really have a hard time anymore going back to windows not because its not a good operating system (that spies on you constantly) but because it looks so bad compared to Linux. This most important thing you can lay down Linus is a guideline or a Constitution about how the system should be maintained. Put that with a Bill of Rights and let nature take its course. It should be a guiding set of principles that everyone should strive to follow.
I see people saying, of someone must step up and learn C, Rust.... but that phrase "SOMEONE" worries me. We all know we need more and that SOMEONE should do it - but not me. How many of you guys are ready to sacrifice your personal time to study? How many of you will stop watching so many tv shows? going out with friends? playing video games, and start studying? I have a feeling that not many.
One who cares about it. I have already started to learn linux, c, rust and dual booting with arch Linux.
You live in another universe or is just too naive about the world.
Our lifes are not a matter of willingness, is about material conditions, most of the time to survive.
Why would I spent time trying to be a professional in a subject that is more complex, have less job opportunities and I will have to fight with people who never had the need to work for a living, which will have every facilitation in the process to beat me? Instead I can studying topics that pay sometimes even more, have 100000x more job opportunities and not enough competition or subpar competition most of the times??
Is a non brain move for someone in need(which is majority of humanity). If people had the choice, if they had basic needs met, like housing, health, food, water, fun, etc, you would certainly see more people willing to dive deep into this, but that is Alice's world not ours.
In your own comment there is some conceptions that are not true for the majority of humans nowadays, there is no such thing as "personal time" for most people, we already work 1/3 of our day, sleep the other 1/3, and the last 1/3 we eat, sometitmes we do studies in our own current work field to mantain our jobs and survive, we do chores, we take care of our family, etc, and in the end of the day we are not machines, we need to have fun or we literally die.
Your world view is one of the tiniest I have ever read.
@@davidchristenes9062 What you said is true, but… very sad. I am certainly one of those who would spend decades of their life working on open source projects, but I need a home, I need internet access, I need food, I need healthcare. And I failed to get a position related to the field of computer science, so I am now trying to get a position in government bureaucracy (I am not American). Once I have basic needs, maybe I will have at most 3 hours a day to do what I want (gorvernment workers work 30 hours per week). I use a lot of this time to study subjects I like, but I won't ever have enough knowledge and training to do things beyond very simple and straightforward ones, like fixing obvious typos, reporting bugs, translating (I speak Portuguese from Brazil), run a server with simple pages with articles I may try to write, create very basic games and so on.
@@davidchristenes9062I really love learning computer science, but companies don't want love, they want real work. And I was not capable of really work. I will not compete with millions of people trying to get a computer science entry-level job, this is too much humiliation for a mentaly disabled person like me. I will just try to get another jobs, one that makes me capable of buying housing, food and healthcare. If someone else provides me these basic needs, then I will be able to use all my time contributing to free software until I die.
Not everyone is the same. Some people have talent for programming.Sure everyone can study basic C but only a minority of them will became real programmers. Amongst them, only a few will be good enough to do kernel level coding. So yes, "Someone" should do it - someone who is really capable
Succession planning is always hard. Especially in volunteer organizations.
How do the maintainers earn a financial living?
I believe most of them are part of other companies that work on the Kernel for their own projects and products. Like Red Hat or Google.
IIRC most maintainers (and Linus himself) are paid by The Linux Foundation, which is in turn funded by companies who financially benefit from Linux's existence.
many work for free.
There are so many Linux distros, which means that maintainers are spread very thin. I always thought it would be better if there were fewer versions of Linux, so more attention could be given to the popular versions.
It's ridiculous. Imagine 20 versions of windows..
@@monki_sudo for who? Techno nurds? With such attitude, no wonder linux will never compete with Windoze or Apple.
@sudo but 600 is bit too much as well given limited user base.
Many of those distributions are bold enough to try something unique that the mainstream is too afraid to do.
It's why Linux has so much progress even with much lesser amount of resources being put into it's development.
Some of those ideas actually work and later become implemented into the mainstream.
Windows doesn't allow tinkering, you just use your device the way that "gods" intended for it to be used.
It's just sad to see Linux being its own worst enemy...but it unquestionably is.
Give college credit for Linux maintainers?
Open Source is safe in practice and principle. People will rise to fill the need
Probably not.
I like to be optimistic
Always their good people who trying to limit evil actor ability
Honestly as a normal user, I probably don't appreciate the immensity of the task. But I think that funding is the biggest problem in open source. If the future generation don't see a clear path to financial stability I doubt they will stick around and become maintainers.
Maintainers are paid (and paid well) by The Linux Foundation, which is in turn funded by it's member companies, which are companies that choose to financially support the development of Linux because they financially benefit from it's existence (AKA most companies that deal with servers). I don't see that going away anytime soon.
The problem is not funding, it's mindset. As Linus points out, most new kernel developers/maintainers gravitate toward the parts of the kernel that don't have to be so perfect, mistakes are okay, etc. The hard part is finding people that are both capable of doing the job, and who has the mindset like the old timers to really be picky about the code that enters the kernel.
@@Daktyl198 My point wasn't about the developers that actually become maintainers it is about the young developers that first start out. How many payed positions are there in most open source projects?
As someone that provided critical software for a community, thanklessly and without monetary payment it's what killed development on that project.
People wanted more, for nothing, I felt so burned out and unappreciated that it killed my drive to produce. It's taken years to force myself into a better mindset to tackle the coding challenges I used to love. I'm a C/C++ programmer by hobby, it's the best skill I have.
It's a shame I'll never be able to get paid for it and if I did I'd be miserable I'm sure.
The little IT Admin work I did killed all drive to help others with their IT problems. Just 2 years of corporate IT admin did that.
People just really do not understand the level of technical skill needed to keep up with these technologies they take f or granted, one day I fear there will be so few of us that the technology will start to fail around us as those old maintainers pass away.
Don't even get me started on getting others to learn and contribute to projects either...
It's drastically hard to maintain a behemoth project so it's too hard to join and think what you will gonna distribute to this project. The new programmers that joining is mostly relay too much of the prefabricated frameworks and library so deep knowledge of the programming become so thin. Only the true enthusiast gonna spend more brain cells to join the maintaining.
The world of software development has changed. When we started off we were the odd ones out, the only people who could do this stuff. Computers were for nerds and us programmers were the kings of the nerds. It was a rare skill and there was a feeling of belonging to a special group. Nowadays it's just another common profession. They're not gonna do it for free anymore just to get the geek cred. It's a natural evolution.
New developer here. I really don’t give a f*** about being a nerd. I love programming and the open source, so I’m open to contribute and hopefully maintain. You old folks are gate keeping knowledge and for what?
I see it the other way around. What do people do when they get that old? - Retire from their jobs and spend more time doing their hobby, so they will continue to work with the community.
does the kernel really need to be updated as often as it is? what would happen if they just pushed only security updates for a year? i imagine this team feels like they are responsible for keeping the whole world up and running. and if they stop for a day then all servers will come to a grinding halt and humans will cease to exist on this planet (joking...a little). there has to be some sort of compromise they can make to ease the burden no?
I have some Intel Arc systems that are new and they won't boot with older kernels. I use the latest kernel because my only alternative is to use it or Windows. The rate of technology begs that the kernel drivers at least are updated or Linux falls behind other operating systems. Telling someone they have to wait six months to buy the updated hardware or just use Windows till our next update will just lead to a 'no thanks' from a lot of people in regards to using Linux.
Unfortunately, yes. Linux has a very questionable system whereby drivers are compiled directly into the kernel executable. To update any kernel-driver code for drivers (e.g. GPU drivers) it requires an entirely new kernel release.
@@Daktyl198 use kernel modules to add drivers, I've got modules installed for AMD CPU sensors and Xbox wireless controller adapters.
Linux "build the drivers into the kernel if they're stable" means you have secure and stable drivers running in your PC as soon as you boot Linux without having to go out and trust that the random driver package you download is real or not.
as well as having it ready to go cuz you need to get stuff done @@ThePlayerOfGames
...i remember a time when 2.6 was a good kernel and then they changed the schedules.
...i remember a time where programs didn't get a "full version-hop" for a minor fix.
...i remember when quality was the goal. and not quantity.
...i remember when the sourcecode INCL. MAN-PAGES were well documented. and "new guys" could "learn to code".
...i remember when maintainers weren't hostile to "new guys". and there were much more people involved in dev.
...i remember when developers wanted to include users, and not exclude them. leading to non-documented code not doing what the users need thus one version after the other must be pushed out fixing issues that wouldn't have appeared if devs would talk to users aso.
i remember that time; it was the golden era of linux, bc back then, linux was different than the BSD (who back then acted like linux-maintaiers and -devs act today)
This video surprises me given the accomplishment of Linux in our life. It is the most used OS utilised in the world (Android, server). At the same time it is ironically unknown by the vast majority of people. Imagining this project to disappear is appalling. I want to believe it will survive because of its predominance and importance in so many fields. Future will tell.
So Linus thought that many of the Linux developers suck at programming, and at the same time they worked too hard after all that bad work? What the heck is he expecting from his contributors?? Not everybody can be Torvalds, Carmack or Wozniak. Some people at the top are out of their damn mind.
Linux community is not toxic. Toxic are some people safeguarding their precious distributions, I'm looking at you Arch community.
I don't know about open source maintainers of linux alone, but in the near future we will be solving the open source provenance, remuneration and in general resource allocation mismatches that currently exist. This will ride on the back of the next wave of revolutionary version control software, just as git itself broke us out of the closed-source only paradigm that existed. It took the technological leap of git to make open source really work, and it will be a technological leap that will get us to the next level.
I'm curious how the new rust additions solves what I feel is a sort of contagion of crates in rust where a certain bloat happens. At least user space apps easily seems to suffer from this. Or is rust used so lightly that every component use special adaptations slimmed down for kernel use?
Ada would be an excellent choice to reduce the maintenance burdon. Ada is also the best language for hardware registers and network protocols.
we should move to microkernel architecture like minix or redox, more redox, in rust we trust, and it is much small and much easier to maintain, and less steep lern curve for new comers
People tend to forget one thing that while softwares can be FOSS, some services are kept paid to sustain the software as a whole and especially, a "wage" of developers. I have no problem with any paid service as long as the cost is reasonable with respect to the services provided. Also, a bunch of softwares can have some premium features like MS 365 scheme, but if subscription amount is fair for both ends, then all is good
Even not professional programmer, programming is still a fun thing to do.. Not a sacrifice at all to ponder few hours on certain code,.. Looking forward to join the kernel team in 2 years.. Thanks Linus..
I don't have time for contributions, this is something that I will gladly do in my retirement. Right now I fully focus on my software development career and the world of software is constantly changing and evolving. But, everyone who uses Linux as an everyday tool as I do and can afford it should consider donating. In fact a small sum of money to every open source good quality software that we use is not something that we as a tech community can't do. The fact that the alternatives for commercial software that is more and more intrusive, unfairly licensed and keeps messing with our private data is something we must really appreciate.
Is the ethos that resulted in the development of Linux still present in younger generations? The free spirit of sharing and creating something free for all to use isn’t something I see much of anymore. I see a lot of people trying to monetize everything and make a quick buck. I read a story of a company that took pictures free from the library of congress, charging companies for licensing and cease and desist letters. I hope Linux gets passed on to a solid group of younger people with the strength and determination to fight off vultures and grifters.
What happens if development stops for say 2 months? I'm really curious because if he leaves on a stable release, what's the problem?
New hardware doesn't work or keeps working partially and Linux adoption keeps falling behind. Now it is even more important as worse spyware is recently included in Windows such as AI stuff that records everything you do, and privacy invasive laws are being passed were governments can check anything in your devices by default. Look for chat control law in EU for example. We also need more Linux in mobile devices to avoid such crap.
Linus is the man. He will be severely missed when he moves on.
So i guess what Linus wants to say is that it's over.
Where do we go next, now?
It's far from over, but the writing is on the wall. Android, and Arm with probably Adreno GFX will eventually take over. So the Linux kernel in some form will live on but the contributions from actual linux will become increasingly smaller as they diverge further apart..
@@grndzro777 Linux already supports arm up to some point and development is in progress. Many people will prefer an OS that doesn't come with Google bloatware and spyware.
Over time, converting even core parts of the Linux kernel to Rust would have the effect of making it simpler to understand the intentions behind the code, making it easier to attract fresh talent. For instance, Rust's strong type system would make it easier to navigate the project, and its module system combined with its zero-cost abstractions would make it easier to break down the various subsystems into smaller parts without missing on performance.
Rust is not the only language with those attributes.
Even modern c++ is better when written well
Why does Linus suddenly look 20 years older?
This is the updated version
As long as Linus has a similar-minded leader lined up, I don't fear for Linux. As long as you can keep the big companies out of the final decision making, I believe the Linus-types will lead Linux into continued eras of success.
Well I have ben a Windows user since 1982 and have had every OS since then. I have tried Linux in many different forms Mint Cinnamon being my preferred version however I am only semi savvy and do have a problem with the way that one has to navigate around in the Linux systems particularly that sudo command. Having said that I have also wondered how these different versions are maintained seeing that in my own opinion (there are perhaps too many versions) it must cost something somewhere along the line to do so. This discussion must be very concerning to those who use the Linux OS's as their main base OS and I am now wondering if Linux will not be around for much longer??
i agree partially - the fragmentation has gone too wild. especially on the "debian based"-distros. they all do the same, try the same, and redo the same steps and coding and improvements aso. every distro that wants an own windowmanager builts its own distro (instead just focussing on the WM and making it installable on other distros)
...lots of overhead generated due to that...
i can see the advantages of "specialised distributions", but how many attempts of the "best desktop-distro" is needed until the maintainers of said distro start working together and reducing the workload for each other? (especially if they're all bases on the very same distro!?)
but well - asking this question is very dangerous ...
Nobody will need to worry about that because the bigger problem is many distros like Ubuntu are killing the system by choosing to default on slow
snap package installers over the standard formats , also they continue to use
heavy desktop environments over a basic good quality gnome desktop which
is making many people leave the Linux environment.
companies like canonical keep choosing to do what they feel like
the regular user wants , while all along knowing the end users don't want
what they are trying to push .
What I find odd is current developers are experiencing burnout and aging and so more developers coming in could be useful….and then the foundation cuts spending for Linux development
If you pick apart a C executable you realize that they are patching in security checks and memory scalability in a context where the ABI on Android is just pure crazy Looney Tunes levels of backwards compatible insanity. So "runs everywhere" becomes a burden which I guess they are trying to fix with rust.
I always hear this argument, and perhaps it makes sense going forward, but how will Rust solve the problem of maintaining the existing C/C++ codebases?
Rewriting them.
Hmm...that is, assuming different bugs or other unexpected outcomes are accumulated at an equal or lower frequency in Rust as compared to C?
There's us, we will keep it running
I think we all need to step up. I am sure it will happen because I rather not go back to Windows or Mac at in the future.
All the new, up and coming developers and those curious to join, are far too busy developing Text Editors and configuring their text editors than spending more time around these folks learning from their mistakes and contributing or learning how to contribute.
Well, 40 years ago, Richard Stallman was busy developing a text editor. He went on to working on GCC, GDB and GNU Make. Linux kernel development absolutely depended on this tool chain that was created by one of those text editor-focused programmers 40 years ago.
I used Windows for 10 years, and then Linux for 15 years.
You don't go back once you are on Linux. My computer would feel like a toy for children.
I will become mentainer after graduating from college only
it is not "just Linux" it is also the computers going to far! (new gears & no oldies)
hope it wont turn into winux!
gray hair is a must these days 🎉
I have a question. Does Linux and the "maintainers" use AI to improve it? Specifically, has much software been used for Linux that can rewrite itself in response to upcoming needs? I've read about such programs, and am wondering if they've been implemented for Linux. I've never used Linux myself, but for many years have thought about switching.
I don't think they will use AI to do anything in kernel. AI is very buggy and cannot solve very complex problems. It even give you incorrect results for very simple problems sometime.
@@shailmurtaza9082 True. Somtimes it has given me wrong answers to even very simple problems, when I looked personally was quickly solvable by me but ChatGPT was making mistakes again and again.
Sir I can't access my laptop keyboard can I do my programming using ssh with another lap??
You can, but what distro are you using? You will have to enable ssh on the machine, and depending on the distro, ssh could be either enabled by default or disabled for security measures.
I don't think the younger generations have the cognitive or psychological architecture required to replace them. Plainly put. A problem which will only get worse.
Dumb take. Boomer?
@@BGraves Early 90's. It's obvious.
Younger computer science graduates don't know how to program well enough to help maintain Linux!
What’s new? They need experience. A cs degree is not a silver bullet.
I think low level skills will be in need in 10 years again. Even now some industries are looking for cobol programers and such. It is looks like same stuff
Ugh, they could just take it easy. It's not like it requires rushing the development in any way. If by any chance Linux goes rogue way and gets under influence of corporate dictatorship, there will always be alternatives: forks or *BSD systems.
Thank god for openbsd and freebsd
There are literary hundreds of Linux distros and significant amuont of all developers use Linux based os. How can we have not enough fresh maintainers ? is none willing to contribute to open source these days ?
There are too many distro's, WM's, and fractured competing projects. Ubuntu had a chance to unify Linux but they started Unity, instead of putting their might into making Wayland work for everyone.
2:05 sounds like a fantastic circlejerking club we all should join, but no one does.
Microsoft will always be the condender especially now when they are no longer fighting Linux but instead has joined linux. Linux as long a it is free will always be around because even as a hacker such as me I can do what I want in coding Linux that one can not do with any other operating system. Coders will always be there they will never die and challlenges is whst keeps them going.
I used to be os dev. But i get paid now to manage risk and see most folks scripting in javascript. I miss the real foundation.
These guys need to work harder, I can’t afford any downtime, if there are problems they MUST be fixed immediately.
Why should a sane man voice over this video and explain? 😅
STOP WITH 68.9 MILLIONS OF DISTROS.
UNITE AND MAKE A WHOLE OS THAT HAS ALL.
FLAT/REPO/OTHER PACKAGES WHAT HAVE YOU, ALL IN 1 SINGLE FLIPPIN OS EDITION.. NO MORE DISTROS 😮💨
The average recruiter treats experience contributing to open source as a demerit. There's no incentive to become a mature coder or participate meaningfully. It's easier to schmooze your way into a deadend role rubber stamping divide by zero programs.
What will happen after Linus dies?
What fatigue? Lol
Life is too short to worry about all these things , you guys have gone mad 😂 , there will always be something that will replace something that you are using or the way you are using it , it's life.
One thing is all the younger people want hip, flashy, and trendy... and they are all going to mac. Which sucks... I'm 39 and don't follow trends...I been an on n off linux user from 2003 to 2018 and 2018 on I am full time Linux mint user
I don't think it's about trends as such. People are, by default, given these corporate products and never even hear about such thing as Linux in their lifetimes.
I'm a college student in my mid 20s and I'm the only person I know that is interested in system programming. The last semester I was the only one in my class who regularly uses bash. It's insane how disinterested youngsters are in non-GUI environments.
I was born in 2003. And want to be linux core dev.
@@isuckatthisgameCisco is all CLI.
There is less and less people learning how to code or how a pc works and under 30's have never heard of Linux.
Ill serve coffee. 😂
Why cant we use Artificisl Intelligence to maintain the kernel ?
Because despite all the hype about it, AI is actually still really really bad at programming. Right now it's only really useful as a "google search for when you aren't sure what keywords you need to search for to find the results you want".
Because it is a glorified google/ search
Greg's just sitting there getting roasted for no reason
lool
What a wonderful problem to have though. An open source project that has been so successful, for so many decades, and ever growingly so, that it becomes a generational ship issue of maintainership. As far as the future, in the worst case if everything collapses without key people, I see just a tree of forks that people will use. Linux in one form or another is here to stay quite possibly for a century or more as it is so integrated into so many scales of device and infrastructure. The only other person who I think we should go out of our way to commend in this is RMS. Yeah, he has his personal... stuff, but his work is also a massive legacy for humanity. Without his almost 10 year old toolchain, compiler, even emacs (vim guy here), and the GPL itself, when Linus started, the project might not have taken off nearly as sharply because of limited minix utils and functionality. Sorry Andy - at least your OS is used to spy on, I mean remotely manage millions of machines around the planet. So there's that.
I think with the energy crisis and the oil production/stock going down around 2050, linux kernel is our best opportunity to build a low energy consumption computer. It should focus on specific hardware and probably remove some technology : I don't see bluetooth being a standard technology with less accessible oil, copper, and other raw materials
Going this way, I guess you need less maintainers, and it contributes for the future of computing science
Windows machines support low power states. No energy problem is going to be solved with it
I have been learning Linux over 30 years now, sadly my brains full of so many computer things now, I am starting to need cheat sheets for the simple stuff, like a Data Buffer Overflow, I think ... However maybe I wonder if anyone has looked at or investigated a A.I to "Maintain", instead of Flogging you crap you don't need in Windows, maybe the Linux version could do this job or Help Organics manage it easier ..
Well, if it pays well, its better then being a web developer.
When I read linux source code for the first time I said I will never read it again. It's too much for me. Too many lines of code per file. Too many functions which are not documented. Overall looks like spaghetti.
The code is the documentation. It's right there.
@@1pcfred well good luck Linux to encourage new generation of young developers who will read that documentation
Well, I already know C/C++, but them putting it in Rust is likely to be a reason I don't contribute in the future. I have no interest in learning Rust, and frankly I don't particularly want to use an OS that's written in it either. Some constantly changing language developed by Mozilla that's not very portable, and doesn't have a well-versioned API, with code that constantly bitrots as the language evolves? No thanks. Mozilla can't even develop a decent browser, why do people trust them to make a programming language? I hate to say it, but Linux is making BSD look good these days... I feel like a lot of the problems people were trying to get away from Windows to solve are now kinda happening on Linux too. The moment it gained a little popularity, they started embracing all these normie design trends and ruined it. :(
BSD is already beginning to use Rust, and looking to expand it.
Maybe AI can take up some of the slack so that less people can maintain it.
I think, I may be wrong, butv isn't his name pronounced "Lie-nus" and the operating system "Lyn- ux" ???
The "Lienus" is just an American thing, it's pronounced "Liinus"
@@tinkerjar396 Thanks
From the horses mouth ruclips.net/video/5IfHm6R5le0/видео.html
@@geevee9728 thanks
And the kernel name is similarly "Lee-nux", both with stressed first syllable.
Is it possible to train an AI on all code update on Linux to make it a Linux AI assistence?
AIs most certainly don't have your best interests in mind.
AI is based on prior works. It do not create new code , it just return something someone did before via a weightage/probability system.
@@pehclark7256 There must be enormous amount of GitHub repositories on the development of the Linux system.
What if you train an AI on all that data and then use it as a Linux AI assistance ?
It would probably understand what is allowed or not or ok, if any new code are to be added.
It would be able to explain if new code is ok.
Maybe even be able to give better code because it would understand what the new code is to do.
@@pehclark7256 Linux have a hugh GitHub repository history that a AI can be trained on.
The AI would probably then know what code update work and could probably understand the code update and recommend a better working version.
I bet it is possible to make a Linux AI assistent that can help Linux developers ....
So, maybe it's time that Linux Foundation dedicate more then 3.2% of their resources to Linux.
Do you have a reference for this assertion? If it's true, then it's disturbing. Where would the rest of the money go to?
@@joseoncrack Linux Foundation 2023 annual report. And yeah, now it's down to 2%.
@@CptMarkAh, I checked. You're right. If you look at how the expenditures span out, it becomes obvious that they are driven by the main contributors, which are big tech companies.Cloud stuff coming out on top.
Over 260 millions of revenue, and thus spending about 5 millions on the kernel.
This may look odd, until you look at Mozilla, they are making nearly 500 millions a year (or more now?) and their flagship product, Firefox, only has a tiny fraction of market share. And what they spend their revenue on is, to put it nicely, odd as well.
People have day jobs, social life, kids and do not want to spend hours on end to get Linux software working. Good luck in the Linux rabbit hole. I'm not going to find out how deep it is.
The whole IT world is/has been burning out for over a decade, not just Linux....
it would survive it few linux distros are focused instead of 100s of useless distributions.
Rust represent
Start selling the OS. I wasted 200 in windows 11 pro for nothing because I just went back to Linux...so I would LOVE to buy any distro for about 100
I'm sure Red Hat would take your money. You should ask them.
1:20
Use AI to automate?
According to Nvidia CEO you soon won't need maintainers... How wonderful! (we are all doomed)
Linux is getting old. Time for something new
Too many distros, too many abandoned github projects, too many projects.. full stop, too many apps all doing the same thing, too many desktops, too much argument over the basics like package managers or GUIs.. it's endless. Linux is a mess.
That’s why maintainers are needed. To cut through the fluff and make the useful Gems stick or available.
Evolution is a messy process.
It's so cynical and sad, that the time when it seems that Linux finally achieves the step to the desktop, with Windows dying even possible desktop dominance in not many years, on the other hand Linux is already dead - at least as an open source project. Doesn't Linux have any way to stop the Linux foundation from actually not caring for Linux at all?
Let AI maintain it.
Create a LLM AI from the source code, and let the AI maintain the code. 😅
Make it even more windows like. Then you won't have to beat your head against the wall. When I plug in anything I expect it to play. Not so with Linux !
Why do you expect anything you plug in to work? Linux supports far more hardware than Windows does. But to have the support you do need the driver. Most Linux kernels don't have every driver built into them.
I guess me like most others expect it to be at least equivalent to windows. If you can't do that, don't expect it to ever go anywhere.@@1pcfred
@@icheckedavailability Take it up with the HDMI Forum. They're the ones that keep their proprietary standard closed. It is not Linux's problem. You don't even know who to blame. But that's OK because we don't need users like you. Net negatives. WTG chief!
Programming in C is overkill. Redox is the future. Rust is the successor of C (and the C++ replacement). There's a reason Linus accepted Rust as an alternative language for the Kernel. But, you work with C interfaces in Linux. In Redox, even scripts are made in Rust.
You could setup a nice Ai to help you. Not via open Ai these shills will try to own it. Use the French open source one.
New generation is so lazy they want to wrote everything with AI. There are some gen zs are coding but most dont want to use linux. I am afraid unless there is a way to monetise linux it will never be mainstream.
they are survivors but the job of making ai real and accessible is looming - the hw/sw stacks need to catch up and it will probably be 4-5 years before this really happens but once it does look for smb mkt and prosumer mkt to thrive and produce much more innovation than big tech could with ai - the looming ai realization crisis is sort of an existential crisis for linux but once it hits we should see a renewed interest in the kernel maintenance and associated drivers and hardware and sw - we can at least hope - the graybeards will have to pass the torch to a new generation and maybe even pass it on to ai itself if they want to stay alive and vibrant and vital. lots of things are going to happen and many many innovations will get produced and discovered - now is boring but in 5 years things will get exciting again - semiaccurate conclusion of current and future paradigms