OLinux: First of all, tell us about your background.
Ben Collins: I am generally speaking a programmer and systems administrator. In the past I have also worked as a Desktop Publisher and a web designer. I've worked for NASA LaRC, several ISPs and currently am working at Winstar .
OLinux: Please give a brief summary of Debian's History, Philosophy and Organization on handling free software development?
Ben Collins: Our philosophy goes back a long way. Mainly we believe that it is possible to create a completely free operating system with all of the things you need to do your daily work. That's what started Debian, and prompted Ian Murdock to write the Debian Manifesto. From there began our project, and from it has come the Debian Free Software Guidelines (DFSG), which defines the type of software licensing we consider to be Free in the sense of Freedom. Also came the Debian Social Contract, which defines what we will support for our users. Later, as we grew, came our Constitution, which defines our operating procedures, and breakdown of authority within the project.
We've basically given full control of each package to the maintainer of that package, so long as it falls within the guidelines of our well defined Policy. Our Policy is one of the strengths of the Debian distribution. Without it, we would not have a cohesive set of packages, and installs/upgrades would be a nightmare.
OLinux: How excited you are about being in front of the Debian Project? Do you have something in mind for the Debian Project? Are you going to make changes on the way the work is done?
Ben Collins: I'm extremely excited. This is my third run at the DPL position, and it is a goal I have finally achieved thanks completely to those in the project that have faith in my ability to handle the job. I have plans to clean up some loose ends that have been plaguing our internal organisation for some time. After this, I plan to tackle some of the more difficult situations that still linger, or are threatening to be a problem in the near future.
OLinux: What are going to be the differences between your leadership and the predecessor's?
Ben Collins: When I first came to Debian, Ian Jackson was finishing his term as DPL, and he was very inactive (to his defense, I do not know any details of his situation). Wichert then followed for two terms. I believe he did an excellent job keeping Debian going. However, my plans are to get Debian moving rather than continue to limp along with some of the problems we face.
OLinux: How are people organized and what are the tools used to control the results of the work being done in different projects and parts of the world?
Ben Collins: Within Debian, we have the maintainers (some 800 it seems now). Each of them is responsible for maintaining one or more packages (some do not maintain packages, but help with other projects internally, such as ftp archive, www site, etc.). They have complete control of their tasks within the guidelines and policy. Within this, some developers have grouped together to manage large specific tasks. Examples of this are the Debian Junior project, as well as the ports (such as sparc, arm, alapha, powerpc, etc.) and language projects.
All work is coordinated via mailing lists. Some people also use IRC as a way of immediate interaction (via irc.openprojects.net). We also have the Debian Bug Tracking System to manage bug reports for all of our packages and systems. This system is available publicly via our web pages. Anyone can file a bug, and track it's progress directly with the maintainer.
OLinux: How many people are working for Debian nowadays? Are you satisfied with the results?
Ben Collins: Last I checked, about 800. I am satisfied with the results. What I am not satisfied with is the influx of maintainers without a better scheme to manage them. Work is being done, but I want to see some other things in this area discussed and looked at.
OLinux: What do you think about people saying that the Debian 2.2 has too much bugs? What are you going to do in "Woody" to change this point of view?
Ben Collins: I was not aware that people said that. We have an excellent security team that fixes all known security related bugs. We also make regular point releases (2.2r3 is being worked on as I write this) to update the security patches into a new release. For woody we have a new "testing" mechanism which should help reduce the amount of time needed to release. Hopefully this will make more frequent releases possible.
OLinux: What are your expectations about the "Woody" launch?
Ben Collins: I look forward to a lot of the things that are going to be available in woody. Woody also promises to be the most architectures we have ever released at one time (by any distribution, that I am aware of).
OLinux: What are the active projects at Debian? How are they divided and coordinated in terms of content and staff for each project?
Ben Collins: Usually a project within Debian creates itself to fill a need. The project manages itself, and delegates within its own ranks who is responsible for what tasks. I'm not aware of all such projects, simply because most of them work in the background, silently making Debian better.
OLinux: Here, in Brazil, there is a project called Debian BR . This is a project that is translating the Debian content to Portuguese. Do you know that? If yes, what do you think about it? If not, you are invited to visit the Debian BR web site at debian-br.sourceforge.net. Do you know other projects like this in other countries?
Ben Collins: I had not heard of it before. I think it is an excellent thing, much like the JP and similar projects. The more people we can get Debian to, the better. I'll have a look at the web site, and I wish the best of luck to the project for it's efforts.
OLinux: Do you consider Debian the leading GNU/Linux distribution in the world?
Ben Collins: On many basis, yes. However, I measure Debian on what's important to me, and am well aware that it lacks in areas that are important to others. A recurring topic is our installer. I'm happy to report that a new modular installer is being worked on, and it so far appears to exceed, or will exceed, all of the goals that the group set for itself. It will probably not be done in time for woody, though.
OLinux: How is Debian's relationship with the GNOME Foundation? And with the KDE league?
Ben Collins: I'm not able to answer this question. I do know that we have some developers that work closely with both projects, and that GNOME and KDE are both fully integrated within our distribution.
OLinux: What are the advantages and what differentiates Debian from other popular distributions as SuSE or Red Hat, besides being a non-commercial distribution?
Ben Collins: I think we have three major strengths. One is our development model. No other distribution has all of its developers available first hand to take bug reports and suggestions from its user base.
No other distribution has as extensive a set of policies that allows it to distribute as many packages as we do, all integrated into our distribution, with easy installation.
No other distribution offers the ease of upgrades that we do. There have been reports of people being able to effortlessly upgrade from as far back as Debian 1.3 (bo) to the current stable 2.2 (potato) (note, this is a libc5 to libc6 upgrade path). Debian not only supports, but guarantees upgradability. It is one of our primary goals.
OLinux: How do you describe Debian Project achievements and what are the prospects and goals for the next years?
Ben Collins: The fact that Debian is still around, and is still growing is a major achievement. We have not lost site of our primary goals; to produce a free and stable distribution. Over the next few years I hope to see Debian prosper from commercial acceptance via companies like Progeny. I'm hoping that vendors will see us as a more viable solution for desktops and pre-installed systems.
OLinux: Give us some predictions about the growth of the GNU/Linux operating system for the next 2, 5 and 10 years.
Ben Collins: That's hard to predict. Unfortunately, as free as it may be, GNU/Linux is directly affected by the economy. The current trend of Internet companies starting to fail, will likely scare away of a lot of the venture capital that has flooded Linux in the past years. Hopefully this will be a good thing, and the Linux companies will have to start working to make their money, and not ride the wave of hype. I would guess that over the next 2 years, Linux's hype will settle down, and people will start taking it more seriously (not just those in-the-know).
In 5 years, I suspect that GNU/Linux will be as common as MacOS, Solaris and Windows in the home. In 10 years, who knows. That's like an eternity to the technical world, so Linux may be obsolete by then.
OLinux: What are the improvements that GNU/Linux needs to be more deployed in by the corporate market?
Ben Collins: An accepted, easy to use interface. KDE and GNOME are working toward this with great strides. But even with a good interface, getting accepted and being "common" take far longer than a development cycle.
OLinux: Debian is definitely the best Linux distro, but its hardware configuration interface and its installer are not so friendly. Is the Debian Project going to focus on a best interaction with the final user or it still a distribution for the systems administrators only?
Ben Collins: Yes, the debian-installer group is working very hard on this. We do not want to remain a niche distribution only used by administrators and hard-core hackers.