The Debian GNU/Linux operating system continues to generate interest from developers around the world, keen to sign up and contribute code to the open-source project now in its 15th year.
But this popularity has been a mixed blessing. The project came under fire recently when programmers who wanted to get on board were unable to sign up and become registered participants.
Some analysts foresee a less than rosy future for projects such as Debian, claiming free coding is all well and good, but that without a solid financial backing — such as the models adopted by Red Hat and to a greater degree Novell/Suse — Debian will ultimately hit a brick wall.
Elected Debian project leader on 17 April, Steve McIntyre is the man charged with leading the organisation to the next level. Despite being faced with a backdrop of organisational challenges, while at the same time working away fervently on his own Debian bug fixes and development projects, McIntyre found some time to talk ZDNet.co.uk through his vision of where the Debian project is headed.
Q: Why did you take on the role of Debian project leader, and what do you hope to bring to the development of the operating system?
A: I've got a few ideas about places where we can improve things, as I laid out in my election platform. The main issue I want to work on is communication, both within the project and externally. We're doing some really excellent work, but it's often not visible. I want to encourage all our developers and core teams to talk more about what they're doing. In my opinion, the best way to recruit more developers and users is to show off the cool stuff we're doing.
Debian author Ian Murdoch was also a founding director of the Open Source Initiative. Why do you think he saw the need for a new flavour of Linux with the particular look, feel and structure of Debian?
Ian started the project back in 1993 to make a new, openly developed, Linux-based distribution. He was very much inspired by the ideals of the GNU Project and the Linux developers at the time. He believed there was a place for a new OS developed in the same manner, designed by dedicated volunteers for themselves and everybody else to use. In many ways Debian has changed and evolved hugely over the years, but that core spirit is still the same.
ZDNet.co.uk recently reported on the "administrative" delays you've been experiencing surrounding community registration. What do you intend to do to tackle this issue?
I'm actually in the middle of a review of all our core teams right now, as I promised during the project leader election. There are potentially several places in Debian's core teams where we could be more active and we definitely need to be more communicative about what's going on. The furore on Planet Debian is just indicative that this review is well overdue, I believe.
I'm expecting that there'll be more news about how we're shaping up and what we need to change and improve in the next few weeks. I can't promise it will be especially interesting to anybody outside the project, though.
Debian's software-development methodology hinges around the provision of precompiled software "packages" that effectively form reusable components. How do you vet newly proposed packages, and what type of components would you like to see developed in the future?
New packages are typically proposed on our main development mailing list so other developers can comment. Depending on the package in question, that process can lead to some extended discussion. Once a developer believes he has something ready for inclusion in Debian and uploads it for the first time, a thorough review is carried out by the "ftpmaster" team. The team checks for potential licensing problems, security considerations and packaging quality. If the new software meets those checks, it's allowed in.
Personally, I'd like to see more and more end-user applications developed and improved in the future. We already have...
...one of the most stable systems available, but more and more exciting applications are being created as more users and developers turn to free software on the desktop.
If we look at the two leading Linux distributors that are out there, Red Hat and Suse occupy the top slots and have received considerable financial backing from vendors such as IBM, HP, Dell and others to cement those positions. Is Debian looking to become more commercially focused, and will this help to overcome some of the development inertia that appears to be afflicting the project?
Firstly, I disagree with your suggestion that we have development inertia. We have more and developers wanting to join Debian and help us work on our ever-improving operating system all the time.
On the main question, many of our competitors in the Linux world might be corporations with enterprise-level friends, but I strongly believe there is a place for a not-for-profit group like Debian. As well as packaging software that other people develop, Debian developers are often also upstream authors and collaborators on much of the common infrastructure that all the Linux distributors benefit from. We're not in this just to make money, but because we're passionate about making the best free operating system we can.
We also have many friends among the vendors. For a recent example, HP has described how they have made significant amounts of money from selling hardware with support for Debian. Others including Sun, AMD, Intel and IBM work with us to provide support or hardware. Large companies also regularly sponsor our annual development conference, Debconf. This year we're heading to Argentina for the conference and HP and Nokia are the two biggest supporters so far.
Commentators have said that Debian needs to look at its workflow processes and approach the project more like a commercial business organisation if it is to flourish. Would you agree with this approach?
Frankly, no. Debian is a volunteer-run organisation. Much of our workflow is driven by people working on the things that interest them. We have a large community of contributors with all kinds of backgrounds and skills. Trying to impose a more formal business process would cause many of those people to leave, possibly killing the project.
Do you still receive funding from not-for-profit umbrella organisation Software in the Public Interest (SPI)? And what path towards a more approach do you see Debian taking in the future?
SPI is not responsible for funding Debian; rather, it's an organisation that was created specifically to hold funds and act as a legal backend for Debian. It has now evolved into a more general umbrella group that provides similar functions for other projects as well: for example, PostgreSQL, OFTC, Freedesktop.org. We're not looking for direct commercial fortune for Debian — that's not what we're about.
Will Debian always suffer from existing at the hobbyist programmer level and its inherent proximity to the archetypal non-business-minded software engineer mentality?
We may suffer a little in terms of external perceptions of us, but we're proud of our roots and what we achieve. Many of our developers are happy to be able to work on a high-quality project where doing things "right" is important. That's all too often something that's lacking in more business-minded organisations.
Debian will be 15 years old this August. Where would you like to see the project in another decade and a half?
That's a simple question! Bigger and better is my own hope: more developers, more software, more users, better quality and better features.