Author: t.j.duchene Date: To: Martinx - ジェームズ CC: dng@lists.dyne.org Subject: Re: [Dng]
No,the majority doesn't knows. Long life to the Scientific Method!
>> Seriously though, having come from opensuse I've seen the debates,
>> they stink. Most people don't even have a clue what systemd is, what
>> it does and even more disappointing, what's wrong with it. The
>> ignorance is life threatening. I have read some of systemd's code,
>> it's acctually very well formatted C, but also very, very, glibc and
>> gcc specific. >That's very sad... :-(
That’s true. Just to make a fair comment, to my knowledge, the Linux kernel does not really compile very well or at all without GCC either. Last time I checked, that situation was improving.
>> Here, democracy does NOT work.
Sorry to hear that. If I might make a suggestion, in the US, we have a system of “checks and balances.” Even though we have a majority vote, the minority has some legal recourse to protect them from bad decisions. I’m not comparing Brazil to the US, or making a critique. I am suggesting that if Devuan drafts a project constitution, that such a “checks and balances” might be a good idea - with decisions backed up by evidence rather than popularity. Considerable weight should be given to practicality, IMHO.
>Good point! I'm seeing that it will be harder to make Devuan support
>both systemd, as an option, and sysvinit-core but, it is doable.
>The worse thing I'm seeing with this systemd fiasco at Debian, is the
>lack of choice.
If I might make an observation, Debian has a way to do this. The way to make this possible is create a “virtual package” set that either sysv or systemd can satisfy. The problem that Debian faced and basically sidestepped diving into systemd was that they did not have to support two very different init systems. What I would suggest is that Devuan look ultimately at forking APT.
The reasoning is very simple. APT does not provide the mechanisms to ensure that things like systemd do not happen again.