“In
Part I of our multi-part series of discussions with Canonical CEO and Ubuntu founder Mark Shuttleworth, Shuttleworth covered the delay in the release of Dapper, and something of a history of the open source and free software movements. In Part II, Shuttleworth spoke specifically about Dapper in the Enterprise, and a bit about how Canonical will make money on Ubuntu. In this final installment, we delve into the geographics of Ubuntu’s appeal; the investment by Shuttleworth’s venture Capital fund, HBD in ImpiLinux, and ever so gingerly broach the religious topic of KDE versus Gnome – not just as the Linux desktop, but as Mark Shuttleworth’s Ubuntu desktop.”
Mark Shuttleworth: “Our strategy with Ubuntu is very much to allow local, regional groups to customize it and produce something that is strong locally.” … “Canonical’s role is to create a global ecosystem and to make that ecosystem self-sustaining by keeping the cost of production as low as possible and keeping the results, the product, effectively as general as possible.”
From what i have seen, the tools are all here that allow anyone to build from litle-costumized to highly branded and specialized versions (no, not forks) of ubuntu.
IMHO It is a good corporate strategy: it allows smaller companies to make a buck without having to make huge investments (money and time), and end clients to get customized products and high quality support at very low cost.
(Offtopic: just finished installing ubuntu LTS and even it’s not perfect, it flies!)
By the way: there’s nothing much to say on “the religious topic of KDE versus Gnome”.
Mark runs Gnome on his laptop, KDE on desktop, and XFCE, AFAIK, on a Mini-ITX system.
Edited 2006-06-10 16:59
Ubuntu is not exactly an OS geared for customization at the level you are discussing. In fact, it’s far from it. You need a meta distribution for the level of customization that gives clients flexibility and control. Meta-distributions like Gentoo and to a little extent debian are good for these. Ubuntu should focus on what it is doing best, attracting new users to Linux.
I think i understand your point (and believe you didn’t understand mine).
In theory, companies can simply “brand” their own ubuntu version, and sell services arround it, without the need to ever get into highly technical details about the underlying system.
If you need “real” customization and flexibility, ubuntu probably isn’t the best choice, but if you look at projects like edubuntu, you realise that ubuntu can in fact be a good choce for a base os because of it’s simplicity and ease of use, amongst others.
Some projects might just need a simple, clean, “just-works” base where they can add aditional layers of software and services into.
I think rPath would be a better example than Gentoo when talking about that strategy.
I’m not quite sure how you can say in the same paragraph that Ubuntu is not geared towards customization and say that Debian is. They are in essence the same thing. The only real difference is that Debian’s standard install is very basic and that is the only choice with Debian at the moment. Whereas with Ubuntu you can either install one of their Desktop installs (KDE, Gnome, XFCE, etc.) or you can install off the alternate disk that has the server install (which is basically the same as installing off of a Debian net install, but with more packages on the CD)
The only real difference is that Debian’s standard install is very basic and that is the only choice with Debian at the moment.
Not quite. A default Debian “Desktop” installation will have GNOME and GDM setup and working properly.
Of course, you’ll need at least disc 1.
Actually, the last time I installed Debian, the “Desktop” option in Tasksel literally installed both KDE and Gnome. Debian is great, I love it. But really there isn’t a whole lot of difference between Debian and Ubuntu. With the exception of package versions that are in Ubuntu’s Main may be newer (Packages that are in Universe and Multiverse are most of the time older than what is in Sid and sometimes Etch as well).
I will say that as far as running either as a server I would prefer Debian. Simply because it’s so very well tested before it goes into a stable branch.
I’m not quite sure how you can say in the same paragraph that Ubuntu is not geared towards customization and say that Debian is. They are in essence the same thing.
I’ve seen a review of Ubuntu which suggests that Ubuntu ties certain programs together by changing their dependencies.
http://techrepublic.com.com/5100-10877_11-6066271.html
Although the review/interview is aggressively overcritical, there may still be some truth in its claims. X/K/Ubuntu is developed as a pre-selected set of applications and I wouldn’t be surprised if they changed some dependencies for their packages to ensure that certain programs are always installed together. Of course, Debian packages have also their own dependencies but Debian is not developed or packaged with a similar tightly pre-configured set of applications in mind as is the case with Ubuntu.
I haven’t really studied this supposed hard-coding issue in Ubuntu, so I can’t say for sure if it exists or if such hard-coding can be avoided by using the minimal “server” installation and adding individual applications afterwards. Anyway, if Ubuntu changes dependencies to tie together some programs that aren’t similarly hard-coded in Debian, then Debian obviously would be more geared towards customization than Ubuntu.
Ubuntu doesn’t really need to change dependencies to have certain apps installed together. Metapackages allow just that.
Guadalinex from Spain is a good example of Unbuntu used as the base and Spanish localizations added to it, not only the language but also programs the local group want for their project. Guadalinex 3.0 released in February is based on Breezy released in October. I expect Guadalinex 4 (?) to be released a few months after Dapper. Guadalinex is endorsed and supported by the local government and schools. I am sure businesses and individuals use it also.
It’s “because of its simplicity”, not “because of it’s simplicity”.
http://unUbuntu.com/
Cool! UnUbuntu! Where do I download it? Is it current with Dapper? I’m supposing this is a DE derivative. If so, which DE (or WM) does it use? Is it non-brown? If so, may I suggest midnight blue, since that’s the color opposite brown on the color wheel.
The general idea expressed in this section of the interview is that by drafting a good development model and handing the power over to the community, you end up getting a lot of innovation outside of the core product. In addition, the umbrella Ubuntu Project has been very accomodating about including these extensions into the existing model (the *buntus). Same goes for the handling of localization and translation.
This was the least interesting of the interview sections (part II was my favorite), but my overall impression is very positive. Mark is obviously a very sharp guy who is a “believer” in the truest sense. He thinks that no sacrifices to the ideals of a free, community distribution need to be made in order to create a profitable business around it. Sure, there are distribution projects with lower barriers-to-entry for developers and more flexibility for power users, but none with the legitimate commercial aspirations of Ubuntu.
Of course, Dapper has its bugs and rough spots, but what we see here is a project that has a development model that’s working and a community that’s strong and rapidly growing. For me, Ubuntu is a vastly more successful fork of the Debian development/release model, the clear victor of the large number of Debian-derived distribution projects of the last 2-3 years. Ubuntu is Debian’s long-overdue initial offering into the commercial Linux arena (IMHO held back by the sub-par packaging/licensing models of the RPM-based giants), and an entry that the free software community can be proud of. It’s just a matter of months (two years tops) before Ubuntu is way more than a blip on Red Hat’s and Novell’s radar.