Bringing Up Hardware First In Linux, Then Windows

[droidhacker]

Windows doesn't have 99% marketshare anymore on desktops and they lost almost all of their marketshare in the mobile market by removing all backwards compatibility in Windows Mobile 7. Just saying.

Now a common hardware architecture among all operating systems would be good for both the companies paying for driver development and for Linux HW compatibility list.

Having wrappers might be even better for Linux, given not all smartphone companies release newer versions of Android and having proprietary drivers kills it for the customer; he/she has to by a new phone just for an Android update.

On the other hand this may lead to "Oh we can just make proprietary drivers for Linux.", while companies are currently more or less forced to GPLv2 it.

On one hand I like to have Android updates for the rest of my phone's life, but on the other hand I like that the world needs to bow for FLOSS.

But writing drivers for Linux first and then for Windows might not work.

It is a little premature to be talking about android updates requiring new hardware due to lack of source for the hardware drivers... at this point, I can't think of a single android handset for which there are no drivers available supporting the latest version of Android. The very first android handset sold -- htc dream (one of which I have sitting about 10 inches from me this very second), has hardware drivers for everything up to and including android 2.2.

Not that I don't agree with you in theory though, since clearly you can expect at some point that there could be a problem with hardware driver support. Good news though, is that as Android matures, the hardware driver requirements should/will_hopefully stabilize to the point where the older drivers will be reusable on newer versions. What always has been open source, and will therefore continue to be useful for future updates, are the kernel/driver glue. We can thank GPLv2 for that. This open source driver glue was very useful in the initial attempts to get 2.2 running on older hardware -- by using the 1.6 driver blobs.

Now Android is a very different case to general desktop linux. Google has arranged it as a commercial linux desktop and has managed to be extremely successful with the platform. One of THEIR goals is to make hardware support EASY, so having a stable API/ABI is much higher on the list of priorities for this distro than for other distros.

[DebianAroundParis]

I never said anything of the above and looking at your first replay you are the one that sounds frustrated to me. Yeah it's okay, trying to put bullshit into other mouths. And you wonder why I called you idiot?

Just because I am not working on the Linux kernel does not mean that I am not entitled to have an opionion, that I am not entitled to talk about that opinion. If you aren't happy with that don't read my posts. And if there is so much wrong in my posts be free to correct them.

Regarding kernel devs, well those are human too and contradict themselves sometimes as well, so just because they/some say something does not mean that it is always "right" -- as if there was something like that -- and in fact I do _not_ imply here that it is wrong, just that you can shove your "shut up" up your ass.

You are inferior to kernel devs as far as understanding what is best for the Linux kernel. Inferior people should not lecture superior people, unless they absolutely want to make themselves ridiculous. You should talk to a shrink to understand why you are neither wise nor modest.

[nanonyme]

Inferior people should not lecture superior people, unless they absolutely want to make themselves ridiculous.

Then why do you keep on doing it?

[RealNC]

Uh, no they don't. They just need to release the source so it goes into the kernel and is updated as required for API changes.

You're so smart, aren't you.

[PsynoKhi0]

So erm *cough* back on topic...
Although a nice idea I fear that would fall on deaf ears for companies at large: patents, misc. trade secrets and what not.

That or some major corporation would supply earplugs. (And unfortunately I can't see all vendors telling said major corp to stick it up their ass...)

Isn't it more profitable to artificially control the life span of your products through proprietary drivers even if it means wasting development costs reinventing the wheel?

[Prescience500]

I may not be a computer programmer, but I do know markets and the economics. I don't think it likely that firms would bring up hardware support in Linux desktops/laptops first in the near future unless the drivers were open source AND at least partly community developed. If open source video drivers are eventually able to improve past a certain point relative to the proprietary drivers, then this may happen. It might also require components of the open source driver to be used on other operating systems like Windows.

[Yfrwlf]

Uh, no they don't. They just need to release the source so it goes into the kernel and is updated as required for API changes.

Stable APIs are the reason why Windows is such a colossal pile of security holes; they're the last thing Linux needs.

I'm still struggling to understand why you can't have a driver interface that is high level enough to where any kind of security holes won't actually need to require changes in that standard interface. Even if it did, you just require a new standard and the driver devs update their driver for the new standards which should hopefully be really easy.

It's good that Linux encourages/demands open source drivers the way it does things currently, but I simply wish that things could be made more modular so that things like driver installation was much easier for users, and driver upkeep was easier for developers. Both things would benefit the Linux ecosystem in many ways, like this whole mess with needing to backport drivers, and it would allow companies to ship Linux drivers along with their products.

WHY CAN'T WE HAVE THE BEST OF BOTH WORLDS???

There has to be a solution to every problem, that's all.

[madjr]

i hope this becomes true

[BlackStar]

I'm still struggling to understand why you can't have a driver interface that is high level enough to where any kind of security holes won't actually need to require changes in that standard interface. Even if it did, you just require a new standard and the driver devs update their driver for the new standards which should hopefully be really easy.

It's good that Linux encourages/demands open source drivers the way it does things currently, but I simply wish that things could be made more modular so that things like driver installation was much easier for users, and driver upkeep was easier for developers. Both things would benefit the Linux ecosystem in many ways, like this whole mess with needing to backport drivers, and it would allow companies to ship Linux drivers along with their products.

WHY CAN'T WE HAVE THE BEST OF BOTH WORLDS???

There has to be a solution to every problem, that's all.

How does that differ from the current approach? Right now, the abstraction doesn't change unless it must and, when it does, affected drivers are updated (with usually simple patches).

Interested companies seem to be able to ship drivers without issue along with their products. It's been years since I've seen unsupported hardware. In fact, companies seem to abandon Windows first, not Linux. I have (musical) keyboard, a Creative SoundBlaster, an AverTV card, a nForce2 motherboard that work perfectly fine on Linux but do not work *at all* on Vista/Win7.

[alizard]

For the most part, bringing up a given device in Windoze is going to be a matter of updating the driver(s). That gets the hardware vendor a chance at the extremely overcrowded Windows market said vendor already had access to as of the last Windows-only model.

Doing the work needed to add "Linux compatible" to the device web page gives one access to the overwhelmingly Linux server market and of course, the small but growing Linux desktop market. Even computers not intended for server usage will work at the low end of the webserver spectrum, and a vendor who doesn't care to support Linux will simply have professional customers go somewhere else.

Everyone who puts "Linux compatible" and [device type] into google will have a list with the vendors that sells Linux compatible devices on them.

And for every distro that's listed on a given device page, the chances that people will find that device increase.

Vendors spend lots of money on search engine optimization.

The best way to increase one's page hit rate and sales if one is selling commodity computer hardware is to have keywords on a page like in this case, Linux , Debian, Ubuntu, SUSE, Red Hat, and for that matter, OpenBSD and Solaris. It should be obvious that the more keywords that describe product features, the better for sales. It only takes a few sales to pay for the developer time required to make any given product Linux-compatible, after that, Linux sales are simply more profit.

NOT doing this is leaving money on the table, doing this on purpose or due to ignorance is an unsound business practice.