Re: project status

On Sat, 2005-01-15 at 19:27 -0500, Michael Miller wrote:
> ok guys... you are misinterpreting what i am trying to say. I know you
> guys might not care about the average Joe, but look at it this way: it

Actually, we do care about the average Joe that chooses to use our
distribution.  You have completely interpreted what I said incorrectly.
We don't go out of our way to try to hand hold to steal market share
from others, as so many other distributions and operating systems do.
Instead, we try to support our users as best we can, while moving Gentoo
into a better technical position.  As things get better, they tend to
get easier simply because the manual work lessens.

> can turn into profit by selling more cds from the shop-or by selling
> support. While i know the redhat updater crashes a lot, that does not

Gentoo gets no money from the Gentoo Store, which is a separate entity
run by Gentoo Technologies Inc.   GTI has agreed to donate a certain
portion of their proceeds to the Gentoo Foundation, but we are not one
in the same.

As for support, the current way Gentoo works leads to being
unsupportable, especially when contracts are on the line and timetables
must be drawn.  It is one thing to file a bug on bugs.gentoo.org, and it
is another thing entirely to be paying money to have someone support
you.  With one, there is little expectation other than the responsible
parties trying their best, with the latter, there it the expectation
that as a paying customer, your needs will be met in a timely manner, or
you will take your business elsewhere.  When Gentoo does decide to
actually provide commercial support, we'll only have one shot at it, so
we have to do it right.

> mean that the gentoo one has to :). In all seriousness you guys are
> amazing programmers. I just think that it would be benificial to
> everyone if gentoo reached the mainstream. If gentoo was user friendly
> enough(user friendly != no ability to tweak settings) it would take
> away customers from microsoft, and possibly become the domininant

That was my whole point.  Gentoo does not care about market share.  We
do not have stock holders or investors to answer to.  This is a
community.  We get to enjoy what we do, not slave over code we don't
want to write just to appease some investors.  We are *free* of such
burdens, and we *like it* this way.

As things move along, they will get easier and we will gain market
share, but that is not our goal.  Our goal is to make the best damn
distribution out there.  I think we already do, but there is still room
for improvement.

> OS/Distro. I know that gentoo was made to be tweakable. What i suggest
> is creating tools to change CFLAGS, explaining every option. This

That is a near impossible task.  This seems to be something that most
people don't understand.  The more options and explanation you give to
someone, the harder they perceive it to be.  Why do you think Windows is
"so easy" to people?  They have little to no choice on how things are
done.  They simply learn the way it works, and are done with it.  The
same cna be said about Mac OS X, the pride of the "desktop Unix" world.
Something like CFLAGS would be pretty much impossible to give the user
all of the options, explain it so they can understand them, and still
make it friendly.  Heck, there are tons of possible CFLAGS options that
I don't understand at all, and have had to read pages upon pages just to
understand.  Could you imagine if we had a 400 page "CFLAGS manual" in
an effort to "simplify" things for our users?

> allows for both user friendlyness and ability to tweak. You can label
> everything so the average joe can figure it out, while still
> maintaining the ability to change settings.

Actually, I think the way we are currently planning seems to be a bit
easier.  We provide a sane set of defaults, then basically a blank slate
for the user to make customizations.  The defaults we provide tend to be
good enough for everyone, which is why they're the defaults.  We could
spend months of developer time trying to work out all the possible
CFLAGS combinations, including which ones are safe or not for each
architecture, or we could spend that time adding more packages and
improving the quality of the ones we have.  It is a trade-off, but one I
think we all are willing to make.  We simply don't have the manpower to
make tools capable of informing the user of all possible choices, while
still shielding the user from themselves, which is a requirement for
"user friendly" capabilities.

-- 
Chris Gianelloni
Release Engineering - Operations/QA Manager
Games - Developer
Gentoo Linux

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)

iD8DBQBB6pe8kT4lNIS36YERAvRKAJ0VY4/XuEM7CU8qwBLN7BDCN4MajACgpXsp
EHKE1TX989CmwMgy/G3vImo=
=QxPV
-----END PGP SIGNATURE-----