As a happy Ubuntu user, I was pleased last week when I received a note announcing the Ubuntu Foundation. The Ubuntu Foundation was created to fund development of Ubuntu and to separate Ubuntu development from Canonical, which supports Ubuntu and open source software. Mark Shuttleworth has put $10 million into the Foundation, to help it get on its feet. The Ubuntu Foundation has also announced that the 6.04 version -- due in April 2006 -- will be supported for three years on the desktop and five years on the server. This kind of predictability and long-term roadmap is exactly what's needed.
One thing troubled me, however, while reading the announcement. There was no mention whatsoever of Debian, which provided the basis for Ubuntu Linux. I've been hoping for some time that there would be increased cooperation between Debian and Ubuntu, rather than less, and I'm not the only one.
I had a few other questions about the Foundation, so I e-mailed Benjamin Mako Hill, a full-time employee of Canonical and a Debian Developer, about the Ubuntu Foundation and its relationship with Debian.
The first question I asked is why the long wait? Breezy Badger is due out in a few months, so why make people wait until April 2006? Hill said that the reason is because Badger is already underway. "For something supported for so long, we want a full development cycle to make sure it's rock solid so we're doing Breezy+1."
I was also curious about some of the details of the Foundation, how many people would be employed and how it would be funded after the intial funds run out. The exact number of developers to be employed by the foundation is not yet known, though Hill said that there are 15 people employed by Canonical and "At the very least, half of those hackers will go over." What happens when the intial funding runs out? According to Hill:
With investments and such, this should be a sustainable amount and will not actively work on getting other funding. That said, there will be donations and there may be later influx of cash into the foundation from Mark or otherwise.
Finally, I asked why Debian wasn't mentioned in the announcement, and if there were plans to coordinate more actively with the Debian project to make the two distributions more compatible. Here's what Hill had to say:
We're absolutely committed to working with Debian. I'm a Debian developer and an active one at that. :) Along those lines, I have organized a round table on derived distribution at Debconf next week. Ian Murdock's suggestions around binary compatibility may not be embraced. Ian seems to overlook the fact that with three different distributions, Debian is not even binary compatible with *itself.* We will pursue interoperability with Debian and with other distributions on the source level and we'll pursue on the binary level with LSB and similar initiatives. This accomplishes all of Murdock's goals -- although not his exact prescription. :)
Murdock's comments (at least some of them) on Ubuntu and Debian can be found on his weblog here and here. I've also touched base with Murdock on the Ubuntu Foundation, and will cover his comments Tuesday.
There are a few things that I think we can all agree on. Debian is an excellent distribution, and an excellent foundation for an enterprise Linux distro. There is a lot of interest in building on the Debian foundation, for a number of reasons -- Debian's excellent administration tools (APT, Synaptic), its humongous collection of packages, the social contract that guarantees that Debian will remain free and focused on users (and uncontrolled) and its general stability.
In fact, as open source and Linux become more and more accepted in the mainstream, I think there will be a shift away from a vendor-controlled Linux distros like Red Hat and SUSE and towards a single standard. Debian seems ideally-suited to provide that standard, whether one wishes to use standard Debian, or another flavor of Linux based on Debian.