Jan 31, 2005
Going into 2005, Mandrakesoft is a company that has gone against everything that was predicted. It has survived the dot-com bust, become profitable and continues to frown on the idea of moving away from Free and Open Source Software development in favor of the proprietary approach advocated by many of the newer GNU/Linux distribution developers. To catch up on the changes since their last full interview, Mandrakesoft co-founder GaÃ«l Duval recently agreed to talk with OFB's editor-in-chief, Timothy R. Butler, about where the company is heading and the state of the industry at the present time.
Open for Business:
It has been almost a year since Mandrakesoft emerged from bankruptcy protection. A lot of nay-sayers had concluded that the company wouldn't survive that period, but instead it seems not only to have survived but thirved. What is it about Mandrake that has allowed it to overcome a situation that has lead to the demise of so many other IT companies over the last few years?
There is a lot of business to do with Linux and everything is still to invent in OSS
business. I think it's been the main reason why we survived.
Our problem was mainly a cash issue, and the fact that the company was oversized. This is related with all excesses of the year 2000 and the the bubble-crash which followed.
Mandrakesoft's stock appears to have done quite well in the last year, as well. Does Mandrakesoft have any more capital raising plans for 2005?
We have just closed a capital increase of around 4 millions USD,
so it should be enough for now. Especially because - as we are profitable - we don't burn the cash anymore. But this increase of capital was needed to upgrade our listing to a regulated stock market, which is likely to happen in the next 6 months.
There have been a number of distributions that have arrived on the market over the past few years — Linspire's and Xandros' products, for instance — that have focused on providing a more proprietary angle to GNU
/Linux. By that, I do not mean just on the subject of licensing, but also by creating more in-house replacements for Free Software projects and doing larger scale modifications to the FOSS
projects that are used. This, arguably, has the advantage of tighter integration while providing the disadvantage of higher training costs between distributions. What are your thoughts on this type of design and do you see Mandrakesoft ever doing the same thing?
I wish them courage and luck, because unless being a very large corporation, this is going to be very complicated to maintain all these developments in-house and try to impose their proprietary views and standard candidates. We also had the choice to take this way, a few years ago, but we realized that Open Source Software will always win in the end. This is the trend and the open source community capitalizes the biggest amount of energy in the software world.
On a similar note, it seems that there are constantly people from both the GNOME
communities trying to argue that Mandrakesoft prefers the other community's desktop. I've heard a number of people recently suggesting that Mandrakelinux has become more GNOME
-centric, for instance; do you see any truth to this?
I'm pleased to hear that because it means that we have succeeded in being as neutral as possible. For a few years, Mandrakesoft's policy has been to offer the best of both graphical environments.
|Duval: Open Source always wins in the end.|
Speaking of GNOME
-centricity, Red Hat's Fedora Core has essentially remained quite focused on GNOME.
While this cuts down on choice, it has also allowed Red Hat to integrate its system configuration tools directly into, and use the look of, the GNOME
control applets area. Do you see somewhere in the future where Mandrakelinux's control center might be integrated into either KDE
's or GNOME
's respective system preferences areas?
Our Control Center would be easier to integrate into GNOME
than into KDE
because it uses the Gtk+ widgets. It's currently a stand-alone application (in fact, it's a modular application where you can add new configuration modules easily), and is well integrated in the menues. Users are generally happy with that. It was also ranked 1st in a Linux Configuration Tools roundup in Linux Format UK in January. A full integration, as part of GNOME
would mean deep hacks, and it's always dangerous to maintain.
But of course, this is the kind of questions that our marketing department, our development team and the community are constantly thinking about.
Mandrakesoft, along with several other distribution companies, announced the Linux Core Consortium a few months ago. How is this project proceeding?
A common “distribution-ball” is going to be released online, a bit like Cooker. It will consist of a core LSB
-compliant Linux distribution, and should be available both as RPM
A few announces should follow soon.
Is there any word on whether other distributors invited to join, particularly Red Hat and Novell, have made a decision one way or the other?
Red Hat and Novell have expressed official support for the project, as well as HP,
Computer Associates, Sun and others. I think they want to see first what it looks like to decide wether they should join or not, it's natural. Furthermore, Red Hat and Novell aren't anymore the kind of Linux company which can decide in weeks for such a new project. They are too big.
The site says that the core will be released this quarter. When it is, what can we expect to see; will there be a functional reference distribution that can be installed and operated on its own, or will it be a set of packages that require additional software to turn into a usable distribution?
It should be a usable (but basic) Linux distribution. This is still a bit too soon for such details. We are going to communicate about it each time there is something new to say…
Move is in its second iteration presently. Can we expect anything new in this area over the next year? Have you considered a GNOME
Currently there hasn't been a strong demand of a GNOME
-oriented Move. We don't plan any new product in this area in the next semester. Move is selling well, as well as the Globetrotter.
/Linux distribution is picking up, although it still is not typical to run into a SOHO
user in a non-technical business who is running a GNU
/Linux desktop. What do you think it is going to take to change this situation? Do you see it being resolved in the next year or two?
Linux is ready for the business desktop. I mean at least for well characterized tasks such as Internet/Office. But I'm afraid that it's still not ready for everyday's user at home because of the lack of all end-user applications and games. Open Source replacements have emerged for many of them, but that's not enough. I think we need either a Windows emulator that works all the time, transparently and is very fast, or software vendors have to move themselves and provide Linux compatibility for their applications.
A few years back, Mandrakesoft added ZeroConf (better known under Apple's brand Rendezvous) to Mandrakelinux. Since then there has not been any apparent move, that I am aware of, among other vendors to support it, save for functionality built directly into KDE
which is obviously less desirable than it being supported across the whole system. What is Mandrakesoft's position on continuing to support the ZeroConf spec in the future?
We're going to keep it unless nobody needs it. In general when we add a new feature to Mandrakelinux it's because someones needed it. That's one of the reasons why Mandrakelinux is popular!
Speaking of Apple, I am sure everyone is now familiar with the USD
$499 Mac Mini. It would appear that this is starting to encroach on the fledging GNU
desktop market by placing a BSD
-based alternative in a nice package for less than $500. Some have also posited that this unit would make a nice file server, or perhaps even more powerful server using clustering. Is this a threat to GNU
/Linux on the desktop?
Not really since the Mac Mini is very likely to run Linux. Now the question is: what's the cost of the Mac Mini without MacOS?
Seriously, this is Apple traditional (and excellent) marketing and communication. Now it's easy to find a PC,
faster than the Mac Mini, with a 17” flat screen and Linux, for $500 or less.
My feeling is that the iPod was a better concept than the Mac Mini, but I may be wrong.
Searching looks like it could become the next big thing. Microsoft and Apple are, of course, working on search technologies in a big way, and Beagle looks quite promising in GNOME.
What are your thoughts on this current trend? Is there anything Mandrakesoft is looking into promoting or developing concerning this?
That's the kind of features we want to add to Mandrakelinux.
With the new year already in full swing, what do you anticipate as the most exciting upcoming development at Mandrakesoft?
There are many projects here, and a major announcement for Mandrakesoft is to follow in about one month from here.
Thank-you, GaÃ«l, for your time.