Regarding unresponsive Debian maintainers (was: Re: Open-Source environments for Java)

From: Rogério Brito (rbrito_at_ime.usp.br)
Date: 05/23/05

  • Next message: Jacob Friis Larsen: "Re: dnscache: readproctitle service errors"
    Date: Mon, 23 May 2005 00:55:30 -0300
    To: debian-user@lists.debian.org
    
    

    On May 22 2005, Roberto C. Sanchez wrote:
    > Rogério Brito wrote:
    > > In my very humble and uninformed opinion, some maintainers should
    > > really give up maintaining their packages or should try to get other
    > > people as co-maintainers, if they lack the time to fix their
    > > packages. :-(
    > >
    > > If they applied to the project, then, they committed themselves to some
    > > responsibilities and even though the project is mostly driven by
    > > volunteers, if they can't keep up with their committments, then they
    > > should politely be asked to give space to other people. And I can tell
    > > you that there are other people interested in *working* for the
    > > project.
    >
    > By my count, there are already many maintainers that have done so with
    > their packages:
    >
    > http://www.nl.debian.org/devel/wnpp/rfa_bypackage
    > http://www.nl.debian.org/devel/wnpp/orphaned
    > http://www.nl.debian.org/devel/wnpp/being_adopted
    >
    > I understand that not all are represented however.

    *That* was my point. The fact that a given package is being orphaned or
    adopted actually seems to indicate that some attention is being given to
    it: the maintainer recognizes that he has no time or interest in the
    package anymore and won't be stalling the project (or, at least, the users
    of such packages).

    And the fact that a package is adopted shows that it is important to, at
    least, someone (the person adopting the package, of course -- but probably
    many other users also).

    > I have seen cases of maintainers that have packages with may bug reports,
    > without any indication of an intent to fix them.

    Exactly. It is a frustration to have a bug filed for, say, almost one year
    (Lack of time during one year? Better drop the position of being a Debian
    developer, then).

    I have filed some bugs (sometimes *with* patches included) and not received
    any response from the maintainer. Not only me, but many other users. Want
    an example? See the maintainer of the vrms package.

    I merged some bugs there. I sent the maintainer a simplified patch (after
    another user had already sent a better patch, but possibly more
    "intrusive"). I'm still waiting for a response. And the package could
    easily be fixed (say, a hot fix) and then refactored to include a proper
    fix and some quite useful features (like listing contrib packages).

    I think that Martin-Éric Racine has expressed this quite well in his
    message on the BTS: <http://bugs.debian.org/302504>.

    And this is not the only case. Want to see other examples? See, for
    instance, the bugs for grip (I have at least two filed there, both with
    more than 200 days).

    And there are many other examples of unresponsive maintainers. Would it be
    the case of the project taking a drastic change and dropping such
    developers (and some of the buggy, obsolete packages) more often than what
    is currently done?

    Perhaps this would be a good movement to get "the cruft" out of Debian and
    get it to a manageable size. Perhaps that could lead to slightly faster
    release cycles.

    > They don't respond to pings from other developers or users.

    Yes, I have suffered from that, as expressed above. :-(

    And I think that the Project Leaders (say, Project Scud) could take some
    actions regarding this. It would only benefit the project, as far as I can
    see.

    > Then when someone NMUs their package to fix a bug or something like that,
    > they immediately revert the changes.

    Which means that they actually had time to revert the changes and upload a
    new package. Why couldn't they fix the package in the first place? Or
    simply marking a bug with "wontfix".

    That's enough for the user to know the status of the package s/he is using
    (and to know if a locally compiled version should be made).

    Just my 2 cents, Rogério Brito.

    -- 
    Rogério Brito : rbrito@ime.usp.br : http://www.ime.usp.br/~rbrito
    Homepage of the algorithms package : http://algorithms.berlios.de
    Homepage on freshmeat:  http://freshmeat.net/projects/algorithms/
    -- 
    To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org 
    with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
    

  • Next message: Jacob Friis Larsen: "Re: dnscache: readproctitle service errors"

    Relevant Pages

    • Re: Want to work with a Linux Group
      ... reporting bugs there are basically two things Debian developers do, ... The first step is to decide on which package you want to work. ... Request for adoption - a new maintainer is needed but the old one ...
      (Debian-User)
    • Re: A real package manager in action
      ... The maintainer is the person who does the packaging, ... not necessarily the original author who is upstream. ... It also spreads the support load: package ... It depends what the resource is, and the worst that can happen is ...
      (comp.sys.acorn.apps)
    • Re: why should packages migrate to testing automatically
      ... Sometimes a maintainer might want the packages to stay ... A particular package might call for more ... > attention from Sid users to expose its bugs... ... > 3) Gives the users of sid sufficient time to find a bug and report it. ...
      (Debian-User)
    • Re: maintainer not responding
      ... need a response like "I don't have time to fix these bugs, ... package should be orphaned but I'm not familiar with Debian's package ...
      (Debian-User)
    • Re: maintainer not responding
      ... or checking the upstream project to see if they know about the issue? ... why I could not discuss these bugs with the upstream authors and report back. ... Bugs can be manipulated by anyone, not just the maintainer. ... rc-alert and/or wnpp-alert (finding the package containing these is left ...
      (Debian-User)