Gaël Duval, the founder of the popular MandrakeLinux (later MandrivaLinux) project, was fired from Mandriva last spring in an effort to cut costs. Almost immediately afterward, Duval began work on a new GNU/Linux distribution, Ulteo. As of this writing, Ulteo has not yet released its first beta edition, but it should be available soon. While we’re all waiting, I figured I’d ask Mr. Duval some questions about Ulteo, Mandriva, and starting GNU/Linux projects in general. Answers are below.
It’s been six months since you left Mandriva and started the Ulteo project. Has the transition been going well? Any hard feelings for Mandriva?
Gaël Duval: Well… You know, when at 25 you create a new product, then create a company to support the development of this product, and the company becomes quite successful, when finally the person who took over control of that company decides that it’s time to say “bye bye” in the most hypocritical way it could ever be, it’s very hard to accept.
My relationship with Mandrakesoft/Mandriva has been really more than a “job” thing. It was something very special for me, a big part of my life. I’m afraid I will never have harsh enough words for this guy and for what he’s doing with the company. It’s still a very sensitive question for me. I prefer not comment further.
What is Ulteo exactly? What is its target market?
GD: Ulteo’s purpose is to address several issues:
- How to free the user from traditional system administration and system upgrades.
- How to free the user from application installation.
- How to bring open source applications to Windows users without them having to install Linux.
- How to improve the security and privacy.
- How to address modern nomadism needs.
The target market is individuals for the first step, but the concept is certainly also good for corporations, so the second target market will be small and medium businesses.
What lessons did you learn from founding and developing MandrakeLinux that you have applied to the founding and development of Ulteo?
GD: I think that the key is to delegate the work to trusted people as much as possible, when possible. But never give over the control, even if you are young and not experienced. You need experienced people at some point, but never to tell you what you should do. They should just give their best effort for your plan and vision.
In fact, (so-called) experienced people should just tell you about this point, but of course, they don’t.
That’s something I say to all young people who are starting a project or a business, because I wouldn’t like them to commit the same mistake I did.
On the Ulteo Web site you say that a portion of the profits will go to “humanitarian and ecological organizations.” Which organizations in specific?
GD: That is still to be defined, I have some ideas though. Look at the world where we are living in. Isn’t there a problem? If I can start a successful business again, I want it to help with the problems that are ruining the planet and humans. I think that all managers and entrepreneurs should think about that. The economical and financial powers, by giving just a small part of their benefits to humanitarian and ecological organizations, would certainly help things to go much better on earth.
What’s the question I forgot to ask? What do you want readers to know that other journalists never seem to say?
GD: Eh eh
Maybe you could ask: “how are you going to address the real market, which is the Windows end users market?” That’s certainly where “we,” the Linux vendors, have really failed so far because we made the mistake of thinking that all Windows users would just eventually install Linux on their PC instead of Windows.
Discuss this article or get technical support on our forum.
Copyright 2006 Jem Matzan.
|
|
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 2.5 License.
|
Writing analysis
The author of SUSE Linux must have been paid by the word; he tends to try to tell the story of how to use SUSE Linux instead of getting straight to the useful parts. While this may make for an interesting classroom presentation (the author is a teacher), it does not make for compelling reading. SUSE Linux proceeds at a glacial pace; each subchapter starts with a retrospective on Linux or Unix history, or a famous quote, or something else tangentially related to the subject of the chapter, then you’re told what the goal is and how to accomplish it in a variety of ways. In essence, each lab is like a how-to article that you might find on a Linux news site.
According to the preface, this book was originally designed with SUSE Linux 9.3 Professional in mind, then remodeled to accommodate SUSE Linux 10. Before the book was completed, SUSE Linux 10.1 came out. So which version does SUSE Linux apply to? I found that it was mostly concerned with elements common to both 10 and 10.1, with very few bits specific to 10.1; the book failed to cover 10.1’s post-release problems with ZENworks.
Putting the book to the test
SUSE Linux is best read start to finish. Each chapter is filled with “lab” subchapters where you solve a problem or accomplish a goal. In the beginning the book covers many general, elementary GNU/Linux concepts and ends with super-advanced system administration techniques. It’s entirely possible to find yourself skipping over much of the obvious material at the beginning of the book while finding the system administration sections near the end to be over your head.
Some of the labs in SUSE Linux are — to be blunt — crazy. For instance the lab on adding software to SUSE starts out by saying that the easiest way to do this is through YaST, but then proceeds to tell you how to do it from the command line by browsing the installation CDs and unzipping a directory listing. There is no conceivable situation in which someone would want to do this instead of using YaST, and it doesn’t teach readers how to do anything that might be useful later.
If you buy this book, you’ll learn more than you probably ever wanted to know about SUSE Linux 10 and 10.1, but don’t expect to become immediately productive with it. This is not a “quick start” guide, nor does it cover installation troubleshooting — it is a course on using GNU/Linux in general and SUSE Linux 10/10.1 in specific. If you’re going to be reading it in your spare time, it could take months to complete all of the labs.
Conclusions
SUSE Linux is most appropriate for people who have no prior experience with GNU/Linux or other Unix-like operating systems, or who need to learn everything there is to know about SUSE Linux 10.1 for their career. Think of SUSE Linux as a classroom with paperback covers. By the time you’re finished with this book, you should be able to do nearly any system administration task in SUSE Linux 10 or 10.1. Experienced users will probably find SUSE Linux to be too tedious and pedantic for their tastes, but readers of all skill levels will be able to learn something.
| Title | SUSE Linux |
| Publisher | O’Reilly |
| Author | Chris Brown |
| ISBN | 059610183X |
| Pages | Paperback, 430 pages |
| Rating | 7 out of 10 |
| Tag line | A complete guide to Novell’s community distribution. |
| Price (retail) | U.S. $26. Buy it from Amazon.com |
Discuss this article or get technical support on our forum.
Copyright 2006 Jem Matzan.
|
|
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 2.5 License.
|
How are things going with the integration of Conectiva and Lycoris?
Francois Bancilhon: Integration of Lycoris was mainly done with the release of Mandriva 2006, where Lycoris technology was incorporated into the distro, we were satisfied with the result. Integration of Conectiva is now completely done and we are one single company with one leg in Europe and one leg in South America, with about the same amount of people in each location.
It’s been about six months since Mandrake founder Gael Duval was fired from Mandriva. Any regrets at this point?
FB: The same regrets I had at the time: that our financial situation did not allow us to keep all the valuable employees we had to let go to reduce expenses, and this applies to Gael as well as to the others.
Do you have a chief competitor? Of Microsoft, Sun Microsystems, and commercial GNU/Linux vendors like Red Hat and Novell, whom are you most focused on displacing?
FB: We have two different businesses: consumer and corporate. On the consumer side, we mainly deal with customer or users switching from Windows or who already had a Linux distro. On the corporate side we mainly sell to customers coming from the Windows world or the Unix world.
Many feel that Vista’s delays, high price tag, expensive hardware requirements, and Microsoft’s bad reputation for release quality and overall security offer commercial GNU/Linux vendors the opportunity to capitalize on Vista’s release. Unless there are more delays from Microsoft, this will be your last Mandriva release before Windows Vista hits the market. Have you done anything special to Mandriva Linux (in its various forms) to better compete with Vista, or has Vista in any way changed your strategy?
FB: It’s true that the Vista situation creates a real opportunity on the desktop for Linux. I am also very happy to see the Linux technology taking the lead in terms of desktop with the release of 3D technology. I’m impressed by the wow’s that the demos of Mandriva 2007 generates among the people watching them. Something is definitely happening there.
What’s the question I forgot to ask? What do journalists overlook when writing about Mandriva?
FB: Two things I’d like to insist on. First is the importance of emerging markets for Linux: we see large opportunities in country as diverse as Venezuela, Algeria, Tunisia, Ghana, Morocco, Azerbaijan and Georgia. In Brasil, right now we are selling in the order of 30,000 Mandriva machine per month, which is a sizable portion of the PC market there. This takes me to my second point, which is the emergence of the OEM channel to bring Linux to the PC market.
Discuss this article or get technical support on our forum.
Copyright 2006 Jem Matzan.
|
|
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 2.5 License.
|
What does “carbon-free” mean?
The term “carbon-free” isn’t what it sounds like — it’s not like sugar-free or caffeine-free. It means that in using this CPU, the carbon dioxide produced by generating the electricity necessary to run it is offset or nullified. A byproduct of electricity production (and the burning of fossil fuels) is carbon dioxide, a “greenhouse” gas that insulates the planet. A certain level of carbon dioxide in the atmosphere is necessary to keep the Earth suitably warm, but an overabundance of greenhouse gases could lead to excessive global warming. Fortunately, many plants have the innate ability to naturally eliminate small amounts of carbon dioxide, thereby reducing some of the greenhouse gas that human beings are making through energy production.
So how is the VIA C7-D processor carbon-free? It consumes a maximum of 20 watts at 1.8Ghz, which is considerably lower than Intel and AMD desktop CPUs that, system-wide, can easily exceed 300 watts. For the small amount of carbon dioxide generated from the power consumed by every C7-D processor sold, VIA claims it will work with Carbon Footprint Ltd. to invest in “regional projects in energy conservation, reforestation, and alternative energy,” thereby making each C7-D CPU carbon-neutral.
VIA’s carbon offset measures assume that a C7-D-based computer will be in operation all day, every day for three years, with 27% of that time being spent in standby mode, 3% in sleep mode, 67% in idle mode, and 3% in performance mode. Any use beyond these estimates will increase the carbon footprint of the machine, thereby eliminating its carbon-free status.
VIA Technologies representative Keith Kowal told me in a phone interview that VIA is currently partnered with OEMs in the United Kingdom and China to produce entirely carbon-free “green” PCs based on the C7-D.
C7-D technical details
Before you get too excited about the C7-D, you should know that you can’t just log onto Newegg or run down to your local CompUSA, buy one, and put it into your existing desktop machine. The C7-D, like all other VIA CPUs, is designed to be part of an integrated product like VIA’s own EPIA series of Mini-ITX motherboards. As of this writing there are no EPIA boards that include the C7-D, but VIA seems to be planning some C7-D-based machines for its PC-1 initiative.
The C7-D is the most technologically advanced VIA processor to date. In the past, VIA CPUs were behind the times in terms of raw speed, memory bandwidth, and integrated processor extensions like Intel’s MMX and SSE (Streaming SIMD Extensions). VIA processors have also traditionally been too slow for a lot of desktop work, though in recent months that reputation has been changing. Here’s a brief overview of the technologies found in the C7-D:
- 400Mhz frontside bus (FSB), which is faster than previous VIA CPUs and supports DDR2-400 or DDR2-533 memory. A faster frontside bus means higher memory bandwidth, which translates into improved performance of RAM-intensive operations.
- VIA PadLock Security Engine, which adds RSA, AES, SHA1, and SHA-256 hardware encryption/decryption for OpenSSL.
- Intel Streaming SIMD Extensions v. 3 (SSE3), which primarily speeds up multimedia functions.
- The VIA CN700 northbridge chipset is also part of the platform. It includes an MPEG-2 accelerator for playing DVD movies, 6-channel audio, and S3 UniChrome Pro graphics with the ability to output to two displays.
The complete data sheet can be found here.
Market impact
VIA’s marketing efforts for the C7-D will begin in Europe and Asia, where energy consumption and environmental considerations are of higher importance. “This is something new to the market, and we’re watching to see how it responds. In Europe this is a growing trend — companies trying to reduce their carbon footprint,” VIA’s Keith Kowal told me. When asked if competitors like Intel and AMD would feel challenged by VIA’s carbon-free CPU initiative, Kowal said that it would be much more costly because their processors can use more than seven times as much electricity.
Discuss this article or get technical support on our forum.
Copyright 2006 Jem Matzan.
|
|
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 2.5 License.
|


