Home opinion-and-analysis Open Sauce New Debian leader aims for better communication

New Debian leader aims for better communication

In many ways the Debian GNU/Linux project is unique among all the distributions that dot the Linux landscape. Apart from putting out a high-grade distribution, it has served as the base for a number of others - Ubuntu and Knoppix, to name just two - which have had a big impact on the growth of the Linux ecosystem.

The project has long, fiery discussions and every little bit of it is on the web for all to see; it has survived the predictions of a great many prophets of doom and naysayers and emerged stronger each time.

The annual elections are also a unique feature. A few days back, the project concluded its elections for the year and Steve McIntyre emerged as the victor in a three-cornered contest. McIntyre may well be the most watched elected official of a non-profit group - the direction the project takes is of vital concern to a great many businesses.

A month short of 34, McIntyre has been a Debian developer for the last 12 years. Raised in the northwest of England, he studied engineering at the University of Cambridge where he was first exposed to Linux

He initially used Slackware and later Debian, gradually spending more and more time on the operating system rather than his studies.

McIntyre is employed by Amino Communications, a company in Cambridgeshire that develops Linux-based set-top boxes. He runs the Debian-CD team that creates the official CDs and DVDs to accompany each release and is also part of the team that organises Debconf, the Debian Project's developer conference, each year.


He was quick off the blocks when approached for an interview; his prose was clear and concise, his reactions swift.

iTWire: Congratulations on being elected leader of one of the world's larger software projects. Has the fact that you now head a team which is much bigger and much more fractious than the average company sunk in?

Steve McIntyre: Thanks! It's slowly dawning on me. I've met a lot of the people in Debian over the years and made many friends; that helps me to not to be too intimidated. I'm expecting to have an interesting time "in charge" of the herd of cats, though.

The turnout for the election was comparatively low this time. I've seen comments indicating that this was because there were no "controversial" candidates in the running. Your comment?

That's a reasonable justification, I think. All of us (me, Marc and Raphael) are quite well-known and well-respected within the project already, and we had broadly similar ideas on things to do. I can understand that that may have led to reduced interest in the election itself. We also reduced the length of the election process this year, which will have further contributed to the reduced numbers.

When Anthony Towns was leader, there was a great deal of coverage (mostly negative) about the project. During Sam Hocevar's time, there has been very little written about the project. Which do you think is better? Can you think of any reason why this is so?

Anthony was one of our most active DPLs in recent years, and he wasn't afraid of trying controversial things (for example Dunc Tank). In a project like Debian with over 1000 strong-minded developers, you can probably expect more than 2000 different opinions on any given topic. Given that, a project leader trying wild new ideas is guaranteed to cause a lot of publicity.

In contrast, Sam has been much quieter in the last year. He has been working much more on issues that are less visible externally. That means that the press coverage has been much more subdued.

As to which I think is better, I'd have to say "it depends". Obviously, I'd love us to get lots and lots of positive exposure this year, if at all possible. But something of a happy medium is most useful in my opinion: if we're too quiet then people start stories saying "Debian is dying because we've not heard anything", but alternatively it's also too easy to go to the other extreme with stories of "Debian is killing itself with internal arguments". Somewhere in the middle is nice... :-)


One of the priorities listed in your platform is the improvement of communications within the project.  What are the areas where communication has been a problem in the past?

We've had problems in the past with communications between some of the core teams. It's all too easy for some of us to get so bogged-down in actually doing a job that we don't spend any time telling other people what we're doing, or (more importantly) how and when they could help us.

I'm a great believer in the benefits of positive communication: the more that people talk about what they've achieved, the problems they've solved and the cool stuff they've developed, the more other
people will be encouraged to join us. That's a philosophy I want to try and share with more people in the project.

There has been some discussion in the community about the need for Debian to attract corporate sponsors - not to pour money in directly but to hire developers so that they will be able to spend more time on Debian. You seem to have some talent in this direction as well so will you be taking up initiatives like this?

To be honest, I'm not very convinced that this is a useful place for the DPL to get involved. Quite a number of DDs are already employed by various Debian-friendly companies around the globe: I know a lot of them at well-known companies like Google, Nokia and HP, and more again working for consulting companies like Credativ and Univention who offer commercial development and support for Debian.

Many of these companies end up pushing back their improvements over time, and of course that's useful to us all. Lots of places are hiring Debian people already; there's not much more I could or would do to push that even further.

Do you support initiatives like that which Anthony Towns came up with - the Dunc tank, or paying some developers to get a release out on time? On the whole, how do yo rate that particular initiative?

I was a part of the original Dunc Tank project along with Anthony, in fact. At the time, it seemed like it was a worthwhile experiment to try. In some respects, it was successful: the added effort available in the Release Team was very useful. However, overall I think we must admit it was something of a failure. Fundamentally, we did not release on the exact schedule we had aimed for. More time was probably spent arguing about it than was gained directly. I would not try a similar idea again myself.


In a few months, it will be time to freeze Lenny if you are going to release by the year-end. Do you see that happening? And do you think it is possible to keep to something like an 18-month schedule for releases?

We've already started freezing some of the archive, in fact. Core essential packages are already stabilised, with changes now needing approval from the Release Team. As we go further into the year, we'll be progressively freezing larger parts of the system: core toolchains and libraries, non-core toolchains and so on up to the full freeze leading to the release itself.

We're still aiming to release some time in the second half of the year, and at this point that's looking entirely possible. It'll be hard work (as always!) and there will be some hard decisions to be made. In my opinion, a key thing that has changed over the last couple of release cycles (Etch and Lenny) is that we have a very strong Release Team who concentrate on release issues for the entire period, trying to ensure that Testing is always in a reasonable and consistent state. We used to let things essentially run wild for long periods and only try and stabilise very late. The very long Woody and Sarge cycles showed that could not work any longer.

At the moment, we're aiming for 18-month release cycles. If we can aim for 18 months and deliver within a few months of that each time, we'll be happy. Etch showed it was possible once, and I believe Lenny will show we can manage it again.

How do you plan to go about making it easier and less intimidating for prospective developers to join?

We're always looking for new developers - they're important to keep the project going and growing. Our existing New Maintainer process is infamous for being demanding, both in terms of the technical challenges and the time and patience required to complete it. However, more and more potential developers are working with us on the debian-mentors list (a great place for training) and getting their packages sponsored into the archive.

We also have the newly-created DM (Debian Maintainer) scheme whereby new people can be granted limited upload rights once they have shown a reasonable level of ability. These two together make a much easier level of entry into Debian than previously; more effort is still underway to streamline the process even further.



How much does it affect the project when there is negative publicity? And how would you as leader deal with it in the event it happened?

Negative publicity always hurts: fundamentally, we're volunteers doing Debian work because it's fun, or because it scratches our itches, or because of the ego boost from being recognised for doing a good job. Bad press can kill that motivation and turn people against each other. My method to deal with it is simple: be truthful and honest. If there are misunderstandings, set the story straight. If there are real problems, acknowledge them and strive to fix them.

Despite all the predictions of gloom and all the naysayers, the Debian project still continues to produce a great distribution. How do developers stay motivated?

That's a *very* big question. :-) Just like opinions, we have *many* motivations and I can't possibly hope to cover all of them for all of the developers. I can speak for myself, though. I love the knowledge that I'm working on a complete operating system which covers my needs. If it breaks, I have a chance of fixing it, or I can find a friend who will be able to do it for me. If I want to do something new with the system, I'm confident that it will support me. The fact that millions of other people appreciate what I'm doing as well is icing on the cake!

And finally, what will be the benchmarks by which you measure the success of your 12 months at the head of the project?

I'll know I've done a good job if Debian is in a better shape: more efficient, more people having fun, more people working on the core tasks that are needed. I want to see Lenny released inside our target window. Beyond that, I'll just have to trust to judgement - my own and those of my peers. I know that's a little wishy-washy, but it's not like I can point to objective numbers like sales figures here... :-)

LEARN HOW TO BE A SUCCESSFUL MVNO

Did you know: 1 in 10 mobile services in Australia use an MVNO, as more consumers are turning away from the big 3 providers?

The Australian mobile landscape is changing, and you can take advantage of it.

Any business can grow its brand (and revenue) by adding mobile services to their product range.

From telcos to supermarkets, see who’s found success and learn how they did it in the free report ‘Rise of the MVNOs’.

This free report shows you how to become a successful MVNO:

· Track recent MVNO market trends
· See who’s found success with mobile
· Find out the secret to how they did it
· Learn how to launch your own MVNO service

DOWNLOAD NOW!

Sam Varghese

website statistics

A professional journalist with decades of experience, Sam for nine years used DOS and then Windows, which led him to start experimenting with GNU/Linux in 1998. Since then he has written widely about the use of both free and open source software, and the people behind the code. His personal blog is titled Irregular Expression.