I think that Ian Murdock is right on with regard to speculation
about the future of Mandriva.
I have been a GNU application user since the eighties (using
stuff like GNU Emacs and GNU tar, and later, BASH, prior to getting
into using GNU/Linux software in late 1995 (Slackware was the
distribution I used first). It wasn’t until 2001 that I started to
regularly use desktop Linux software in place of other software at
home.
At first, Caldera OpenLinux eDesktop 2.4 (on my laptop) and
Mandrake 8.0 (on my desktop) were the two Linux distributions I
used the most. Actually, I started using eDesktop 2.4 in 1999 on
the laptop, but because of some graduate classes I was taking that
were (unfortunately) using and requiring the use of Microsoft
Office, I had to do some dual booting back then.
During my graduate studies, I wrote a lot about Linux software.
I felt that a really solid desktop system could be built, even from
the software available at that time (which is tremendously improved
now). However, what I felt was lacking then, and is still somewhat
lacking even today, is a solid business plan from which to grow a
thriving and prosperous business.
I felt that the major hardware vendors who have lost desktop
market and operating system sales over the years would be the most
likely, willing, and able organizations to put a sizable investment
into Linux software. While I felt that it would take several years
for that to happen, I’ve been really disappointed at the amount
(lack) of investment that has been made. Those who have invested
have done so almost exclusively at the server level, where there
already is a lot of great software.
At this stage, there are a few software companies who are at
least solvent enough to stand on their own two feet. Since the
hardware vendors have not really cooperated all that much, a few
good software companies with a good business plan could possibly
turn the trick in a non traditional way, as Ian Murdock is now
suggesting.
I provide the history of my own Linux usage because during the
time between 2001 and 2002, I saw commercial Debian ventures begin
to see a measure of success. Prior to that, Corel had initiated
some great ideas, but were unable to pull them off, so they sold
off to Xandros. Progeny, Ian Murdock’s company got off to a good
start, but quickly found that selling distributions was not the way
to make money. Distributions, at best, were “enabling technology,”
providing, in effect, a reasonably easy doorway into a vastly
different style of computing than what most people were familiar
with at that time.
2002 was an interesting year. Xandros finally put out their
first product, and also helped Lindows.com with their first
release, both Debian based. Knoppix really started to provide some
interesting ideas in hardware detection and usable Live CDs that
were quickly adopted by a rash of small development efforts. By
2003, several other distributions were also producing their own
Live CDs. One of them in particular, MEPIS, really came up with a
terrific idea of using the Live CD as a way to boot up the system,
then place a desktop icon on the Live CD desktop, from which the
system could be quickly and easily installed to the hard disk.
Now we have Ubuntu and many other efforts doing similar things,
and well over half of the cool stuff going on seems to be evolving
out of the Debian and Knoppix projects and the resulting efforts
that have spun off them.
Mandrake did a pretty good job of turning urpmi into a pretty
decent installation tool that nearly rivals apt-get, dpkg,
aptitude, and synaptic, in terms of handling application and
library packaging updates.
By acquiring Conectiva, Mandrake–now Mandriva–has one of the
assets in house that can help accomplish a move from RPM packaging
to DEB packaging: the great synaptic tool. With synaptic, I could
foresee Mandriva eventually rewriting or eliminating the use of
urpmi. Joseph Cheek, Mandriva’s latest consultant, wrote a pretty
handy tool himself in Lycoris to graphically handle RPM packages,
but beneath the covers, I think it really forced packages in place
using the equivalent of rpm -Uvh *.rpm whenever a group of
applications were installed.
I think that Conectiva acquisition, as you noted, helped
Mandriva acquire one community; the Lycoris acquisition, besides
(at least temporarily) picking up a great engineer in Joe Cheek,
helped to build another solid pocket of a desktop community.
If Warren Woodford (MEPIS); Tal Danzig (Libra Systems LTD.’s
Libranet); Michael Robertson or Kevin Carmody (Linspire); Ubuntu;
or even Xandros were to be obtainable, each of them provide a
fairly decent user community, Debian software, and the
underpinnings from which a Debian infrastructure could be built. A
few of them already have sizable (and cooperative) user
communities.
I’d still like to see companies like IBM, Hewlett Packard, Dell
(and am I asking for too much, Sun Microsystems?), to put some real
energy behind not only further entrenching Linux as a leading
server operating system, but to put some muscle behind making it
work on the desk. Maybe that’s not such a good idea,
though, they’ve rarely cooperated with each other in the past.
Perhaps a French company, leading multinational development
efforts and communities, might instead put together a solid
community and finally bring us a winning desktop that can be
deployed, not only by individuals, but by businesses and
educational institutions as well. Perhaps making it an honest
community effort, backed by some loose corporate project
management, might be one way to pull it off. I like the idea of
centering that user community around a Debian approach because the
Debian packaging fosters community, but even more importantly, the
Debian Social Contract is all about community and freedom.
Something to think about, huh?