Re: RH and GNOME
On Wed, Jul 22, 1998 at 10:29:31AM -0500, Zed Pobre wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
>
> On Wed, 22 Jul 1998, Enrique Zanardi wrote:
>
> >If we have to rewrite every app out there to make them do the right thing
> >I doubt we were be able to buil quality distribution, or any
>
> We won't have to rewrite every app out there, because it's a pretty
> reasonable assumption that:
> 1) All our old apps aren't going to disappear
But old versions will became obsolete as time passes (for Linux almost
everything older than, say, one year is obsolete). If new versions use
stupid extensions we will have to modify them. More work for us.
> 2) All the new apps being written aren't going to use every stupid
> extension in existence.
Oh, it doesn't take a whole load of stupid extensions to ruin our life as
a project. Just a few ones in a majority of our packages (remember, we
have ~1900 packages).
> And if an app looks like it's going to be too much work to make
> stable, we don't bother packaging it. Nothing says we have to package
> every scrap of code in existence. Developers who want to muck with
> packages with problems, will, and everyone else can ignore them. No
> great loss.
You base your reasoning in the assumption that only a few non-important
apps will be problematic. What if the problem is in an important library?
What if the problem is in a lot of important apps? What if every
gnomified app becomes totally incompatible with our packaging tools or
our menu-system? (Not that I think it will happen. I just try to clarify
my point of view).
> And I think the net has amply demonstrated that when things
> start to break in a serious fashion, and the "people in charge"
> refuse to change, people will take their work elsewhere and start a
> divergent tree if they have to.
How bad a solution must be to treat it as "broken in a serious
fashion"? If it's simply a worse solution than the one we have developed?
If it's only a different solution? Can we afford to waste all the efforts
we have put in developing a solution every time the "industry" decides to
choose a different one? Can we afford to stop being an important part of
that "industry"? If no one is going to use the solutions we develop, we
may as well pack up and do other interesting things (scuba diving
anyone?).
> Because they can. That's why it made a
> difference in the previous argument that Windows is not GPL. The funny
> thing about GPL is that the worst that can happen to you is that you sit
> at the status quo.
Unluckily, in this quickly evolving world we are like sharks, if we stop
moving we are sunk.
> >distribution at all. When you think about market share, you must remember
> >that includes the developers of the software we package. If more of them
> >use Debian, that will make our work easier.
>
> Sure. Of course, having them use Linux at all makes our work easier
> than if they write their code for Windows, at which point it's
> practically impossible.
Harder, but not impossible (two nice apps I use, xmame and x11amp are
ports of well known windows apps).
> I must be still tired; I'm drifting off topic. The reason that the
> extreme came up is because I see truly paranoid and/or venemous comments
> come flying across this list, and I really don't want that to become a
> serious part of my life. Windows still has a lot of software that is
> either nicer than the Linux equivalent or that has no counterpart, but
> Linux is still my primary work machine because for 80% of the work I do
> it covers my work needs, and that number keeps rising. Frankly, I don't
> see Redhat as a threat to that trend.
Neither do I. But the goals of a Debian developer are a superset of the
goals of a simple Linux user. We want more apps for Linux, of course, but
we also want to integrate them in a high quality distribution. Anything
that makes that goal harder to achieve is a possible threat we should
consider (if only for a few minutes a year).
To keep my position clear. I don't see RedHat as a threat here and now.
But I think that any commercial Linux distribution becoming the Linux
industry monopolistic leader can make our work harder, and I want to work
proactively to keep or expand our "market share" to avoid that.
We don't have to treat them as enemies, but we must be aware that they
are our competitors in the Linux market. And we must be aware that losing
that #@~|@ market share will make our work harder.
> >Of course, that's an extreme example, but look at the RPM vs deb thing.
> >It's not that extreme, and we are facing it _now_.
>
> And I *still* don't see the problem. It's not like Debian is going
> to stop using dpkg in favor of RPM, especially since RPM doesn't do what
> we want.
Sorry to ruin your dream, but if the LSB mandates RPM as the Linux
standard package format, we will have to use RPM (the format), or forget
the "Linux" part in Debian GNU/Linux. We are the ones that follow the
official standards, aren't we?
> >More "market share" means more developers using Debian (not just more
> >Debian developers, but Linux apps developers), doing things "the Debian
> >way", and that means our goal (a high quality distribution) will be
> >easier to achieve.
>
> Hey, I'll agree with that, at least up to a point. I'd be as happy
> to settle for developers not doing things that were inherently wrong,
> though, and Debian isn't the only advocate of that.
If it were that easy to prove them they are "inherently wrong"...
--
Enrique Zanardi ezanardi@ull.es
--
To UNSUBSCRIBE, email to debian-devel-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Reply to: