[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Future of Debian uncertain?



On Wed, Feb 26, 2003 at 12:29:45AM -0500, Alfredo Valles wrote:
> I think that many new people like me are so in love with debian and want 
> it to overcome its faults that we lose perspective. Maybe we should 
> first congratulate you all for the terrific work you've done with debian.

The problems you raised aren't based on a perspective. They are in part
based on market positioning. The need for general-use distro has dimnished.

Currently Debian is the distribution best suited for production servers - it
is rock-stable, it changes rarely, the security updates preserve APIs and
binary compatibility. Current form of Debian can also be used on a desktop,
but...

Desktop distribution and desktop users have slightly different needs than
servers. The users are heatseekers[1] no matter it has sense or not. They do
not want  for the latest KDE to mature, they want it NOW.

The question is if Debian project should cater to those users.

But wait, there's more. There are users who don't want lates KDE3 because
they don't know what it is and they shouldn't - they're end users who want
to do stuff on a computer no matter what it runs. This market segment is
growing fast but it has very special needs - the systems need to be polished
from marketing perspective for the enduser. No loose ends left because there
was nobody to check the final product. Look at Windows or MacosX from the
enduser aesthetics and usability point of view. The UIs concentrate on doing
stuff like working with music files, photos and textual documents, not on
'running programs' or even 'browsing the web'. The systems mentioned are
complete in this aspect. System management is simplified to the maximum.

I don't think that The Debian Project is technically able to cater to third
group - there's a lot of expensive work or such that nobody will want to do
as member project (I'm talking about marketing research and usability
testing for example). Commercial spinoffs of Debian could be able to adapt
the system to needs of those users.

For the second group - this is difficult to decide if they should get what
they want at all - and this is a question on Debian market positioning. If
this is decided, some things should change (in the simpliest approach - the
bare minimum is to provide security updates for testing). Maybe some system
akin to BSD ports that automatizes recompiling testing and unstable packages
for stable. But first, the decision.

In my opinion Debian has enough strong marketshare in the server market that
it probably always stay here to some extent. Most of the Debian people are
admins or programmers and those types best work at own needs.

Alex
[1] http://www.science.uva.nl/~mes/jargon/h/heatseeker.html



Reply to: