31–40 of 62 posts
I worry about Debian’s future. The infrastructure of the project is slow moving and baroque. Not many people want to learn it to become maintainers. Michael Stapelberg’s blog post about leaving Debian describes the general issue of using tooling and process with 1994 leverage to tackle problems with 2022 scale, does not sound fun: https://michael.stapelberg.ch/posts/2019-03-10-debian-windin... I tried to get into Deb…
How badass is Lennart Poettering? Dude just says alright I'm going to improve upon the craptacular init system. He follows through and the whole world starts using his software. That dude is a legend and he's only 41.
Devuan forked in 2015 and is basically a non-existent distro after 7 years. Systemd was never controversial as portrayed in this blog.
So what exactly is the delusion of Debian?
>But that's not all. The fact of the matter is that Debian has long been experiencing a decline in the amount of people willing to participate in the project.
There was another recent debian post on HN. No idea why Debian suddenly is so popular of a subject. People responded to me like as if Debian was doing fantastic.I just don't know how anyone can think Debian isn't in death spiral.
>Carter believes they need two to three times their current volunteer levels to accomplish all of their goals. He also feels Debian is too lacking in the area of diversity and not catching up fast enough. He blames Debian's diversity on large regions of the world not being represented enough, failing to put together a timely message regarding Black Lives Matter, and other concerns/ideas.
https://www.phoronix.net/image.php?id=2020&image=debian_prob...
Debian leadership thinks they don't have a strong position on black lives matter? That's why people are leaving the project?
So what have they done? They have built a diversity and inclusion plan.
They even enforce a anti-harassment team. https://wiki.debian.org/Teams/Community?action=show&redirect...
I encourage everyone to read that antiharassment page in depth.
As Debian leadership clearly indicated. They need significantly more people to contribute to the project, but they have an exodus/decline of developers. Clearly they are not doing enough with their antiharassment team.
People are harassing others at debian to such a huge degree the remaining volunteers are taking on ~3x the workload they should. Eventually they will quit as well.
I think what debian needs to urgently do... is figure out exactly who the harassers are. I very clearly can see exactly who is doing the harassment. If Debian makes a mistake and does not directly address this correctly. Debian is dead. Which is so sad to me.
> At the time when Debian 11 was about to be released, PHP 8.0 was 10 months old yet Debian's 11 was released with PHP 7.4. That makes PHP 7.4 the standard in Debian stable for at least 3 years after its release. But PHP 7.4 only gets upstream support until 28 Nov 2021 and security support is permanently ended at 28 Nov 2022, which is seven month from now. This means that unless Debian has some really good C develope…
But it doesn't work like that. Packages have reverse dependencies. Just updating PHP to 8.0 would have broken all reverse dependencies that didn't already support PHP 8.0. So it isn't just a case of shipping the newer version of PHP; all upstreams of reverse dependencies need to have shipped updated versions, too, and all of those need to have had their packaging updated.
In practice distributions patch support in to laggard reverse dependencies or even remove them to speed things up. But still, the actual work involved is much more than the naivety in the statement "PHP 8.0 was 10 months old yet Debian's 11 was released with PHP 7.4."
A second common misconception is that just because an upstream declares some kind of support period, distributions have to follow them or it's somehow a problem if they do not. Upstreams even having a declared support period is the exception, not the norm. Distributions have been declaring a support period for their release as a whole long before some upstreams started doing this.
I worry about Debian’s future. The infrastructure of the project is slow moving and baroque. Not many people want to learn it to become maintainers. Michael Stapelberg’s blog post about leaving Debian describes the general issue of using tooling and process with 1994 leverage to tackle problems with 2022 scale, does not sound fun: https://michael.stapelberg.ch/posts/2019-03-10-debian-windin... I tried to get into Deb…
Any idea what’s keeping it from doing a revamp?
In other words, I'm questioning whether a revamp is needed at all (as opposed to incremental improvements which are happening all the time). This article might be making some claims in that direction, but has had most of its points debunked in this HN thread already.
It’s a shame that after so many years and so many distributions the community could not come up with a good model for a low maintenance secure distro. IMHO the author of this post is surely the kind of people that eventually only care about their job security and just stir the water without making any progress…
Earlier quoted context omitted.
It is useful: it points out why one should stay away from Debian GNU/Linux. Sometimes, knowing what not to do is far more useful for making the correct choice.
Tearing down volunteer effort is fundamentally problematic. Yes, it’s important to not give anything flying under the banner of volunteer a free pass, but a huge project with decades of work and use like Debian is well beyond this. If you’re going to make criticism make it constructive or just move on with another distro. Edit-The easiest way to make some points on the internet is just simple criticism. Doesn’t even…
> At the time when Debian 11 was about to be released, PHP 8.0 was 10 months old yet Debian's 11 was released with PHP 7.4. That makes PHP 7.4 the standard in Debian stable for at least 3 years after its release. But PHP 7.4 only gets upstream support until 28 Nov 2021 and security support is permanently ended at 28 Nov 2022, which is seven month from now. This means that unless Debian has some really good C develope…
A common misconception is that just because an upstream released some time before a distribution release, the distribution ought to have had enough time to update, and a failure to update is a failure on the part of the distribution. But it doesn't work like that. Packages have reverse dependencies. Just updating PHP to 8.0 would have broken all reverse dependencies that didn't already support PHP 8.0. So it isn't ju…
This article is full of inaccuracies and errors. > Due to the sheer size of the current Debian release it is infeasible for a small team to be able to audit all the packages, so there is a system of prioritizing packages which are more security sensitive. This was, at best, poor communication. Of course nobody would ever audit all of 90,000 packages, easily billions of LOC. Especially not when the vast majority of th…
Devuan is in rank 32 on distrowatch. What distros haven't moved to systemd at this point? I would appreciate a list so I never use them thanks. How badass is Lennart Poettering? Dude just says alright I'm going to improve upon the craptacular init system. He follows through and the whole world starts using his software. That dude is a legend and he's only 41. Devuan forked in 2015 and is basically a non-existent dist…
The one that has the biggest impact on my life is Alpine, because I run postmarketOS on my phone which is an Alpine derivative. It has OpenRC instead, which doesn't have the ability to auto-restart crashed services (it doesn't do any action when a service crashes, by design), doesn't do socket activation, and doesn't support running as a user instance for a user session.