LinuxQuestions.org - Interview with Mandrake Linux Founder Gael Duval

Gael Duval, the founder of Mandrake Linux and co-founder of MandrakeSoft, agreed to an LQ interview. Here is what he had to say. Thanks Gael.

###

LQ) Tell us a little about yourself, how you got into Linux and why you started Mandrake Linux?

GD) Actually I firstly discovered UNIX at University where I learned computer sciences. It was mostly on Sun with Solaris or SunOS, and I really was impressed by UNIX. In 1995 I had a 386-based PC at home with MS-DOS/Windows 3.1 runnning on it and of course it was... extremely frustrating. In particular when you are a student with absolutely no money, it was impossible to purchase all the development software for programming in C/C++/Common Lisp and others, or you had to copy it illegally. And of course it was without the documentation. So I spent more and more time at Uni working with UNIX. It was the early WWW times, and I remember I searched for "free Unix" on the Net. If I remember well, I used Yahoo! which started less than one year before, and the browser was... Mosaic :-)

The search results showed several Linux pages. That is how I discovered Linux. A few days later, I was at home with my Slackware on 50-diskettes, still not believing that I could run a Unix-like with X11, Emacs, GCC, Lex, Yacc, Clisp and... all the documentation on my 386. A few hours later, the miracle was here: Linux was running on the PC, with OpenLook on the screen. The next great experience was when I performed the first Internet connexion by modem, through a University access.

Two years later, it was clear for me that Linux had the potential to be an excellent alternative to Windows, or maybe even a full replacement, and at the time I thought that it would be good to provide a Linux distribution that would be as easy to use as Windows. So I started to "play" with Slackware, and later with a Red Hat. It was also the time of the first versions of KDE. After a few months of work, I released the first Mandrake, in July '98, and was the first distro to ship with KDE 1.0 as default graphical environment.

LQ) Before releasing the first Mandrake version (which was based on Red Hat) you were working on a Slackware-based OS. Any regrets on that distro switch? Do you think things would be different had you not made that change?

GD) No regret at all, for a simple reason: it was not serious anymore to release a Linux distribution without a good package management like RPM. I seriously considered to switch to Debian as a base because at the time, Red Hat's reaction was very unclear (as far as I know, forking from a commercial Linux distribution never happened before Mandrake). But back in 1998, Debian's installation procedure was really not friendly at all. As a result, a key success of Mandrake was also that all packages made for Red Hat were compatible with Mandrake, including commercial packages. So the choice of RPM was the good one.

LQ) During a mid-year status update, Francois Bancilhon noted that "Our immediate goal is to exit from this status before the en of the current year" (speaking about the Chapter 11 filing). Does it look as if you will meet this deadline? How does MandrakeSoft's financial future look?

GD) Yes, our goal is now to exit from the Chapter 11 filing soon, but there is no emergency - actually it just limits the level of business we do. We will provide an exit plan on early January and it should make us leave two to three months later. It needs a court approval.

Anyway, we've just released first financial results and they are very positive. There will be a benefit for the current quarter.

LQ) What major changes and updates can we expect to see in the next Mandrake release?

GD) In addition to many improvements, there will be more and more focus on applications that are needed in daily business in small and medium corporations (office, groupware...).

LQ) What are your thoughts on the recent End of Life announcement by Red Hat? Do you feel that it will have an impact on MandrakeSoft?

GD) It already has an impact. Many new users come to Mandrake for this reason. Our users also requested us to clarify our product lifetime and Open Source policies, and hopefully we'll release a statement this week.

LQ) What about your thoughts on the recent acquisition of Suse by Novell?

GD) Good news for everybody.

LQ) What would you consider Mandrake's largest innovation or contribution to Linux?

MD) Proof that Linux is not only for geeks (focus on ease of use), first graphical installer, first remote update utility (including graphical front-end), security levels, transparent access to devices, first Linux releases as an ISO image...

LQ) Are you happy with the financial response to the Mandrake Club? Is it more important to the success of MandrakeSoft then retail boxed sales?

GD) The Mandrake Club is getting well (nearly 20 000 users) and is an excellent business model since more and more people have enough bandwidth capabitlities to download the product.

Anyway we think that the Mandrake Club is still too much for geeks and we're going to improve that.

LQ) Mandrake has announced multiple OEM pre-install partners. Are you happy with the success of these partnerships and is it a trend that you see continuing?

GD) The trend will certainly continue and we're happy with that.

LQ) What do you see as the biggest barrier to the acceptance of Linux on the desktop?

GD) I think there is one and only one reason: the relative lack of end-users applications available, of course in comparison to what is available in the Windows world. As soon as software publishers will consider Linux ready for the desktop and will release their applications for Linux, there will be a Linux big bang.

LQ) Where do you see Linux in 2-3 years? Where do you see MandrakeSoft in 2-3 years?

GD) The state of Linux in 2-3 years depends on the former issue. Linux is now penetrating the enterprise world very quickly as a replacement to other proprietary operating systems. But for "Joe user", there won't be any Linux rush until software publishers release their applications for Linux.

As for MandrakeSoft, the future looks very good because we recently got our first good successes in local administrations (details will be announced later), so at least we will have business in this field and in the corporate field.

At the same time, MandrakeSoft silently continues to expand its presence in the world as one of the first, or the first Linux distribution in term of user-base. That will pay.

LQ) If you couldn't use Mandrake what Linux distribution would you use?

GD) This is the most difficult question I ever had to answer in an interview! :-) That's a frightening question actually because I can't see any alternative that could fit my requirements: friendly, full-featured, powerful, stable, fully open-sourced...

LQ) How would you rate the success of MandrakeSoft products such as MandrakeClustering and Corporate Server? Do you have any new product announcements on the horizon that you can share with us?

GD) MandrakeClustering is very new and it's a special business: it needs good and specialized VARs (we just signed one for the USA), and as many specialized applications as possible. The business has started and there are sales, but we still need to get more partnerships with software publishers in this area. Anyway MandrakeClustering is recognized to be one of the most easy to install and easy to use Clustering solution currently available!

As for the Corporate Server, it's very successful: it's an excellent server product that we sell as part of global deals to admnistrations and corporates. But sales are also good through our online store.

LQ) What are your thoughts on UserLinux? Do you agree with the sentiment of the manifesto?

GD) It's a bit early to comment it because nobody knows where it goes. But my own feeling (not MandrakeSoft's official position), is that 1) it's a Debian thing, isn't it? 2) it's better to follow standardization processes such as Linux Standard Base because all major Linux distributions agree with it, and it's important to avoid the "Unix balkanisation syndrom" if we want to see Linux succeed in the long term.