Re: [Axiom-developer] ChangeLog on Silver

[Gabriel Dos Reis]

If you read my suggestions/amendments to Martin's proposal, you'll
see I suggested:

   * a set of reviewers who review other people's patch.
     Review works by sufficient consensus, not necessarily a voting
     system. 

     (+) A subset of those reviewers are "global" maintainers.

     The set of reviewers should be open so that we integrate new
     reviewers as they have developed enough knowledge of the system.

    
     The set of reviewers usually consists of component maintainers.
     Global maintainers would try to defer to component maintainers.

   * A release manager.  The release manager is NOT a benevolent
     dictator.  Rather, it is someone committed to making releases
     happen on regular basis.  The release manager ideally is also 
     a global mantainer, although that is not necessary.  He set up
     schedules and try to get people's act to make things happen.


   * A College (or Steering Committee) whose job is to make sure that
     the project is not over taken by an individual.  I truly believe
     we need that.  I've spoken to and received emails from some
     original Axiom developers who also promote the College view.

     The College's job is not to make technical decisions.  The
     College will be discussing and making decisions about political
     issues.  Any developer can at any time request the College to
     consider any matter of his/her importance.
     The College appoints maintainers and release managers.

     The College should not consist only of developers.  Users should
     be represented.  In general, people with either long term
     committement or interest in Axiom should be represented there.

     I suggested to keep the number of College members small.  I
     suggested 7 for concreteness.

-- Gaby