ARM, IBM, Samsung and T.I. Form New Company to rollout Linux-Based Devices | Hacker News Reader

ARM, IBM, Samsung and T.I. Form New Company to rollout Linux-Based Devices

linaro.org

39 pointspclark19 comments

There's a vital word missing because the title needed to be truncated.

These guys are already rolling out linux-based devices, this new company they are forming together isn't going to produce devices, just to make that process go faster and easier.

Most of the comments think this is designed to compete with things (Android, Ubuntu etc.) that it's actually designed to work with and make better on the hardware platforms they care about.

Interesting reading in the FAQs:

http://www.linaro.org/faqs/

What's their commitment to actually mainlining their kernel contributions?

TI has git repositories for their OMAP boards[1], (and has for ages). What about the other major players in this space? (I'm thinking of MontaVista [2] and WindRiver[3])

[1]https://gforge.ti.com/gf/project/omapkernel/scmgit/ [2]http://www.mvista.com/product_detail_mvl6.php [3] http://www.windriver.com/products/linux/

From their FAQ:

Q10. How will you work with existing open source projects and the upstream?

A10. We will work with upstream projects directly. There are many existing projects that are creating great software. The best way of helping these projects is to join in, working with them and donating code directly to their development trunks.

Heres to hoping the following things come out of this alliance - a) Cheap and full featured devkits. b) Fast boot times for arm distros. c) An easy to setup cross compile environment, ideally with prebuilt arm packages. d) Eclipse / other ide support with the ability to run compiled arm apps in qemu.

All of these things exist seperately to some extent, but a coherent devkit that brings all this together and gives a newbie friendly dev environment would be great.

How about development boards that don't cost $1,000? I'm talking to you, TI.

People are having fun with OMAP 3430s, but only because BeagleBoard is $150. But $999 for a Sitara EDK? Forget it.

I wouldn't hope for too much, some of these companies just don't get free software at all. Samsung for example manages to pull off the trick of installing a pirated copy of linux on every consumer electronics device they sell by illegally relicencing other people's GLP2 code under the GPL3...

There has been similar other foundations in the past. Generally their work was in the form of a filesystem or distribution but it never picked up as mainstream. I hope this brings out something useful.

I suppose this will deliver a tools + kernel + distro kind of work.

I can understand ARM, Samsung, T.I. and others working together to push their embedded platforms/tools/chips. Not sure what IBMs reason is, though. Do they sell embedded devices/chips?

Also it does mention "Android, LiMo, MeeGo, Ubuntu and webOS". I suppose this will lead to some better driver support on the linux side and very lowlevel tools to support/push development of those operating systems on ARM platforms.

They sell large-scale solutions -- and have invested heavily in *nix.

Off the top of my head I can think of POS systems, hand-held data acquisition devices (car rental, inventory control, etc.), shop-floor control panels, and more.

I think T.I.'s involvement is interesting. Maybe a new TI-100 running linux (IBM) with a beautiful color display (Samsung), and the ability to morph from data acquisition to calculator to communicator to set-top player (ARM based).

TI sell the OMAP. That should be reason enough.

I see. Didn't know IBM builds those. Good to know, thanks. :)

IBM's big new push is "Smarter Planet".. Think: sensors on everything - electric grids, roads, etc... lots of small embedded systems.

(and IBM will try and sell you HW, analytics software, services, etc to run it)

Way to many massive competing companies involved in this for it to succeed. Bigger isn't always better....

Open source only works when you have massive competing companies contributing.

I'm not saying that, in fact I agree with you. However, 6 giant corporations all trying to align themselves to a common strategy doesn't work. Most contributions to open source by big companies are self serving: they enable profits for the company. I can't see TI doing work to enable Freescale to expand its marketshare.

In the best case the outcome will be the lowest common denominator: something so abstract its of no use to anyone doing real work.

I'd be happy to be wrong, but I don't think I am.

TI and Freescale are both in the same boat when we're talking about mobile Internet devices. They both have a variety of products that currently have a very small set of specialized software that runs on them. In order to gain traction against the x86 desktop and iPhone OS, they need platforms like Android and MeeGo to work flawlessly with minimal configuration on all of their hardware.

We're quickly getting to the stage where we want these devices talking directly to printers. All of the companies involved want to make sure that as much as possible, printing companies can write one driver and have it work anywhere. Better still would be if no custom driver were needed.

There's also the current issues with upgrading existing devices to newer Android releases. If they can streamline the low-level stuff to get updates out quickly on Google's heels, it will add considerable value to their products.

So there's definitely a lot of motivation for them to clean up the embedded Linux ecosystem.

That's a pretty absurd statement. Are Nokia and Novell massive competing companies? KDE seems to be doing pretty well. There are also plenty of small successful open projects like battle for wesnoth or the innumerable open libraries that make a programmer's life so much easier. Major projects tend to attract corporate sponsorship as they become useful, and paid developers go at a faster pace than hobbyists, but to say that open source requires contribution from huge companies is just silly.

Where is Google? Will this really fragment Android? They do say they want to contribute to Android.

Google is in Mountain View, California.

This won't fragment Android, this non-profit entity is designed to make installing Linux on an ARM licensee's silicon easier. There is still a bit of catching up the ARM folks have to do, this project aims to enable it.

I agree it looks like there will be lots of competing interests, but there should be enough of a common denominator amongst the ARM chip makers to at least make porting Linux to their platform less painful. That will enable Android - not fracture it.

Reply on news.ycombinator.com