Microsoft’s Tablet Strategy and How Linux Compares
Last week at CES, Microsoft announced their answer to the iPad with their tablet strategy. Computerworldsays, “Microsoft has decided not to follow the Apple and Google route of putting its mobile operating system on tablets. Instead, Microsoft has chosen a more deliberate method where it will migrate its client OS onto tablets.” Microsoft also announced it will wait until its port of Windows 7 to ARM chips is complete. Pundits have criticized this strategy as being out-dated in today’s fast moving tablet market.What exactly are they criticizing?1. The time it will take for Microsoft to ready this system. Because Microsoft is waiting for its next release cycle, this Windows Tablet OS isn’t expected until 2012. That’s a lifetime in a market like this.2. Microsoft is taking a desktop client and force fitting it into a tablet form factor. That likely means tablets running Windows will need more processor power and have shorter battery life than those running Linux-based or iOS variants. It wasn’t designed to be flexible and run in different computing environments — it was designed for resource-hogging desktops.So how does Linux’s tablet strategy compare? Well, it doesn’t. Unlike Microsoft, Linux isn’t controlled by a single entity. Linux is instead powering a variety of vendors’ tablet strategies today. You may have seen the new Motorola Xoom tablet announced at CES ? Based on Linux with an Android VM. The Samsung Tablet? Also Linux. And the hot selling mobile OS Android? Also Linux. Vendors are free to pick and choose what they want from the Linux platform and customize it on their own terms. Because of the nature of collaborative development, this speeds development.And of course Linux vendors benefit from the network effect of shared development. When one of those vendors/projects improves battery life for their project in the Linux kernel, all other projects benefit.Linux already works on ARM, and virtually every other architecture. Linux was designed for architecture portability and because it’s open, companies like Intel and IBM can work on optimizing the code for their platforms. This is one reason there is so much cross-architecture support.The Linux kernel has a release every three months, not the three years of Windows releases. This means you can create software advances that match the state of hardware as they happen.Microsoft is trying to adjust to a new game using old game pieces, which is making it hard for them to move quickly. Microsoft’s legacy model (and its success) makes it hard to change. Of course you can never count them out; they’ve defined desktop computing and continue to grow in those area and others (take a look at Kinect). But one can see that when flexibility and speed are called upon, it’s difficult for them to keep up.
Annual Kernel Development Report: New (and Old) Faces
Today we are pleased to publish annual report on Linux kernel development, detailing who does the work, who sponsors it and how fast the Linux kernel is growing.
The paper documents how hard at work the Linux community has been. There have been 1.5 million lines of code added to the kernel since the 2009 update. Since that last paper, additions and changes translate to an amazing 9,058 lines added, 4,495 lines removed, and 1,978 lines changed every day, weekends and holidays included.
The other good news is that in the list of sponsoring entities we see more mobile and embedded companies participating in Linux kernel development. We see companies such as Nokia, Texas Instruments and Renasas moving up the list of companies who sponsor Linux development. This certainly should not be a surprise given the rise of Linux usage in devices over the last few years. This is great to see, even though the traditional Linux supporters are still at the top of the list: Red Hat, Intel, Novell and IBM.
This paper documents a bit less frenzied development than the last one, which was expected given all the new features of 2.6.30 (ext4, ftrace, btrfs, perf etc) as well as the peak of merged drivers from Linux stable tree. Regardless, this report continues to paint a picture of a very strong and vibrant development community.
I’d like to congratulate Paul Mundt who had the most individual contributions to the kernel, equally 1.3%, since our paper last year, and give my heartfelt thanks to Jon Corbet and Greg Kroah-Hartman, kernel developers and members of our Technical Advisory Board, who really do all the work on this paper. This publication is an important one as it gives a rare glimpse into the world of kernel development. Without their knowledge of the participants and their technical tools to analyze the code, this analysis would not be possible. If you’re interested in kernel development and want to support the work of Jon, please consider buying a subscription at lwn.net.
=====================================
Linux Powering Your Cell Phone Network: A Case Study
We’ve talked a lot about the rise of Linux in embedded devices lately: from our embedded Linux training classes to the Yocto and Meego projects, to a new Linux Foundation fellow. But what about the end users, the people who are deploying Linux in their products?
We just published a Linux training case study on Optelian, a company who designs and manufacturers optical transport systems that send data across optical fiber. That means if you connect your Android phone (or iPhone or IPad or Blackberry) to a telecom carrier in North America (and if you don’t, why do own one of those devices?), you’re likely making use of Optelian’s handywork. It’s yet another example of “Linux is everywhere.” This time it’s in your phone (in Android’s case) as well as the network that is delivering the data to those phones. Optelian migrated to Linux after using a proprietary platform for years. Why?
“It came down to two things: needing to scale and cost,” said Paul Beer, software manager at Optelian. “The amount of money our vendor wanted for the upgrade was just not right for our business. Embedded Linux has really grown as an option since the last time we looked into it.” But Optelian found other benefits. “The more we looked into it the more we realized that the drivers for the specialized devices for optical transport – like lasers – are already in Linux and supported, which really helps us. Cost is only one aspect: our chipset suppliers offer APIs for Linux already. Previously we had to do the porting for the software ourselves. This saves us time and money.”
In order to begin using Linux, Optelian needed to conduct comprehensive Linux training for its developers. After a lengthy search, they selected Linux training from the Linux Foundation. Paul said, “They really had the best credibility out there, and they were flexible and tailored the class to what I needed for my developers.”
Optelian is thriving as their company and its developers have fully made the transition to Linux. You can read the entire Linux training case study if you’re interested in more detail, but I was glad to focus a little light on an actual end user in the embedded space. Also check out our Linux tutorials for free training content on embedded and other topics.
