Re: The end of a dream

[Jean Francois Martinez <jfm2@club-internet.fr>]

That is only needed when you want to change world entirely (and in less
than six days).  A distro, not even the big ones like RedHat and
Mandrake
cannot create application software out of thin air or have the
application
developers implementing a better ergonomy.

What it would have been possible to do even with limited resources:

-Bring real knowledge of user conditions.  For instance Indy was
superior to
Mandrake for interacting with Windows machines.  How as that possible
since
Mandrake and Suse had one hundred times more resources?  Because when
the
Mandrake guy tested an SMB browser it pointed it to a powered on Windows
box
(or a Linux one with Samba) noticed it worked and gladly included in the
distro.
But I worked in a cmpany where Windows was domimant so I know the
Mandrake browser
probes a range of addresses one by one and thus is unusable when it
needs to
probe a couple thousand adresses (most of them unused so it needs to
wait for
timeout).  I also saw my Windows coworkers getting PopUp messages when
their
printing job was finished while I had to walk 20 meters to look for
mine.  And
that is why Indy included a PopUp messager and Samba was preconfigured
to invoke
it when it received a message.  Again this was not in Mandrake because
they are
not real users.

-Avoid the "far too smart developer" syndrom (or more exactly "raised in
diapers").
Your average distro developer has had a teacher to soften the learning
curve, a 
sysadmin to care for the box and the opportuniyty to play with it
without
responsibilities.  And that is why he makes a distro who relies on
sysadmin making
no mistakes because he doesn't know what is to have to adminuster the
box and fix
your (frequent) mistakes when you have zero
knowledge. If he had learned the hard way (like many Linux users) he
would try to
make the distro error tolerant (eg with a smart handling of the fonts
you can make X
not to creash if user has stopped the font server) and easier to recover
in case of
problem (for instance by telling the user at boot prompt how to recover
from unbootable
conditions)

-Care for user needs.   If you want to attract people to Linux the first
thing is make it
USEFUL for them.   Consider that having a personal finance program is as
important as
having Apache.   Do the utmost for including drivers of popular
peripherals and when you
cannot distribute them for legal reasons make a a little app who
retrieves it from
manufacturer's site and installs it properly.  Have an installer who
lets the user with a ready
box instead of with one who does not print or connect to the Net.  A
commercial distro will ever
care more for the single guy ready to sign a 50,000$ support contract
than for the hundred guys
who put together will bring only one or two thousand between them all. 
In Indy we were free to do
the opposite.

So you see at distro level you cannot save the world but you can make a
lot to impriove things and
it does not need hundreds of developers

-
Douglas Carnall wrote:
> 
> Dear Jean,
> 
> I've lurked on your list for a while because I agreed with your ideals: most
> Linux distributions is that are still too hard for the ordinary user to use.
> 
> The problem is that changing that, and making installation and applications
> "intuitive," requires more effort than any few individuals can give. Of
> necessity, the distribution we all dream of would have had to be tested on
> thousands of naive users, and their feedback integrated back into the
> development process by hundreds of (expensive, if it's their day job) UI
> specialists and programmers.
> 
> I'd love to see such an organisation, releasing its output under GPL, but I
> don't know how it's going to happen. Well, if it ever comes to pass, I'm
> sure you're one of the first guys to be hired.
> 
> Courage, mon brave.
> 
> D.
> 
> > I stop Independence.  I noticed that nobody
> > else had tested the software I had put for download and that
> > after release there would be nobody for trying to speak about
> > it and it ideals.  I cannot do this while trying to improve
> > Indy.  Thus I stop.
> >
> > I still believe something like Indy is needed because for one
> > side we have free distros who don't care for the "unwashed
> > masses" and for another side we hacve commercail distros who
> > in those times of CD burners and ADSL lose lots of money selling
> > boxes and try to recoup with support.  But support is bought for
> > mission critical applications (read servers) in companies, not
> > by private individuals.  And that means commercial distros will
> > not make a push for the common user in the foreseeable future.
> >
> > That is why I dreamed in a distro made by people willing to take
> > charge and do something for other users.  But it was a dream.
> 
> --
> Douglas Carnall
> 
> tel:+44 (0)20 7241 1255
> fax:08700 557879
> mob:07900 212881
> http://www.carnall.org/
> dougie@carnall.org

-- 
Jean Francois Martinez

Project Independence http://independence.seul.org
Because Linux should be for everyone