Debian Project Leader election—2017 edition

By Jake Edge
March 15, 2017

After last year's anticlimactic Debian Project Leader (DPL) election, where the one and only candidate resoundingly beat "none of the above", this year's edition will be a contest, at least. Two candidates have stepped forward: current DPL Mehdi Dogguy and Chris Lamb. They both have outlined their platforms and the campaigning period has opened. Voting will take place in the first half of April, with results available on April 16.

While it does seem that DPL elections are seeing fewer candidates these days, it is rare to see a one-candidate election. It only ever happened once before, in 2011, when then-DPL Stefano Zacchiroli ran for re-election unopposed. Most other DPL elections have had three or more candidates. The decline in the number of candidates has led at least one Debian developer to wonder whether the project should consider eliminating the DPL position entirely.

Mehdi Dogguy

Dogguy is the incumbent this year and his platform reflects some of the experience he gained during his tenure as DPL. But some of the problems and areas to work on that he called out in 2016 still persist, so some parts of the platform have been recycled from last year's, which he links to. In particular, he points to the difficulty of recruiting more contributors. In both platforms, he notes a lack of documentation as a key contributing factor to the recruitment woes. It will require a multi-year effort to fix that, but he had some success in attracting students' attention to Debian at an event he participated in while DPL. Events like that, along with internship programs like Outreachy, will be helpful to the recruitment efforts and he intends to continue those kinds of activities if re-elected.

In his talk at DebConf16 (and in a related BoF), he has proposed the creation of a roadmap for the distribution to help guide its development:

Our distribution is the main product delivered by our project to the world. I tend to believe that we don't only package upstream projects and publish new releases. Debian offers more than that and has its added value. Release Goals were one way to show how our distribution is original, relevant and innovating. Even on a social level, it is important to set some goals in order to continue to motivate prospective contributors into joining Debian. We, as the Debian community, should work to publish and maintain a roadmap, and strive to implement it each cycle. It is not necessary to have it done in time for a release, but it is more important to follow its progress.

He intends to spearhead the roadmap effort once the "Stretch" (Debian 9) release is out the door. The plan is to establish SMART criteria to help track the progress of the items on the roadmap.

Dogguy would also like to see the project's fundraising put on a more stable footing by adding a "partners program" that would make it easier to budget funds over longer time frames. These partners would commit to a certain level of support over a few years. In addition, he hopes that the funding provided by the partner companies and organizations will help reduce the fundraising burden for DebConf organizers so they can focus more on the event itself.

He would also like to push forward on the "Bikesheds" project, which has been percolating in Debian circles for some time. It is a way for developers to create their own repositories similar to the Ubuntu personal package archives (PPAs). The name evidently comes from a DebConf15 announcement—complaints have been heard about it—but the idea goes back to a 2013 posting to the debian-devel mailing list.

The project is not well understood within the Debian community, Dogguy said, so he would like to gather the right people for a sprint to fully describe what it is and what is needed to make it happen.

If this event is successful and provides us with answers and a specific plan, I will strongly encourage and potentially organize more sprints to make this move forward. If it requires resources we do not have in our project, we will make a specific call for help or propose an alternative but more realistic plan.

There is more to his platform, of course, but that brushes on the high points. His priorities for his next term if he is elected are: "ensuring the community remains safe and fun", the roadmap, clarifying Bikesheds, and the partners program. In many ways, he is planning to "stay the course" that he set in his current term, though there are some new initiatives he would like to help drive.

Chris Lamb

Lamb's platform begins with a bit of a lament that Debian is not more prominent among distributions: "Although the open source environment is not a zero-sum game, I cannot help feel a pang of regret whenever I hear initiatives — especially headline-grabbing ones — that are based on our derivatives rather than us." He worries that the distribution may be relegated to being a glue layer for containers or Internet of Things devices, which will make it harder to attract contributors.

That said, he believes that "Debian is working incredibly well", but that its image is not what it should be. Some amount of work on marketing the distribution is needed:

Our unflashy image — best encapsulated in our unengaging website — speaks to what P. G. Wodehouse might refer to as a lack of "snap and vim". My short experience in the startup community has taught me that polish and pizzazz are essential parts of any project, be they for-profit or not. We are doing ourselves, our users and potential future developers a disservice by neglecting (or deliberately avoiding) the most basic of marketing.

Like Dogguy, Lamb is concerned that the documentation for newcomers and potential contributors is lacking. In addition, some of the processes that contributors need to understand and use are seen, negatively, as hurdles and obstacles. It is something of a tragedy, he said, since Debian is at the forefront of various important initiatives such as the reproducible builds effort (which Lamb is active in). But without raising the visibility of Debian somewhat, he fears the distribution will be left behind to some extent. "We do not want to become — or even perceived to be — just a high-quality package repository."

Lamb sets out four concrete steps that he would take as DPL. He would like to organize more in-person gatherings and meetings, such as DebConfs, hackathons, summits, sprints, and bug-squashing parties. These meetings have effects that go well beyond whatever work gets done at the time because "they act as a social lubricant for subsequent online communications that can go awry". He would like to encourage and help organize more such events and to help grow the number and diversity of people who attend them.

The "onboarding" process for new users and developers needs improvement and he would like to make that one of his focuses. Usability testing could be one way to identify where the problems are in that process, as well as identifying areas where new users are stymied in their efforts to actually try Debian out. Creating a Debian-specific outreach initiative along the lines of Outreachy is also something he would like to do. Outreachy has been "incredibly successful", and a similar project within Debian would allow "more flexibility in the manner of contributions as well as underline Debian's dedication to 'universality' in all its forms with the public at large".

His final concrete goal would be to remove blockers to working efficiently on Debian. He would like to use some of the project's "surfeit of funds" to provide better hardware where it is needed, but also to change the culture somewhat. If projects could do their jobs better with some additional resource, he wants them to have "no hesitation in asking for it". He would like to find creative solutions for any social or personality blockers as well.

No DPL?

Before Lamb or Dogguy had nominated themselves, Guillem Jover posted an idea to the debian-project mailing list: not electing a DPL at all. The DPL role is largely symbolic in many ways, so perhaps the project could do without it:

The DPL role is overstated and magnified, primarily from the outside, where people seem to believe it's some kind [of] messianic figure that leads the collective Debian souls that would otherwise be lost, by giving technical and moral direction to the project at large. But that's very far from reality.

The truth is that even though the constitution grants _some_ powers to the DPL, they are in general not used, because IMO the project would not see those actions with good eyes. The reality is that the DPL is pretty much a bureaucrat and an ambassador of the project trying to represent its collective views towards the outside world.

Others in the short thread did not see things quite that way. Adam Borowski called the DPL position "a largely thankless, massive amount of work that's contrary to what Debian is about (code hacking and solving technical problems[1])"—though his footnote also added "flamewars". The title is a perk and removing it might make the job even less appealing. Others more or less agreed.

Referendum

The election is effectively a referendum on Dogguy's term as DPL. There are no huge differences in the approach or ideas that Dogguy and Lamb have laid out, so it may come down to a question of whether the incumbent has done a good (or even reasonable) job over the last year. There don't seem to be any external indications that Debian is unhappy with Dogguy's service, so it would not be a surprise to see him re-elected. In addition, Lamb has not run for the position before and first-time candidates have typically not been elected. But we'll have to wait and see who gets announced as the DPL in mid-April.