|
|
Jim Zemlin on Moblin and the well-hedged OS
2008-12-12 Article Rating:  / 0
In an interview with the Linux Foundation's Jim Zemlin, the group's executive director explains why Moblin is "the project you want to watch." Zemlin (pictured) also touts Linux for being ideally "well hedged" to survive the recession, but points to a future obstacle for the OS.
For a man whose job at the LF -- in part -- is to promote the awesomeness of Linux, Jim Zemlin's optimism is not surprising. Yet in the face of what looks to be the worst recession since the '70s, if not the '30s, one might assume a bit of qualified pessimism might be in order. Zemlin will have nothing of it, however, and he offered a compelling argument why Linux should ride out the recession just fine. (And it's not just about lower costs.)
The recession was only a footnote to the main topic Zemlin wanted to discuss: why the Intel-sponsored Moblin.org project could change the face of the emerging consumer electronics platforms that straddle the line between desktop and embedded devices. While commonly known as the open-source development toolkit underlying mobile Internet devices (MIDs), the Moblin stack is also increasingly being put to use in netbooks, says Zemlin. Moblin could also have a big impact in other verticals such as automotive infotainment, he says. The following are some excerpts from our discussion of Moblin and the state of Linux in general: Q1 -- Certainly the connection between MIDs and Moblin has been well documented, but do you also see it being used in netbooks and various verticals? A1 -- The mobile Internet device is one of the most promising categories. But the netbook category has surprised a lot of people, and it has become an unexpected driver of Moblin. I do think that MIDs will drive Moblin in 2009, though. With MIDs, you're seeing a lot of different device makers and form factors. It is kind of like the early days of laptops, people trying different form factors. But I think that will all be sorted out in 2009. Moblin has a tremendous number of OEMs, a very large ecosystem. They've got dozens and dozens of ISVs lined up, and they're just getting started. I tell people that this is the open source project you want to watch. Q2. Why do you see such momentum finally occurring with Moblin? Is it primarily due to interest in the Intel Atom, or the MID and netbook formats, or is it more about the tools and platform itself? A2 -- Moblin has used technology from OpenedHand [Matchbox] to produce a brilliant platform that delivers real desktop applications in a really competitive form factor. It's also 100 percent open source, from the Linux kernel to the GTK libraries -- it's all GPL. You also have features like power management and fast boot time that are pretty compelling, and you've got Intel behind it. Q3. Some observers have suggested that Maemo, LiMo, and even Android could at some point compete with Moblin. Do you see much potential for that? A3 -- LiMo and Android are great, but they're very different. Android is basically a Java-like environment, a different set of middleware and components, so you couldn't expect a Linux application to run on an Android phone. But Moblin runs Flash and Skype and the vast array of Linux applications. Moblin is already using the LSB framework [the LF-sponsored Linux Standards Base], leveraging the work the LSB has done toward binary compatibility. That's very compelling for an ISV that wants to support multiple SKUs in the market, whether the SKU's software is branded as "Moblin" or not. Other projects talk about having an SDK in the market; well, Moblin has done it. Other guys are talking about application portability; well these guys are on it. Q4 -- Is the primary draw of Moblin the fact that it's so tightly focused on a single processor, in this case, the Intel Atom? A4 -- Actually, Intel's biggest competitors could go and implement Moblin on any chip architecture they want. Intel is being very very good about not closing that down. It has not tried to corner market on this code. If I was a competitor I would be jumping all over it. Q5 -- Will x86 ever really be able to compete with ARM in battery-powered devices like MIDs, given that x86 uses so much more power, especially when idle? A5 -- X86 can absolutely win in MIDs. I see Moblin coming to the table with real innovation. The Intel/Moblin PowerTop and LatencyTop projects are just brilliant. Have you seen the demo of Moblin 2? The usability, the 3d physics -- it's cool stuff, really compelling code. Moblin is walking their talk, bringing real innovation. Q6 -- Over the last year, there have been increasing efforts to reach out to the open source community from Via Technologies, ARM, and Texas Instruments. Are they following Intel's lead here? A6 -- There are smart companies out there that are definitely doing that. Intel has a long history of doing this well. They have done it with Linux, when competing with Sun Microsystems before, and they're going to do it well here. Intel has been very sharp about working with the community, keeping it all open source. Q7 -- You recently predicted that Linux may outship Windows on the desktop next year, thanks to the fast-booting Linux environments increasingly bundled with PCs. In fact you wrote, "We may see a world at the end of next year where Linux ships on almost every notebook computer regardless of whether it is loaded with Windows." Considering that the Linux share of PC sales remains so low, how much of an impact can fast-boot really have on the Linux development market and the acceptance of Linux in general? A7 -- Clearly, more people are using the web for applications, and as a result, the desktop operating system has become less important. If people are primarily booting up to check email and browse the web through fast-boot, they're not booting into a Windows experience. It makes Microsoft less essential. They're not seeing the Windows logo, and not seeing a UI that they get accustomed to and don't want to give up. It opens up the door to alternative operating systems, which gives hardware makers more leverage in the market, due to competition. Q7b -- So the recent Linux success in netbooks is going to spread upward to notebooks and desktops? A7b -- Linux will not displace Microsoft on desktops in short order, but there is a combination of trends that is leading that way. One is the cost. If a netbook sells for $200 or $300, and if a Windows license is $50 to 75 of that, it becomes untenable. Linux also gives vendors flexibility. You see this with Asus, which wants to custom-brand their operating systems. As vendors get more experience building their own consumer brand, higher margins will come along. And, if you want to build a cutting edge industrial design device, you're going to have to have access to the source code. Likewise for doing things like fast-boot or power optimization. So, to do those things on Windows, you have to go to Microsoft. Q7c -- Yet Windows devotees will tell you that customers will continue to choose the operating system with the largest selection of software and hardware support. A7c -- True, one of the strengths of Windows is that almost any application is available, and although there are many applications on Linux now, the gap remains. Also, Microsoft Office is a really popular application. It's also true that Linux can be confusing to consumers -- which version of Linux are you running? I think that all this will sort itself out in time. A lot of the functional weaknesses of Linux have largely been closed. Almost all wireless cards are now supported by Linux. Before, you had to reverse engineer drivers, but now manufacturers are increasingly supporting Linux out of the box. Q8 -- Okay, so there's this little thing they're now officially calling "the recession." How bad is it going to get for Linux developers and vendors in this market?
Q9 -- You've mentioned the price advantage. What else does Linux have going for it in these tough times? A9 -- Linux is agile, and it's also hedged properly. Ask yourself, how many companies would need to go out of business for Linux to fail? Nobody thought that DEC would have failed or Sun might fail. The difference with Linux is that there is no single company. It's a combination of collective efforts. It's the first time we have ever had anything like this, such a massively hedged effort. Q10 -- So is there nothing that can stop the Tux juggernaut? No legal threat or other doomsday scenario lurking in the wings? A10 -- I don't think there's anything to slow it in the near future. Certainly, it could slow over a long period of time if Linux was unable to innovate, but there's no sign of that. The other potential problem is the lack of skilled labor. That's why we're trying to run LF events that offer training programs. Labor is going to be the big bottleneck for these companies using Linux. If you are an engineer who was recently laid off, I would go learn Linux. There is no shortage of jobs for Linux developers. Related Stories:
|
news feed
