All I am trying to say is that if you look at where Ubuntu has an advantage over its competitors, the answer is the desktop. For this reason, it makes sense to push the desktop, because winning on the desktop side gives you an advantage on the server side, something they don’t have now.

I hope that clarifies my point some. If you still disagree with me, and I am sure at least some people will, go ahead and post your view.

They have been pushing their server edition in web-based ads a lot lately… or at least it’s like I’ve seen them EVERYWHERE… so it appears they are moving that direction… or at least testing the water.

The only problem is though, that Ubuntu’s main competition in the server space isn’t Red Hat nor Novell, it’s Debian… and in a battle between the two for servers, I think Debian will win.

Eventually Canonical will need to make money, right? What happens then? How much of Ubuntu is Canonical? We hear that Ubuntu is sponsored by Canonical… but what percentage of the packages are made by Canonical developers and what by actual, non-Canonical Ubuntu community members.

Don’t get me wrong, I’m NOT trying to bash Ubuntu the distribution. It is rather nice.

I would like to see something between Red Hat Enterprise Linux (18-24 month release cycle with 5 years of support) and Fedora (6 months release cycle with 18 months of updates). How about something that releases every 12 months and offers 3 years of support… with median price?

Corey

People used to claim that “a real-time kernel is only good for enterprise”, yet when it actually got into the kernel, consumers rejoiced because the didn’t have to choose between having audio and video skip or have a sluggish system. VMWare and KVM are enterprise features, yet it’s something that consumers can use to wean themselves off of Windows. Having a simple system with good defaults that’s easy to learn is prized in enterprise, yet it also wins over new consumers. Having an OEM install is good for enterprise, yet that same OEM install is one thing that allowed Dell and others to pre-install Ubuntu. Conforming to the Americans with Disabilities Act and similar legilation is a very enterprise thing to do, yet it also helps the consumers with visually impairment (including the aged with poor eye sight or people who have problems with color and contrast) or poor motor skills. Having multiple language support out of the box is similarly enterprisy, but it helps consumers at home. Being able to manage multiple PCs simply is also enterprisy, but it’s also important for schools and home users.

The list goes on. As Linus points out, if you do things right, there’s no need to fork a project (in his case the kernel). One project should be able to scale from the very simple to the very complex because there’s no such thing as a pure consumer, pure developer, pure enterprise, pure server, etc. Most people are a mix of these.