|
A senior Debian developer, Australian Anthony Towns, has left some core teams of the project and gone quiet in the last couple of weeks, according to project sources.
Towns made an oblique reference to his decision on April 18 with an entry in his personal blog which read, in part: "One of the freedoms I value is the freedom to choose what you spend your time on and who you spend it with. And while I’ve spent a lot of time arguing that people in key roles in Debian still have those freedoms (hey, 2.1(1), don’t you know), reality these days seems to be otherwise."
The reference to 2.1(1) was linked to a section of the Debian constitution that deals with decision-making bodies and individuals; a second link in his post was to the announcement by last year's project leader, Sam Hocevar, that he had appointed fellow developer Joerg Jaspert as a member of the FTP master team and as a full Debian account manager.
Hocevar and Jaspert were two of those who were prominent in their opposition to an experiment which Towns launched during his term as Debian leader in 2006-07 - a project called Dunc-Tank which paid some developers so they could spend more time on Debian and try and meet the release schedule for Etch.
Jaspert also referred to Towns' move in his blog: "...lets (sic) ...write a little about what happened since I gained that extra group: Aj (meaning Towns) resigned from anything that requires additional group privileges."
I wrote to Towns thrice asking him about the situation, twice before I had verified that he had quit the core teams and once after that, informing him that I would be writing this story and asking for his take. Initially, I was under the impression that he had resigned from the project altogether; he answered this bit of my query, saying; "I haven't done that, no."
He did not respond to queries about his quitting core teams. I also asked whether Jaspert's elevation had anything to do with his decision but he did not respond to that either.
One has to go back a few years to trace the events that appear to have led to Towns' decision. In 2005, the Debian stable release Sarge had come out in July after nearly 36 months of development. It had resulted in a lot of bad publicity for the project and everyone seemed determined to release Etch, the next stable version, by December 2006. The rough schedule which was aimed for was 18 months, a reasonable period by the standards of any project which is as large as Debian and which is so committed to quality releases.
In August 2006, Towns, who was then project leader, started a project called Dunc-Tank, to pay some Debian developers to put in more work on the project in the hope that the release schedule would be kept. Release managers Andreas Barth and Steve Langasek were two who were paid from this mini-project.
The media release about Dunc-Tank said, in part: "The experiment's initial goal is to be able to raise enough funds to pay both release managers enough to work exclusively on the release of etch for a month each, having Steve Langasek available full-time during October and Andreas Barth available full-time during November, with the release expected to follow soon after in the first week of December."
But this did not find favour with many other developers. They began to slow down their work on Debian so that Etch would miss the scheduled release date. Barth made mention of this in his blog, saying: "On the other hand, there was a large disadvantage of the whole experiment: Some people who used to do good work reduced their involvement drastically. There was nothing I could do about, (sic) and that happened way before I started full-time on release, but on the global picture that still counts. So, as a first summary, I am happy with my own involvement, but that doesn't necessarily apply to the full experiment."
Some other developers who opposed the Dunc-Tank idea even went to the extent of demanding that Towns step down. When it came to a vote, however, the Dunc-Tank idea won approval in October that year.
Hocevar was open in his opposition to the Dunc-Tank idea. On a page titled "welcome to the Dunc-Bank" Hocevar wrote: "The Dunc-Bank is an experiment to see how aggressive bug reporting can delay the release of Debian Etch."
The vote to continue the Dunc-Tank experiment did not end the debate. Jaspert, in a posting on October 18, raised a number of questions about the experiment and asked for further discussion. "It has demotivated a lot of people who now either resigned, simply stopped doing (parts of their) Debian work or are doing a lot less than they did before DT was started. The freeze got delayed and getting the release out on schedule has become nearly impossible. We are unable to see any good virtue in this 'experiment'," was his conclusion.
Etch was finally released on April 8, 2007, after 21 months of development.
The animosity which arose as a result of differences over the Dunc-Tank experiment appears to linger. Hocevar's announcement of Jaspert's appointment was made on his last day as project leader for 2007-08. The delegations he made are "effective until revoked by the DPL or by a resolution."
According to Jaspert, the current Debian leader, Steve McIntyre, has expressed support for his (Jaspert's) appointment.
The day after Hocevar's announement, Towns made the blog entry referred to earlier.
A couple of days later fellow Debian developer Russell Coker made reference to Towns' move without naming him. Coker discussed it in a post he titled "Planets and resignations", a general discussion about what should, and should not, appear on a Planet site, which is an aggregation of blogs from people involved in some similar activity or organisation.
Towns is listed as a member of the technical committee and the Debian maintainer keyring team as of May 6.
|
