[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Orphaning php-codesniffer, then take it over by the PHP PEAR team

Hi Charles,

On 12-05-31 at 08:29am, Charles Plessy wrote:
> Le Wed, May 30, 2012 at 11:11:51AM +0200, Jonas Smedegaard a écrit :
> > 
> > *nothing* qualifies for a hijacking.
> Dear Jonas,
> your reaction seems to imply that hijacking is an implicit statement 
> of failure.  But this can be dis-ambiguated by thanking the maintainer 
> for his past work, bringing the package in a team where everybody is 
> welcome, which is not the same as moving it between two single 
> maintainers, and by underlining that the reason for transferring the 
> package to a team is that the mainainer, while not on vacation, could 
> not be contacted.

No, my point is not about lack of recognition for past contributions.

No, it is not about teams being welcoming or superior in others ways.

It is about Debian not being the Wild West.

I distinguish between "hijacking" and "friendly takeover" and "project 

What you describe is the proper way of a friendly takeover: You agree 
with the former maintainer to take over, and in the changelog entry 
formally announcing it you thank the former maintainer for the past 
efforts. It can be a person or a team taking over, and if the team also 
includes the former maintainer it might make sense to skip the "thank 
you" part.

"project overruling" is when Debian by its defined procedures judge that 
the maintainer is unfit to maintain, and therefore relieve her/him/them 
of her/his/their duties,

Hijacking, in my vocabulary, is when a non-maintainer takes matters in 
his/her/their own hands and takes over maintainership without the 
consent of the former maintainer and outside formal Debian procedures.

> There is indeed a problem with packages not maintained by DDs as they 
> can not formally declare themselves on vacation, but search engines 
> would be able to find such a statement on mailing lists if it had been 
> posted.

My non-Debian-member-as-maintainer rant is not about vacation notices, 
but about bonding and commitment.

It's somewhat like making a baby versus having a baby:

Making a Debian package requires certain skills, and some work once (for 
an hour or a year, depending on who you are).  We dearly encourage 
anyone interested to try it out, and share with the results with us and 
the rest of the World.

Having a package (a.k.a. being a Debian maintainer) additionally 
requires passion and devotion for a looong time - ideally for the full 
lifetime of the package.

I don't mean to say that only the Debian elite should raise babies, 
others should use protection - that Debian members are better parents.

My point is that the social structures of Debian matters for the quality 
of maintainance: When you do a poor job in Debian, you get remarks by 
your peers about it. When you do a poor job outside of Debian, there is 
silence. Your bonding with Debian helps you either care more or pass it 
on to someone who cares or get rid of the package.

Debian should avoid elitism by welcoming newcomers to our village, not 
by encouraging raising kids in the jungle.

> I think that it is important to have the flexibility to transfer a 
> package for which the maintainer is not responding, in a neutral way 
> that is not making any judgement on why he is not responding.

I interpret "neutral way" as "through defined process" (i.e. our MIA 
process or the technical committee) rather than ad hoc.

I disagree that we need a defined process for "not responding" regarding 
a single package as opposed to our "missing in action" affecting all 
packages a maintainer is involved in: I believe there is no common 
pattern for the special cases of a maintainer being active but 
irresponsible towards a subset of packages.

MIA process should not be nosy: We should not care _why_, only _if_ a 
maintainer is not doing the job responsibly.  I am unaware if current 
process is too nosy - but if so it seems ripe for improvements.

 - Jonas

 * Jonas Smedegaard - idealist & Internet-arkitekt
 * Tlf.: +45 40843136  Website: http://dr.jones.dk/

 [x] quote me freely  [ ] ask before reusing  [ ] keep private

Attachment: signature.asc
Description: Digital signature

Reply to: