Weekly Edition
Return to the Distributions page
| |||||||||||||
Shuttleworth: Linaro: Accelerating Linux on ARM
Mark Shuttleworth posts his thoughts about Linaro. Canonical will be working closely with the project, which has plans for six-month release cycles like Ubuntu. Linaro will also be using some Ubuntu infrastructure such as Launchpad.
"And finally, there are teams aimed at providing out of the box 'heads' for different user experiences. By 'head' we mean a particular user experience, which might range from the minimalist (console, for developers) to the sophisticated (like KDE for a netbook). Over time, as more partners join, the set of supported 'heads' will grow — ideally in future youll be able to bring up a Gnome head, or a KDE head, or a Chrome OS head, or an Android head, or a MeeGo head, trivially. We already have [good] precedent for this in Ubuntu with support for KDE, Gnome, LXE and server heads, so everyone's confident this will work well."
(Log in to post comments)
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 3, 2010 22:56 UTC (Thu) by DDevine (subscriber, #60717) [Link]
Canonical may not contribute upstream much but it is possible that they could redeem themselves by working with manufacturers to make more Linux compatible hardware.
Just my $0.02
As for the idea of heads, I think that is great but I shudder to think of a KDE Head on Ubuntu. We all know how bad Kubuntu is. Sadly it's likely that they will want to fiddle with Pulse Audio's implementation too.
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 3, 2010 23:53 UTC (Thu) by pflugstad (subscriber, #224) [Link]
What's bad about Kubuntu? I certainly like it better than stock Ubuntu. This is an honest question - my only recent point of comparison is Debian, which has KDE on it, but I always just SSH into it.
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 3, 2010 23:59 UTC (Thu) by dmitrij.ledkov (subscriber, #63320) [Link]
As far as I noticed "Kubuntu is bad" comparisons are usually done with distributions with KDE-by-default history e.g. openSUSE vs complementary projects on top of e.g. Ubuntu/Debian foundations.
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 4, 2010 2:32 UTC (Fri) by modernjazz (guest, #4185) [Link]
> As far as I noticed "Kubuntu is bad" comparisons are usually done with distributions with KDE-by-default history e.g. openSUSE
...and even then the supposed advantages of other distros relative to Kubuntu seem (to me at least) overstretched. Given the hype about the wonders of openSUSE's implementation of KDE, I was frankly disappointed when I finally tried it out and discovered that, by and large, most of the bugs affecting me personally were the same as in Kubuntu. And the ones it had squashed were largely compensated by some exciting new bugs (though not all in KDE). Moral: upstream matters more than anything else. So, it's a good thing that upstream is getting better all the time!
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 4, 2010 10:18 UTC (Fri) by dlang (✭ supporter ✭, #313) [Link]
I actually like the idea of the different 'heads'
I think it will lead to better infrastructure that can work with the different desktops (and/or with no desktop at all) rather than the current 'winner take all' approach that the major distros are taking where they work hard with one particular desktop, and pay very little attention to how the infrastructure/apps they are building would work/look on any other desktop.
diversity and competition is good
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 4, 2010 15:07 UTC (Fri) by drag (subscriber, #31333) [Link]
Welll....
If you want to have a clear division between what is the 'Core System' and what is the 'Desktop' then it's useful to have the ability to support multiple 'heads' to keep that division nice and clean.
But diversity for it's own sake is worthless. It's a tool, a means to a end. Quite often the value of different approaches are very minor compared to just the amount of work that goes into just making one approach work properly.
With a solid foundation for running applications the system as a whole becomes much more useful and flexible then a system, which traditionally has happened with the Linux desktop, which users and application developers are forced to choose between solutions that are broken in different ways in a hope that they find something that is broken in such a way that it will not affect them.
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 4, 2010 14:07 UTC (Fri) by salimma (subscriber, #34460) [Link]
Experience leads me to doubt this is happening -- they collaborated with Broadcom on creating the STA driver, but it's not only closed-source, the documentation is so puny that last I recall, they don't even have a ChangeLog. Nor is there a centralized bug tracker, Launchpad or otherwise.
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 4, 2010 15:19 UTC (Fri) by crimsun (subscriber, #13750) [Link]
If by "fiddle" you mean "use Mandriva as the reference", then I suppose it's all kilter/fodder for grumbles, eh?
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 7, 2010 4:09 UTC (Mon) by luya (subscriber, #50741) [Link]
I think PulseAudio itself is not the problem but the way a distribution implemented it and does not submit patches to upstream.
So Ubuntu has new friends Posted Jun 4, 2010 10:41 UTC (Fri) by rvfh (subscriber, #31018) [Link]
This is good news for Ubuntu, as Intel left them for Fedora as the base of Moblin 2, which is now Meego. This means we have two main mobile Linux stacks:
* Meego for Intel and Nokia * Linaro for ARM, Freescale, IBM, Samsung, ST-Ericsson and Texas Instruments
This is also good news for Debian I think, indirectly. Maybe some of the Maemo developers will decide to switch to Linaro just because of the Debian/Ubuntu base.
Of course, one may regret no to see all these guys working together in one project, but I hope it will bring the customers (us) a healthy competition and thus better products. As long as they don't develop another behind closed doors locking mechanism...
What I wonder now, is whether ARM on server is what explains IBM's presence in this company.
So Ubuntu has new friends Posted Jun 4, 2010 11:22 UTC (Fri) by Frej (subscriber, #4165) [Link]
You do know that nokia doesn't use intel chips, except for the booklet, or produce their own. You can't divede meego into chipsets.
So Ubuntu has new friends Posted Jun 4, 2010 11:44 UTC (Fri) by rvfh (subscriber, #31018) [Link]
Sorry, I am missing the relation between your comment and mine. I was talking about Linux distributions for the mobile world (up to my last sentence) and not about chipsets...
And the whole point of the blog post is to show how Linaro will allow different ARM-based chipsets to use the same stack, which is apparently not the case now.
What did I miss?
So Ubuntu has new friends Posted Jun 4, 2010 13:16 UTC (Fri) by emk (guest, #1128) [Link]
Well, three main mobile Linux stacks, if we count Android. It's definitely a _weird_ Linux (at least in userspace), but it's there and it ships with a lot of devices.
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 4, 2010 14:56 UTC (Fri) by aryonoco (subscriber, #55563) [Link]
Android is already well established and will soon be on tablets and other such ARM products.
MeeGo had a very strong showing in Computex, and looks like there are going to be many Intel Atom netbooks and even tablets by the likes of Acer and Asus with MeeGo coming in 2011. No one really knows what Nokia is doing with it, but they will be using it on ARM.
ChromeOS is coming "this fall", that will be interesting to watch as well.
And then, you've got the Ubuntu on ARM initiative. Personally I don't give Canonical much chance of standing up to Google and Intel and Nokia, but if they can execute this Linaro thing well, they might have a chance.
The linux space has never looked so interesting. The year of Linux on the desktop never actually came, but desktop itself is at least is becoming very much a marginalised form factor, if not a passe one yet.
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 4, 2010 20:34 UTC (Fri) by rvfh (subscriber, #31018) [Link] You're perfectly right. And not only is it Canonical vs Intel/Nokia on the software stack, but of course it is also ARM vs Intel on the hardware stack. Intel coming down from PCs;
Hmmm... are we not forgetting one player here? One that runs its OS on Intel and ARM... one that just got its tablet out, after making a hit with its phone... Indeed, the
Year of the Linux netbook Posted Jun 6, 2010 14:49 UTC (Sun) by man_ls (subscriber, #15091) [Link] desktop itself is at least is becoming very much a marginalised form factorA bit exaggerated. The desktop is still the main computing platform for the majority of people out there; people may use their cellphones for casual browsing and for social networking interaction, and little else. As for tablets and PDAs, they have been around for ages and still haven't caught on with the desktop. And netbooks, while largely desktop replacements, have not changed things much. Well, apart from making Microsoft extend XP and redefine Windows 7; they were major achievements (the year of the Linux netbook came and went), but conceptually they did not redefine desktops too much.
Shuttleworth: Linaro: Accelerating Linux on ARM Posted Jun 10, 2010 11:10 UTC (Thu) by NRArnot (subscriber, #3033) [Link]
Software availability is one side of the problem, hardware availability and pricing is the other.
I was recently looking for an always-on quiet low-power (i.e low watts) server and occasional www-access station. I'd have liked it to be non-Intel (less watts for one thing), but I can't find any equivalent to an Intel Atom board at a sensible price.
On the offchance: can anyone reading this suggest an ARM alternative that meets the requirements at a similar price?
Requirements: Ethernet, USB2, 1280x1024 VGA, SATA disk.
(Prices in UK#) Intel BOXD510MO 49, RAM 20, Case/PSU 43, 500Gb HD (2.5 inch) 51, Total 163. Option to save a bit on both watts and price with D410 instead.
|
|||||||||||||