tag:blogger.com,1999:blog-32725340.post6617457822696936964..comments2008-06-10T15:33:31.709-07:00Gerrithttp://www.blogger.com/profile/05234056324083766726noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-32725340.post-58633209038977771192008-06-10T12:16:00.000-07:002008-06-10T12:16:00.000-07:00thanks.. that was usefuljaisonhttp://www.blogger.com/profile/08131786618167798996noreply@blogger.comtag:blogger.com,1999:blog-32725340.post-57064287066698807072007-09-21T14:38:00.000-07:002007-09-21T14:38:00.000-07:00Gerrit, glad to see that you are spending a little time in desktop space. As initiative manager for the LF Desktop Linux workgroup, I have spent some time thinking in desktop space as well. You have a good handle on some of the major problems, especially the problems that ISVs have in porting applications to Linux. Your solution is also interesting, but it reminds me of the United Linux fad that we experienced a few years ago. You know, that might have worked if one of the flagship distros hadn't bolted with the rest of the community.<BR/><BR/>Some of the blockers for mainstream Linux desktop usage are lack of legal codecs, hardware integration and certain types of basic functionality like software suspend not working reliably. This is not news, of course, but rather the typical gripes people have had for years.<BR/><BR/>While we have waited around for the Linux distros to solve these problems for their customers, this has not really been the case. Sure, there are the promises of CNR and other "easy" mechanisms to bring a Linux desktop to the point of "just working". However, the landscape has changed over the last few months and now there are at least 4 companies producing general purpose computing platforms (desktops/laptops) with pre-installed Linux. These companies include Dell, HP, Lenovo, and Acer.<BR/><BR/>So what does it take to deliver a pre-installed product that consumers will buy? Can you take a popular distribution like Ubuntu and slap it on some hardware (that has the Linux driver support that is needed) and you are off and running? In a recent Wall Street Journal article, a popular pre-installed Linux product was tested by an "average" consumer. Out of the box, it was discovered that the user could not play mp3s, could not play DVDs, and the touchpad continually got in the way. When I read this, my heart sank because I figured that when an OEM produced a Linux-based product that they would handle the integration of codecs and usability capabilities that would "just work" and meet the expectations on consumer desktop users. MY expectation was that the OEM would add value to the integration of an operating system and hardware. Integration that is difficult for platform-independent distros to accomplish should be done by the OEMs that are actually delivering hardware to the market.<BR/><BR/>So what can the Linux community do? Now that OEM vendors have entered the ecosystem, it is vitally important that the OEM vendors participate in the Linux communities (kernel, system software, middleware, desktop, applications) so that OEMs can deliver real products that meet customer expectations with components that are actually mainlined (not one-offs for specific distros).<BR/><BR/>What strategy can OEMs take? I would suggest that OEMs target products for market niches, rather than throwing products out to the general computing markets. These products should be pre-configured with everything needed by that market niche. In some cases (like Tivo), the user may not even need to know they are running Linux. However, if a product is integrated for a specific set of tasks, such as multimedia and video editing and if the product "just worked", the use of Linux would start to gain ground in traditional desktop markets.<BR/><BR/>John Cherry<BR/>Linux FoundationJohnhttp://www.blogger.com/profile/17477205500874184244noreply@blogger.com