Comment 1304 for bug 1

> > It means that the OS of apple doesn't need to be able to work on every
> > platform, with every proc, every graphic or sound card, etc....
>
> And I want it running on different hardware - desktops, servers,
> notebooks, netbooks, ... - oh yes, even mobile phones and routers.
>
> That said...

And i want all the suicide girls as my personal slaves. What is the most important thing, that Ubuntu/Linux grows larger, or that it fills every possible purpose we can think about? The second thing requires the first, so market share is more important. That will give us the resources to build more barracks... i mean features.

Also different hardware != every hardware. Supporting a small set of desktops, servers, notebooks and net-books will still qualify as focusing on a more narrow platform.

> > So, my question is:
> > What if ubuntu were manufacturing and selling it's own "all-in-one", apple-like, computers ?
>
> ...I always have difficulties in recommending hardware for new
> Linux/Ubuntu users.
>
> I am with you partly. I don't think that Canonical should be really
> enter hardware manufactoring. I think it would be sufficient to have
> some official recommendations of hardware combinations (maybe which
> are specifically tested). There could be shops then following the
> specs and build the appropriate machines and getting some sort of
> "Ubuntu optimized hardware" or something like this.
>
> But there must be at least one recommended combination of each genre -
> server, desktop, notebook and net-book.

I think some kind of certification program is the best way to go, in order to make sure that we have quality hardware and drivers. That combined with an OEM program that lets distributors sell a version of Ubuntu with commercial support included would deliver something that matches the Windows OEM program. Then there also should be central database where all Ubuntu-OEM:ed computers can be searched. And all of them must be certified and have hardware with top quality Linux drivers.

Remember that getting Ubuntu on the store-shelves is important for wide adoption. Most people do not buy Windows, they buy "a computer". And they let the salesman at their store tell them what is best for them. Therefor we must have a shiny commersial package that makes the salesman happy, and sells Ubuntu to the customers. A commercial edition combined with proper marketing will get us more users then free downloads ever will. Even if all they really buy is the service of buying "a computer" with pre-installed operating system.

Common users do want something that just works. They do not want to assemble their computer or install an operating system. They want to turn it on and start their browser. Most users even buy a computer with Office pre-installed, because they dont want to do it themselves. This is why a pre-installed OEM version will sell and cash in money that can be used to speed up development.

We currently have 1% usage share. How fast can we double it? When we reach 5% we are as large as OS X and commercial developers like Adobe will take the platform seriously.