I regrets Debian/kBSD...

[Danielsan]

A bunch of months ago I read an interesting article about called "Why Open Source Failed", and I found it really sharable, it is true it is what RMS repeats as a mantra for years, you need an ethical approach even of software otherwise you are going to lose freedom. Most companies use floss, foss, open source software to sell service that aren't open (almost all) or do not return anything back.

Most companies are focusing on the Linux Kernel but not because the community or because the freedom just because they understood how to make money with it.

At Linux North Fest 2018 Bryan Lunduke made the point about that: https://invidio.us/watch?v=TVHcdgrqbHE

The corporations are seated in the Linux cockpit and are leading the Linux development far away from freedom and the users community and since the broad adoption of systemd we do not have any kernel alternative, we don't have a plan B and I regrets the lost of Debian/kFreeBSD. The most ridiculous thing is with Microsoft seated on Linux Board Director the only GPL alternative to GNU/Linux is ReactOS. The absurdity is the corporations prefer the GPL instead of MIT or BSD licenses, so what was at beginning created for the goods of all has became a way for the corporations to defend themselves from the community.

Now we are stick with a giant and fat kernel without alternatives and this also our fault...

[CwF]

I'm not sure I see a problem in the kernel involvement in the coming downfall. Can't we still build our own? Won't always be able to?

I see the issue more mid-level, when the programs we'd like to run on our custom kernel creations require hooks into/onto the internet and the model is terminal/server and not my computer.

At the point of evolution where a PC acts like a smartphone and needs to check in on the net to do anything, then what does the kernel matter anyway?

[n_hologram]

At the point of evolution where a PC acts like a smartphone and needs to check in on the net to do anything...

I disagree with the overgeneralizing statement about the kernel's significance, but I did want to acknowledge how smartphones have shaped the way we interact with software since even this time ten years ago: I'm thinking of the number of specialized programs that have been replaced with web apps and the like.

It's also interesting how smartphones have amplified closed-source issues which bypass any kernel. On the hardware level, we're still dealing with both closed-source firmware/microcode and proprietary hardware (which contains the microcode), and we now know how easily those can be exploited. While kernels can be hardened (I'm thinking of openbsd's), hardware exploits linger. And that trend seems only to get worse with every year, every smartphone, everything "smart" device -- all of these pose new vulnerabilities way below the init or kernel. The current state is, best-case scenario, the kernel mitigates the lingering shortcomings of a hardware/microcode that cannot be easily altered (or altered at all).

Now we are stick with a giant and fat kernel without alternatives and this also our fault...

I always forget how ingrained systemd is becoming with debian until someone brings up a point like this one. Nevertheless, there are a number of systemd-less projects, as well as custom kernels (the linux-libre project removes a lot of binary blobs). One could also build/harden/prune a kernel catered for specific needs. Then there's always jumping ship to projects like freebsd/openbsd which use subjectively "better" kernels. It is ridiculous that the heads of Linux/-based projects ignore or are impotent to these changes (I'm thinking of Google's shady tactics against grsec), but there's still hope.

*Edited for reorganization and to expand on a new point*

[pylkko]

The future will probably see all people increasingly using computational services like voice recognition, machine learning algorithms that adapt to user input and the ways of life of the user themselves. These services will likely be running from a server at Google or similar. 5G-mobile networks are already up and piloting and in the future all devices will be on 5G and have "smart" sensors that rap around the user's way of typing, using a camera, their voice and therefore mood, motivation, but also physical location, are they paying attention to an ad or not etc. Wifi or cable network will be something that Grand Dad did as a kid. Mobile network that is impossible to turn off... "haha, he used to have to access a network before using the internet"

I think the problem is way deeper than capitalism, or unfair distribution of wealth. Technology gives the "feeling of power". Even a lowest person (by any measure; income, political influence, sexual appeal) can "feel that they can do stuff and have power over things" on a mobile phone app. There is no room for revolution when all the peasants feel like kings.

[Danielsan]

Well everything is really cool however we still need an alternative for Linux, and if you read about ng-hurd you get depressed

Hurd-ng is an effort to build a new operating system that preserves the main design goals of the Hurd while fixing some of the Hurd's shortcomings. There is not yet an official roadmap or a concrete specification; indeed, much of the work is research oriented.

[pcalvert]

The corporations are seated in the Linux cockpit and are leading the Linux development far away from freedom and the users community and since the broad adoption of systemd we do not have any kernel alternative, we don't have a plan B and I regrets the lost of Debian/kFreeBSD.

Debian/kFreeBSD is not dead. Development of it has not stopped.

I just picked one package to see when it was last updated. I chose the light, fast, web browser NetSurf. Using wget, I downloaded the package from a mirror that I found on this web page:

https://packages.debian.org/sid/kfreebs ... k/download

Code: Select all

$ wget -c http://http.us.debian.org/debian/pool/main/n/netsurf/netsurf-gtk_3.6-3.2_kfreebsd-amd64.deb
$ ls -l
-rw-r--r-- 1 user user 1238156 Jul 23 23:42 netsurf-gtk_3.6-3.2_kfreebsd-amd64.deb


As you can see, the timestamp on the package shows that it was updated only a few months ago.

Phil

[debiman]

I think the problem is way deeper than capitalism, or unfair distribution of wealth. Technology gives the "feeling of power". Even a lowest person (by any measure; income, political influence, sexual appeal) can "feel that they can do stuff and have power over things" on a mobile phone app. There is no room for revolution when all the peasants feel like kings.

quoted for truth!

[Danielsan]

I believe it is a ghost shadow in the system because the automation, if you go on the daily builds there are the folders but are empty.

As you can read from this mailing list the project was dropped:

https://lists.debian.org/debian-devel-a ... 00005.html

Architectures
=============

There remained yes/no decisions for arm64, ppc64el, and kfreebsd.

arm64 and ppc64el have made enough progress to be release
architectures for Jessie. Britney no longer has special handling
for these two. Therefore, FTBFS regressions for arm64 and ppc64el
are now release critical (but non-regressions are not).

We discussed kfreebsd at length, but are not satisfied that a
release with Jessie will be of sufficient quality. We are dropping
it as an official release architecture, though we do hope that the
porters will be able to make a simultaneous unofficial release.