On Tue, Apr 29, 2003 at 04:25:07PM -0500, David Krider wrote:
> Nathan E Norman wrote:
>
> >>Well, then "they" ought to get used to the idea of being used by an
> >>increasingly-smaller cross-section of the Linux-using community.
> >
> > This threat is heard repeatedly year after year; yet debian "user
> > share" keeps growing. I wonder why?
>
> Really... Do you have a reference somewhere? I suppose you're giving as
> much support to your argument as I gave to mine, but whenever I see
> polls on the net, Red Hat, SuSE, and Mandrake are typically in the lead.
> (Except for the desktoplinux.com poll, which seems to have been horribly
> "stuffed" compared to all the others.) Debian usually places a
> respectable 4th.
You just provided the support for my argument and refuted yours;
thanks. Think about it: if debian's "market share" were "an
increasingly-smaller cross-section of the Linux-using community",
debian should place lower and lower in successive polls. Yet, as you
say, debian is consistently fourth.
Reading what I wrote, I can see that my comment could be misread. I
should not have used the word "share". I was attempting to point out
that the number of _users_ of debian is growing, consistently. The
userbase is not shrinking. It is not growing more slowly than that of
other distros, as far as I can tell, because as new Mandrake and RedHat
users srping up, many existing users decide to take the plunge to debian
because they have heard it is better. I'd guess about 60% of those stay
with debian (but now I am simply guessing).
Are you saying debian should be in first? Why do you think that? Since
these polls are in no way scientific, I think debian in the top 5 is an
accomplishment; it wasn't long ago no one had even heard of debian.
Remember, to most non-linux users linux == redhat.
> Six months or a year ago, I could see that.
How about two years ago? Five years ago? Why chooose the arbitrary dates
you've chosen?
> What's the future look like for Debian? Assuming for a moment that
> sarge's release is 6 months off, and *if* KDE 3 and Gnome 2 make it
> into that release, that puts Debian at a year and a half behind the
> other 3. Do you think that will foster adoption of the distro? I'd
> qualify my comments about the difference between a desktop and a
> server distro, but if you want to master a distro, you really need to
> use it in both places.
Could you explain the last sentence? I don't understand what you are
trying to say. If my job is to have a solid, stable cluster of servers
and I choose debian (I'm basing this example on real life BTW) I believe
I've made a good choice. I also feel that I know quite a bit about
debian, if that's what you mean by "master". I seem quite capable of
installing debian on my desktop. I've noticed other people seem to have
trouble installing debian on their desktop, often because they come from
a world of "Internet Connection Wizards" and the like. As far as I can
tell, there are other distributions that cater to those people, and I
think that's fine.
In other words, I think the future looks great. See my other post in
this thread for more thoughts.
[ snip links ]
> Yeah, yeah. I know. "If you don't like it, don't use it." I get it.
> I'm still trying to figure out if it's worth it, *to me*, to live so
> far behind the curve for the sake of getting out from under a
> corporate interest. Red Hat just burned us with their licensing
> changes, and I don't want to get hooked on SuSE (it's _great_ from
> what I've seen so far) only to have them pull the same stunt in 6
> months. You can bet that if the move doesn't hurt Red Hat, the other
> commercial vendors will follow suit...
I don't understand what curve you are living behind. Can you seemlessly
upgrade your redhat or suse box, as you can do with debian? If you are
running servers, debian stable is sufficient for most tasks. There are
plenty of people who are backporting packages from sarge/sid to woody,
so if you _must have_ the latest and greatest software (I cringe to
think how many people deploy such software in a production environment
_without testing it_), that software is available. If it isn't, you can
grab the source and compile it yourself; no other distribution I've ever
used respects admin changes as well as debian. I can install stuff in
/usr/local and use dpkg-divert to prevent dpkg from surprising me, or I
can create my own packages (this is useful when you need to deploy your
home-grown software to several machines).
Admittedly debian on the desktop is a tougher proposition; often newer
XFree86 is needed to support new hardware. Again though, backported
packages exist.
I have a bias towards the server side of the equation as I believe
that's where debian has, and will continue to make inroads in
corporate-ville. Perhaps projects like debian-desktop will spur adoption
on the desktop which is what you seem to be worried about. I don't think
the home market is a big deal really; people will use linux at home once
they get used to it at work (and likewise will not use it at home as
long as they have to use Windows at work). If what people used at homje
influenced what was used at work, a lot more workplaces would be using
Macs, wouldn't they?
All in all, if you decide it's not worth it to you, we (at least I)
won't be angry. I'm all for free software; I don't really care too much
what distribution you are running unless you hire me to support
it.
I hope this explains my position. Good luck with debian, and with linux.
Best regards,
--
Nathan Norman - Incanus Networking mailto:nnorman@incanus.net
A good plan today is better than a perfect plan tomorrow.
-- Gen. George S. Patton, Jr.