You are not logged in.
Pages: 1
A Microsoft employee recently decided that since Debian is now a systemd only distro that he could help everyone out by modifying one of the debhelper scripts to automatically inject a systemd package dependency even when one may not be needed.
https://salsa.debian.org/debian/debhelp … 9dd3a58b1c
So far we know that this has affected at least sudo (https://packages.debian.org/unstable/sudo), it now requires systemd, systemd-standalone-tmpfiles, or systemd-tmpfiles to be installed. Since Devuan provides a systemd-standalone-tmpfiles package maybe the solution is to install it on every system? Is there a better option?
But rewriting applications to remove and re-license it, is unethical and without honor.
If you look you will find DebConf videos from several years ago before all this blew up where the main developer working on uutils at the time makes it clear that he did care what the license was as long as it was FOSS and his main motivation was to have fun with Rust. We need to be careful not to rush about on witch hunts. Is Ubuntu being dumb? Yes. What's new?
ETA:
Found a post said DD made about this.
When your side COVID hobby project to learn Rust ends up behind Microsoft’s CEO Satya Nadella at Build 2026…
Maybe you did something right? 😅
Great moment for Uutils, Rust, and open source!
Yes, I previously have watched the video of that presentation and at one point your voice can be heard in the background.
That doesn't sound very minimal to me. How does it relate to that program that the OP got made?
As @SteveM vaguly indicated, "minmal" would mean that the thing is no greater in any comparison dimension than other of its kind. So I misunderstood your meaning.
Come to the meeting in 6 hours so I can explain to you how it relates to the OP. https://vdc.dyne.org/devuan
In the mean time stop pretending like you just cleared up poor use of language when the following is easily found on the devuan.org website.
The core of the vision remains the same today: Devuan is a minimal, stable system that honors Debian’s legacy and embraces innovation while maintaining backwards compatibility and interoperability. In this context what dimension is Devuan measured to be lesser?
Maybe these release notes for the minimal ISO will jog your memory on the meaning of :minimal" https://files.devuan.org/devuan_excalib … l-live.txt
Maybe more examples would help? Aprolla measures 1x1 no doubt.
Amprolla is the heart of Devuan that merges various apt repositories in a minimal and modular way.
Here is someone on this forum posting about a minimal X.org server setup that you neglected to scold for using language that you think no one can understand https://dev1galaxy.org/viewtopic.php?id=7650
Let's look up minimal in a dictionary, the standard we use to define what a word means:
the least possible
barely adequate
very small or slight
I don't know if your dog died or a loved one died or what, but I fail to see how your conduct reflects well on Devuan and is in keeping with the "almost" CoC of this forum. The meeting in 6 hours will provide an excellent opportunity for you to set me straight.
Come to the Devuan-Dev meeting in 13 hours and we can get you up to speed.
What do you mean by "minimal"? ... 1x1 pixels?
Considering that the term "minimal" is well understood in this context your comment not only seems pointless but a violation of the almost no CoC item #2 of "a foundation of mutual respect is expected from visitors to this forum."
A little positive encouragement would not have gone amiss.
I have found that nala mostly works as a replacement for apt. I say "mostly" because it can't do an apt dist-upgrade equivalent and when packages get in a real mess it also lacks the full features of apt to unbreak things.
And here I thought my 11th gen Intel running Coreboot was getting old.
It's also a bit like asking why paint a picture or write a book given the options that exist.
I've reworked the Debian package for log2ram to work with sysvinit on Devuan. Please take a look and let me know if there is anything that I should change.
https://git.devuan.org/SteveM/log2ram
Code contains relatively large numbers of non-alphabet characters, which are not as easy to type.
Yes, and code tends to not contain complete sentences. How useful is 120 wpm when typing out this random excerpt from the Linux kernel?
u64 dma_direct_get_required_mask(struct device *dev)
{
phys_addr_t phys = (phys_addr_t)(max_pfn - 1) << PAGE_SHIFT;
u64 max_dma = phys_to_dma_direct(dev, phys);
return (1ULL << (fls64(max_dma) - 1)) * 2 - 1;
}I must confess to not being fast at typing and never having trained. Over the years I went from the guy using 2 fingers and staring at the keyboard to being able to type much faster, but still only with a few fingers and occasional glances at the screen then back to the keyboard to keep myself aligned on the right keys. I know I should learn to type better and faster, but I don't think it is as big of an issue as he claims.
I am an embedded engineer which means I probably spend more of my time reading datasheets and schematics than I do writing code. Quite literally you need to know from reading footnote 2 of table 5.1 that pins 2 and 6 of each port have a different interrupt behavior than the other pins as described in the asynchronous interrupt section 12.2.3. I spent all day yesterday trying to figure out why a display that was working fine on the development hardware now isn't displaying correctly on the prototype. It is difficult to 120wpm your way out of those kinds of problems. I also interface frequently with mechanical engineers, best to just get out of your chair and go chat with them. I'm sure his points are valid in his niche there suspiciously close to Microsoft HQ.
Just a simple music player, that's all I want it to do. Opens waaaaay faster than other music programs and plays until I tell it to stop.
Nothing wrong with that! No one knows what suites you better than you do and no need for all of that bloat and slowness.
It reminds me of calculators. There must be hundreds or thousands of them out there, but I could be a lifetime trying to find one that suites me. I do a lot of low level work in binary, hex, and decimal and need to quickly convert between them and manipulate the data. The Windows calculator isn't terrible, but I have to switch back and forth between "programmer" and "scientific" to perform floating point operations (very needed in FPGA DSP work). That gets really old fast. Today I needed to do a fairly simple task of reducing 0x1FFF to 10%. In Windows I can enter 0x1FFF in hex then switch to decimal and divide by 10 and see the answer displayed in all of the number formats, including hex. In Kcalc I couldn't find a way to do this other than to divide 0x1FFF by 0xA. And in Kcalc you switch between bases in a drop down menu which is painfully slow and doesn't convert the number in the entry field for you. Imagine if I just rolled my own calculator in the spirit of what you have done, tweaked to fit my own quirks.
I have been using your TimeShift for the longest while, if I recall correctly, since ascii.
To be fair, I've only been helping with TimeShift since 2020 as an uploader and not as a maintainer. The DD that sponsors the TImeshift uploads has been doing the vast majority of the updating the past few years, but I jump in when I notice that a new version is out and he's not updated it yet.
You might get a faster reply to your query via IRC, on Libera-chat go to #devuan-dev, that's where all the big cheeses usually hang out
Thanks, I might give that a try later on. I've not had great success with chat based communication, especially if there is a large volume of conversation. If one party of the conversation suddenly needs to leave then the dialog gets lost in the noise. If there is a large time gap between people in a dialog then it quickly falls apart as one person is trying to communicate while the other is sleeping and the message get lost in the noise.
Hi I seen this posting on the Debian Planet and immediately thought of this thread I read the other day. Especially when I read the mentorship offer part at the bottom of it. Could be worth a try to reach out and see if he can help.
Thanks, I might reach out to him. I don't agree with him that having to build packages on Debian is a problem. How on earth is one supposed to test the resulting package if not on your own Debian machine?! I'm also a little uncertain how containers could be used for some packages. For example, when I think I am done building the Swift compiler the very next step is to install those new .deb files on my local system on top of the Swift compiler that I just used and re-build the package to make sure that I have produced a compiler that can still compile itself. There was one time that I did indeed build a compiler that could not rebuild itself due to some changes in upstream that my packaging needed to account for. I fear that all he has accomplished is to worsen the complaint of "Debian has too many tools to learn, often with overlapping or duplicate functionality." One of the challenges I face is that I work with 4-5 DDs to sponsor my uploads and each one has his own preferred tools and flows that they want me to use.
I enjoy contributing and seeing the work that I do spread out to Devuan, Ubuntu, and their derivatives such as Pop OS.
I don't see it in the repo.
It's in unstable https://repology.org/project/debcraft/versions
I maintain some Debian packages and am often held back from releasing updates due to needing to contact a sponsor to do the uploading for me. One of my packages is going to be released to Debian Stable at an older version containing a bug simply because the DD that normally sponsors that package has not replied to my e-mail for months. I have been working to get maintainer status for these packages, but that hasn't been successful yet as that too depends on DDs that are very busy.
Does Devuan have any interest in having me develop packages directly for Devuan or would you rather that I continue to contribute upstream? None of my packages are particularly important or popular. Here is a list of them for context:
https://qa.debian.org/developer.php?log … 40swm1.com
Pages: 1