| |||||||||||||
|
| |||||||||||||
Another Look At Linux Netbook Return Statistics (bMighty.com)Another Look At Linux Netbook Return Statistics (bMighty.com)Posted Apr 9, 2009 18:42 UTC (Thu) by jspaleta (subscriber, #50639)Parent article: Another Look At Linux Netbook Return Statistics (bMighty.com)
That article draws some very interesting conclusions at the end.
"In this case, the supporting data suggests that a satisfying Linux netbook experience depends heavily upon an OEM's willingness to invest in effective, appropriate adjustments to the underlying operating system."
The take away for OEMs is this...they need to actually spend money doing their own QA and customization work to make sure the functionality of linux based devices they are offering meet the expectations of their users. There is no free lunch for OEMs. There is a gap between what a linux enthusiast is willing to tolerate and what a retail consumer is willing to tolerate in terms of expectation. OEMs are only going to cross that gap by spending money funding development work.
Trying to take a general purpose distribution, installing it and shipping it as is isn't going to work. OEMs are going to have to spend money tuning a linux based OS for their device. Once OEMs really take that to heart, the really interesting question is how to most cost-effectively do that work. Is it more effective for OEMs to contract out to linux integrators to do those adjustments or will it be more effective for device OEMs to have their own internal teams in-house? That will be interesting to watch as the value proposition surrounding consumer oriented OEMs services evolves.
-jef
(Log in to post comments)
is that a question? Posted Apr 9, 2009 18:47 UTC (Thu) by eean (guest, #50420) [Link]
I've never heard of a netbook OEM not hiring someone else to do the
software.
is that a question? Posted Apr 9, 2009 19:32 UTC (Thu) by jspaleta (subscriber, #50639) [Link]
Sure, and when the first traditional "mobile" devices started to run linux hardware vendors initially partnered with linux integrators...but things evolve. For example Motorola eventually unveiled its own linux platform for its mobile devices: http://www.motorola.com/content.jsp?globalObjectId=8411
and now things are evolving again as motorola looks at android as a platform. http://phandroid.com/2008/09/28/motorola-android-team-350... The point however is, is that over time Motorola has a device manufacturer has made a larger and larger investment in in-house manpower the deeper their commitment to a linux based operating system has become.
The evolution of the OEM netbook experience could take a similar turn as the OEMs in that space become more aware of where the real value lies in going with linux..direct control over what the device experience is for customers.
This could be more true of the upcoming ARM based device OEMs, which are going to end up marketinging their products narrowly as tailored web-centric devices more than general purpose computing devices to differentiate them from the intel atom based devices which can run a full windows desktop. The ARM manufacturers may have more of an incentive to take more control of the software stack and push software where they need it to go to get that tailored experience.
-jef
Another Look At Linux Netbook Return Statistics (bMighty.com) Posted Apr 9, 2009 19:48 UTC (Thu) by dbruce (subscriber, #57948) [Link]
I have a Dell Mini 9 Ubuntu netbook (actually it belongs to my fifth-grade daughter) and it is impressive - polished, attractive, fully usable with no previous knowledge of Linux or Free Software.
Our local Best Buy had a Asus eee with Xandros on display that seemed obviously inferior to the adjacent Windows netbooks.
That "last 10%" of tuning is immaterial to anyone who is already a GNU/Linux expert, but makes all the difference in the world to the general public.
I don't think linux netbooks are going away. The progressively lower price points will either drive Windows licensing down to utterly cannibalistic levels, or cede this market to linux, not to mention the possibility of ARM-based netbooks that won't run Windows at all.
It's a one-way street.
Another Look At Linux Netbook Return Statistics (bMighty.com) Posted Apr 9, 2009 20:27 UTC (Thu) by leoc (subscriber, #39773) [Link] the possibility of ARM-based netbooks that won't run Windows at allThey could run Windows CE/PocketPC (or whatever they call it these days). The thing about that is there are not nearly as many applications for this version of Windows as there are for their regular Intel versions, so it would level the playing field quite a bit more.
Another Look At Linux Netbook Return Statistics (bMighty.com) Posted Apr 10, 2009 13:47 UTC (Fri) by mfuzzey (subscriber, #57966) [Link]
Not only that it would break the Windows consumer "grab this setup.exe and run it" reflex. Suddenly package managers and their associated repositories of already ported software which automatically get the right architecture start to look more interesting to the average joe...
Another Look At Linux Netbook Return Statistics (bMighty.com) Posted Apr 10, 2009 17:00 UTC (Fri) by jspaleta (subscriber, #50639) [Link]
Maybe, netbooks as a class of devices aren't going to be expected to be as flexible as traditional computers in terms of range of software applications. Maybe netbooks will be a platform for the web browser as middleware with the vast majority of applications will actually be run in the browser interface. Maybe netbooks really will evolve to be more like smartphones than traditional laptops.
http://texvc.com/2009/03/21/sxsw-taught-me-that-netbooks-...
I think its worth noting HP's approach to installable software in their MI interface.
Built on Ubuntu, but by default the gui package updater only exposes HP pre-approved applications, not a full range of available Ubuntu packages.
In this way HP takes control of the quality of the experience for the target user by limiting the range of applications that are "easily" installable through the designed interface. But at the same time gives more traditional linux users the ability to install a wider range of software. Are those two distinct audiences with two distinct sets of expectations?
The real question is how good of a job does HP interface do at keeping the target audience inside the walled garden (a very low wall to be sure, more like a raised patio edge than a wall) of pre-approved applications? Are the majority of of HP MI purchasers choosing to live within the functionality HP exposes by default..or in the pre-approved additional applications listing? Or are most consumers of the device hitting synaptic and piling in additional functionality that HP has not vetted? I'm not sure HP has the telemetry to really answer that, but it's an important question. Is the walled garden approach good for the less technically inclined mass audience or is it just an unneeded barrier?
-jef
Another Look At Linux Netbook Return Statistics (bMighty.com) Posted Apr 10, 2009 19:56 UTC (Fri) by iabervon (subscriber, #722) [Link]
How many distro package maintainers work for any given OEM? I think the most effective system (setting aside cost factors) would be for OEMs to do their own QA and contract out the adjustments. The internal QA is good, because it gives tech support access to people with experience on the quirks of the system, but there's a huge benefit to contracting out the adjustments to the distro, since there's a good chance that you'll benefit from the results of somebody else contracting out similar adjustments. And you also benefit from having the changes go upstream; that is, if you install Ubuntu Hardy on your netbook and tweak it until it works nicely, that'll get good results until your customers decide to install Jaunty on it and have no idea how to make the trackpad feel right. If you pay Ubuntu to detect your trackpad and use defaults that make it nice to use, this carries over to people installing Jaunty or, if those defaults get into upstream projects, even Gentoo.
|
|
||||||||||||