Did anyone think to actually ask the developer who is maintaining the LTS kernel versions why he made that change (back in February?), i.e. me?
{sigh}
No, I guess that would take too much effort, and wouldn't result in such a click-bait headline "LTS kernels are no longer supported for 6 years because it turns out no one used them." doesn't have that same fun sound...
> LTS kernels are no longer supported for 6 years because it turns out no one used them.
Looking quickly at Debian, old old stable (10) uses 4.19, old stable (11) 5.10, and stable (12) 6.1. All these versions are LTS kernels.
I don't think Debian uses LTS kernel branches as is, instead cherry picking patches, but it seems they do leverage LTS branches, at least to ease the back-porting I presume.
I'm probably missing something, and looking through the Debian mailing lists, I've not seen any discussion about the end of long term LTS kernels, but it seemed they did rely on it to a degree.
Maybe it was more of a nice to have, but from the outside (I'm neither a kernel or Debian maintainer), it leaves me a bit confused.
The one does not discount the other. In other words: disregarding the lts misunderstanding it still appears true that the Linux foundation is abandoning Linux kernel development. At least financially. Worth a look I guess.
Also: thank you for all your contributions to Linux for all these years!
The amount of resources and other stuff that the LF provides to the Linux kernel community has increased over the years, including last year. Just because new people are brought in with new projects (that the LF member companies want to host) does not mean that somehow less is being given to the kernel community at all. It is not a zero-sum game here at all, that's not how the LF works in any way.
Again, this would have been easy to verify if someone just asked us.
So to repeat, no "abandonment" is happening here at all, the opposite is happening, just like it has for the entirety of the LF's existence, support has grown every year.
Can you explain what this means? E.g. Debian very much sticks to LTS kernel versions in their stable releases. Were they not working with upstream, you? If they'd be doing their own maintenance, there's no need for them to stick to LTS versions, but clearly they didn't tell you. Any idea what's going on there?
> LTS kernels are no longer supported for 6 years because it turns out no one used them
I assumed companies use a lot of LTS software for servers to avoid frequent upgrades that always carry some chance of breaking stuff. Could you please elaborate more or link me to some place where I can read more about this?
A proven and repeatable way to scale project communities via a comprehensive portfolio of support programs for aspiring industry leading projects:
Neutral home for code and collaboration - We aim to democratize code and scale adoption, for all projects.
Ecosystem curation and community building - We strive to create new technology categories by identifying trends, accelerating the growth of nascent technologies, and removing barriers to adoption.
Enterprise ready, the OSS way - We provide turnkey technology and support programs for developer enablement, business operations, training & certification, marketing and events, and membership development to help projects scale fast.
Project insights and management tools - We help projects streamline operations and boost community engagement with cloud-based, collaborative tooling, contributor and participation analytics, and infrastructure management.
Lunduke has been picking so many fights, I'm a bit lost as to whether I should pay attention or not. Sure, it doesn't seem great, but the Linux foundation putting money in other areas of the ecosystem is hardly a good point. Show us a real breakdown, instead of a gratuitous Linux kernel expenditure vs not-Linux kernel expenditure.
I remember when I discovered the Bad Voltage podcast he was a co host. I knew nothing about him at the time, but I remember thinking good Lord why is this guy allways a contrarian asshole. Since then he's left the podcast and it got so much better and I've learned who he is and decided I really don't like the guy.
But the author and many others aren't wrong in highlighting this and have covered the LF issues for years.
Given that is from a guy that cares a lot about how the Linux Kernel project is funded, that is very concerning.
The first step in solving a problem is to identify that there is an inherent issue, which the author has done, but goes ignored by those who do not wish to care about the project.
I agree, Journalists should pick fights, but I would prefer if they built up their arguments a bit better than this.
I am one who is not familiar with the author previously, and from what he brings up I am highly inclined to agree with him, but I have to agree I became more sceptical to his arguments when all he brings up to support it is one very sparing data point (that graph feels very misleading). Give me data on how the kernel support has changed over time. Give me data on how the foundations income has changed over time. Not showing me any of this makes me very sceptical. Would showing this data not support the authors point?
This is weird. LF has spent less on Linux than everything else. LF has to deal with lots more than LF. Even based on the numbers this article mentions, LF spends close to a million a year on Linux. I don't understand what the aim is here.
How many engineers do you think are actually managed by LF? From my understanding, the majority of engineers contributing to Linux are on the payroll of other companies like Google.
I would say even less because you have to take into account all the costs of hosting, dev machines, laptops, testing, travel etc. $1 million for such huge project is peanuts this days.
I think this is really just a naming problem. The Apache Software Foundation was founded for the Apache webserver, but today they have hundreds of projects unrelated to that. The same is true for the Linux Foundation.
Good point. In a sense, foundations have to appear much more interesting than they are nowadays. Developing an OS kernel or a web server isn't appealing to newcomers, so I guess there should be alternative pathways to that. There should still be alignment in behaviour to the identity the entity claims to be. Or we just could have the Whatever Foundation. So, are these other projects in any way related to the development of Linux kernel or not?
I always find a kind of ironic that Linux Foundation is the home of Zephyr, an embedded OS that has nothing to do with Linux, other than being competition against Linux distributions for the embedded market.
It isn't only Zephyr, also being the home of several other projects completly unrelated to the Linux kernel.
> Zephyr, an embedded OS that has nothing to do with Linux, other than being competition against Linux distributions for the embedded market.
Zephr isn't "competition against Linux". It primarily targets microcontrollers that Linux would have no hope of running on (e.g. parts with <1MB of memory).
It certainly is, given that it isn't the only target, if you bother to actually watch their talks, specially its license makes it much more appeling to OEMs than a GPL tainted kernel (from their point of view).
Does anyone know if the Linux foundation makes their financial statements available?
My cursory glance turned up the slick annual presentation type report, but I am looking for the 3 basics, income statement, cash flow, and balance sheet, audited or unaudited.
An aside, I think Linux is a true modern marvel of humanity, it really is people at their best. It's a bummer when in-fighting and drama pop-up, but I guess it comes with the territory and its part of the sausage making.
The 2022 report (which was not linke even though it was the source of this article) doesn't look so dramatic. The listed grievances look like small contributions alongside their main projects:
Readit News
{sigh}
No, I guess that would take too much effort, and wouldn't result in such a click-bait headline "LTS kernels are no longer supported for 6 years because it turns out no one used them." doesn't have that same fun sound...
Looking quickly at Debian, old old stable (10) uses 4.19, old stable (11) 5.10, and stable (12) 6.1. All these versions are LTS kernels.
I don't think Debian uses LTS kernel branches as is, instead cherry picking patches, but it seems they do leverage LTS branches, at least to ease the back-porting I presume.
I'm probably missing something, and looking through the Debian mailing lists, I've not seen any discussion about the end of long term LTS kernels, but it seemed they did rely on it to a degree.
Maybe it was more of a nice to have, but from the outside (I'm neither a kernel or Debian maintainer), it leaves me a bit confused.
Also: thank you for all your contributions to Linux for all these years!
Again, this would have been easy to verify if someone just asked us.
So to repeat, no "abandonment" is happening here at all, the opposite is happening, just like it has for the entirety of the LF's existence, support has grown every year.
Can you explain what this means? E.g. Debian very much sticks to LTS kernel versions in their stable releases. Were they not working with upstream, you? If they'd be doing their own maintenance, there's no need for them to stick to LTS versions, but clearly they didn't tell you. Any idea what's going on there?
I assumed companies use a lot of LTS software for servers to avoid frequent upgrades that always carry some chance of breaking stuff. Could you please elaborate more or link me to some place where I can read more about this?
Deleted Comment
https://www.linuxfoundation.org/about
The Linux Foundation Method
A proven and repeatable way to scale project communities via a comprehensive portfolio of support programs for aspiring industry leading projects:
Neutral home for code and collaboration - We aim to democratize code and scale adoption, for all projects.
Ecosystem curation and community building - We strive to create new technology categories by identifying trends, accelerating the growth of nascent technologies, and removing barriers to adoption.
Enterprise ready, the OSS way - We provide turnkey technology and support programs for developer enablement, business operations, training & certification, marketing and events, and membership development to help projects scale fast.
Project insights and management tools - We help projects streamline operations and boost community engagement with cloud-based, collaborative tooling, contributor and participation analytics, and infrastructure management.
Given that is from a guy that cares a lot about how the Linux Kernel project is funded, that is very concerning.
The first step in solving a problem is to identify that there is an inherent issue, which the author has done, but goes ignored by those who do not wish to care about the project.
The Linux Foundatiom should an engineers-first organization.
The leadership team breakdown shows me it’s full of MBAs specializing in “on a computer”.
I am one who is not familiar with the author previously, and from what he brings up I am highly inclined to agree with him, but I have to agree I became more sceptical to his arguments when all he brings up to support it is one very sparing data point (that graph feels very misleading). Give me data on how the kernel support has changed over time. Give me data on how the foundations income has changed over time. Not showing me any of this makes me very sceptical. Would showing this data not support the authors point?
why?
what do you feel it's purpose is? do you think the people who give it gobs of money agree with you?
... for that company's own interest.
> I would argue Linux barely needs anything from the LF.
Linux? No. It's a computer program. Computer programs have no needs (yet).
Users, real natural persons, YES! Remeber when Linux couldn't play audio without hiccups? Remember Con Lolivas patches?
Have a look at linux kernel maintainers. How may systems in there are "supported" vs. just "maintained".
Deleted Comment
It isn't only Zephyr, also being the home of several other projects completly unrelated to the Linux kernel.
Zephr isn't "competition against Linux". It primarily targets microcontrollers that Linux would have no hope of running on (e.g. parts with <1MB of memory).
My cursory glance turned up the slick annual presentation type report, but I am looking for the 3 basics, income statement, cash flow, and balance sheet, audited or unaudited.
An aside, I think Linux is a true modern marvel of humanity, it really is people at their best. It's a bummer when in-fighting and drama pop-up, but I guess it comes with the territory and its part of the sausage making.
https://project.linuxfoundation.org/hubfs/LF%20Research/2022...