This year, the election of the Debian Project Leader (DPL) is proving more sedate than in previous years. The change, perhaps, reflects the diminished importance of the distribution, or the efforts over the last few years to increase civility within the project. Or possibly the change is due to the fact that only two candidates are running — Stefano Zacchiroli and incumbent Steve McIntyre, two long time project members who respect one another. But, for whatever reason, this years DPL election is noteworthy mainly for its civility. Zacchiroli and McIntyre share many views, and, on many others, differ mainly in emphasis. The result is that the election will likely be decided mainly on personality and management style rather than issues.
As outlined in the Debian Constitution, the DPL is elected each year. The election begins with a week long nomination period, followed by a three week campaign period and a two week voting period. Only official Debian Developers can vote, and results are tallied by the Project Secretary using a Condorcet method, in which the winner is the candidate who would beat each of the other candidates in a run-off election (although, this year, the results will be the same as a simple majority vote, since there are only two candidates). This year, voting began on March 29, and is scheduled to end on April 11, a second before midnight UTC.
The Candidates: Steve McIntyre
Steve McIntyre, the current DPL, has been a Debian Developer since 1996. According to his election platform, his main contribution to Debian is as a member of the CD Team, "both developing the debian-cd software itself and using that same software to create the official CDs." In addition, McIntyre hosts servers on Debian's autobuilder network for building packages for multiple hardware architectures, and helps to administer the project's Summer of Code program.
McIntyre ran largely on his record in his present term. In his platform, he
emphasized how, in the past year, he organized a review of how various
Debian teams were performing. To try to improve inefficient teams, he has
tried to recruit new members and suggest new work methods. He also
emphasized communication, saying that he has helped resolve disputes within
the project, and represented Debian at trade fairs, as well as to the press
and other free software groups. He listed "communications within the
project. The Debian community tends to discuss everything publicly and at great lengths, but, this year, debian-vote seems subdued compared to previous years." as his first goal in a new term, followed by finding ways to improve processes within the project, such as the one by which new maintainers join the project.
However, McIntyre did not specify how he would approach these goals. Instead, he simply said, "I want to see improvements made where they are clearly needed. There are cases where communications are lacking, or people and teams might not be working as effectively as they want to or need to. In those cases, I want to help and encourage the people involved to make those improvements. If that is not enough, then we may need stronger actions."
Rather than focusing on details, McIntyre urged people to vote for him on the basis of his experience, his organizational work, and his personality. "I'm a programmer, which means I have strong opinions on many subjects (*grin*). Despite that, I believe I am honest, generally approachable and easy to work with. I am a good communicator and negotiator, and I have made many friends in the Debian and wider Free Software Community over the years."
The Candidates: Stefano Zacchiroli
Stefano Zacchiroli, better known as "Zack," has been a Debian developer since 2001. He maintains a number of packages related to OCaml and Vim as well as "tools for mathematics in XML, [and] Python libraries for dealing with XML," according to his platform.
In contrast to McIntyre's statement, Zacchiroli's statement was much more organized and concrete. Even so, it is at least twice as long, listing key points in bullet lists and listing specific actions he wants to take. These key point range from a promise to be "transparent and present," responsible for setting the project agenda, and active in communicating his activities to the rest of the project to encouraging consensus and the control of "vocal minorities" in discussions, and finding new ways in which new developers and non-programmers can become more active more quickly. He also promised to encourage face-to-face meetings, to see that core teams within the project consist of at least three members, and to encourage members of inactive or poorly organized teams to find a replacement and "something to work on that is more fun for them."
While Zacchiroli's platform included such concrete ideas as methods for continuing to improve the levels of politeness on mailing lists and redesigning the web site, a large part of it centered on the role of the DPL. In Zacchiroli's view, the DPL is responsible for conflict resolution within the project. He explained that, while he will not be a "'post-in-every-thread' DPL" that he "will try to be present in 'hot' discussions . . . which concern the organization and the big picture of the project. I will also encourage seeking the DPL opinion on specific topics by pinging me explicitly. The DPL['s] opinion can also be a reasonable first attempt to solve a conflict; if it fails, we do have other mechanisms to settle it." He suggested that these proposals will be aided by his personality: "I am thoughtful, listen to others, and open to be convinced by good arguments."
For those who want details about the campaigning, Zacchiroli provides a convenient summary of most of the questions raised on the debian-vote mailing list during the campaign period.
However, for those hoping to differentiate the candidates by their stands on the issues, debian-vote offers little help. McIntyre and Zacchiroli are in broad agreement on many — even most — subjects, including the use of non-sexist language in standard project documents. Both agree that collaborative package management is desirable, particularly as a way to help new maintainers gain experience, and that important packages should usually have core teams to maintain them. Similarly, the two candidates are in broad agreement that Debian's reserve funds of some $125,000 could be greatly reduced, although McIntyre suggests that they be halved, and Zacchiroli suggests that they could be reduced to $50,000.
One small difference that emerges on debian-vote is the role of the DPL in discussions of policies and actions: Zacchiroli sees the DPL as the leader of discussions who sets the agenda, while McIntyre favors a less direct approach of encouraging sensible people to voice their opinions whenever possible. However, this is a difference in approach, not a dispute over whether anything needs to be done.
Even McIntyre's announcement of Luk Claes, another long-time Debian Developer, as a running mate has not been much of issue. The idea that the DPL needs trusted assistants is not a new one; for instance, in 2005, Branden Robinson and Andreas Schuldei ran, promising that if either was elected, he would be assisted by "Project SCUD" — a group of half a dozen long-time Debian contributors. However, while Project SCUD was controversial in its day, nobody has publicly expressed a comparable level of concern that McIntyre proposed a kind of partnership that the Debian Constitution does not explicitly provide for, explaining that he and Claes are "planning on splitting the workload between us, depending on what fits and who has the time at any given point. My plan is for both of us to receive leader@mail."
Most likely, the proposal is uncontroversial because Zacchiroli's position is not far from McIntyre's. Although Zacchiroli said in his platform that he does not believe in a "DPL board" because "the DPL term is too short to spend time with extra coordination hops," he immediately moved closer to McIntyre's position by adding that he "will be looking for a 2IC (second in charge, vice-leader)." This admission led McIntyre to add a rebuttal to his own platform: "That I can understand completely, but I'd be personally much happier if he was to name his proposed assistant before the election. I did that deliberately with Luk, as I want people to be able to vote for a known quantity rather than somebody unnamed."
The largest difference between the candidates is in their answers to the question of what actions will improve Debian the most. In reply to the question, Zacchiroli wrote, "When I joined in 2001, Debian was The Distribution that a lot of users were using and all my friends knowing Free Software were dreaming of contributing to. Things have changed since then: newbies now use Ubuntu or Fedora, and contributors can easily join their communities. Debian is too often seen as the old distro that some old timers still use, having a process to join which is not worth trying. The Debian value that needs to be improved the most is changing that: putting Debian back into its place." Zacchiroli maintains that, within a year, Debian can set an example for the rest of the free software community, particularly in contributing changes to upstream projects.
In comparison, McIntyre replied
to the question by writing about improving communication both inside and
outside the project. "We need to encourage more people to experiment
with new ideas, rather than simply live with the status quo all the
time. Larger projects have inertia, and we have to acknowledge that: either
push things more to overcome it, or work around it for the new ideas until
they're ready for wider adoptions." He added that "the most
important thing we need to do is to work out exactly what we want the
Debian project to be, in terms of leadership. Making it clearer and easier
for people to gain recognition for whatever work they're doing on or with
Debian is far and away the best way to encourage more people to get
involved." Noticeably, though, the solution for both candidates to the main problem that they see is to improve coordination within and without the project, the only difference being the reasons they want to do so.
Rather than attack each other's policies, both McIntyre and Zacchiroli tried to point out the weakness in each other's statements. For instance, in the rebuttal added to the end of his platform statement, Zacchiroli attempts to characterize Claes and McIntyre's candidacy as vague, claiming that "their platform lacks hints and strategies that explain how they are going to attack those problems . . . . Also, I'm missing the vision of the two candidates on relevant topics such as Debian membership, mailing list discussion quality, [and] derivatives [distributions based on Debian]."
For his part, McIntyre attempted in his own rebuttal to counter Zacchiroli's bullet lists by saying, "I'm not sure he will have the time to do what he's planning: time always runs away more quickly than you expect" — a comment that serves to remind voters of his own experience as DPL.
After you read McIntyre's and Zacchiroli's comments, you soon realize that there is no major difference of opinion between them. When you find an apparent one, on close examination, it tends to be a matter of emphasis at a particular time more than anything approaching a major disagreement. In fact, one candidate's answer in a particular context often echoes the other's answers in another. Both show an awareness of issues and maturity that could make an effective DPL.
The main question about this years' election, apart from who wins, is whether such reasonableness will affect voter turnout. Although many decry the acrimony of previous elections, it may have encouraged people to discuss issues and vote. This years' reasonableness seems to have reduced discussion on debian-vote (always taking into account the Debian community's tendency to discuss everything publicly and at great length) and the possibility seems strong that the number of votes cast could be similarly reduced. We'll know some time in the early hours of April 12 when the results are announced.
to post comments)