On 10/6/05, Yedidyah Bar-David <[EMAIL PROTECTED]> wrote:
We cannot accept any non-free software in our repository. Not even links to non-free software. So that's enough of a reason to dump any project that doesn't respect user freedoms.
Yep, we also have legacy stuff, but compared to other projects, we are much more younger, so it's easier to get rid of it, and much easier to do good planing not to get to that point.
Big projects have the tendency to forget where they came from. Not to begin a flame war, but we are a GNU distribution, not a Linux distribution. So, linux is just a package, the same for kfreebsd. Also, the majority of users don't know where the software they are using came from. When you see gcc-lib you know it's part of gcc. When you see libstdc++.. you get my point.
I update the packages I need for my work. Others update packages that they need for theirs. It's hard to manage updating non-critical packages, so that's why we have only half working gnome and xfce, no openoffice yet, etc... But the stuff we work with it's good and stable enough (for us) to be used in production environments.
But are you sure the right thing is to dump Debian (or gentoo, for that
matter) and start from scratch? Debian isn't just dpkg and apt. In fact,
We cannot accept any non-free software in our repository. Not even links to non-free software. So that's enough of a reason to dump any project that doesn't respect user freedoms.
Of course, if you consider the entire Debian archive and infrastructure
to be full of legacy (which it probably is), etc., then you have no
option. But my own experience says that real life isn't _that_ simple,
and if you manage to get enough users and momentum, you'll turn up to
also have legacy cruft etc. You'll probably also find out that you do
not agree with all of the decisions of your volunteers, and either let
your archive loose quality (by your definitions) or spend a lot of time
fixing their work.
Yep, we also have legacy stuff, but compared to other projects, we are much more younger, so it's easier to get rid of it, and much easier to do good planing not to get to that point.
> packages, while we name them after the sources. Just some examples, we have
> gcc and gcc-lib instead of gcc and libstdc++, mysql and mysql-lib instead of
While it makes some sense, why do you think it's important?
Big projects have the tendency to forget where they came from. Not to begin a flame war, but we are a GNU distribution, not a Linux distribution. So, linux is just a package, the same for kfreebsd. Also, the majority of users don't know where the software they are using came from. When you see gcc-lib you know it's part of gcc. When you see libstdc++.. you get my point.
That's admirable, but do you really think one person can manage keeping
up with updates of 2000 packages? You already say it's too hard.
I update the packages I need for my work. Others update packages that they need for theirs. It's hard to manage updating non-critical packages, so that's why we have only half working gnome and xfce, no openoffice yet, etc... But the stuff we work with it's good and stable enough (for us) to be used in production environments.
"Suppose I want to help Free Software. Given my set of abilities and
interests, and the limited time I want to devote, what is the best way?".
For me personally, it wasn't helping some specific distribution, but
mainly participating in this mailing list and a few small projects. But
as I said, if I decided I want to help a specific distro, I would choose
one of the big ones.
--
Didi
As long as you help GNU or other free software projects that we use, you are indirectly helping us, and we appreciate that. If you want to help us directly we'll appreciate it even more. But I can't convince people that we are better than other projects. We are not (yet). But you'll never know if you won't try it :)
--
Robert Wallner
