svn commit: r335281 - in head: . audio audio/gnump3d

[John Marino]

On 12/3/2013 02:59, Adam Weinberger wrote:
>>> (2013/12/02 @ 1347 EST): Mark Linimon said, in 1.5K: <<
>> My reasoning is that contributing to FreeBSD is sometimes
>> hard and thankless, and if someone just hears too much negative feedback,
>> they may just find something else better to do.
>>> end of "Re: svn commit: r335281 - in head: . audio audio/gnump3d" from Mark Linimon <<
> 
> The style police (I'm looking at you, Alexey) should be allowed to fix
> commits for "correctness," and committers should be allowed to fix
> obvious mistakes for each other.
> 
> 1) You have the confidence that if you accidentally commit something
> with an obvious error, someone is likely to help you out and fix it for
> you.
> 
> 2) You don't have to sit there and stew when a commit accidentally
> breaks that port you love. Because let's face it, nobody wants to file a
> PR for ${PORT_OPTIONS:MODCS}.
> 
> 3) If everybody is empowered to fix obvious mistakes and style nits, you
> gain a consistent experience for the end-user, who we all theoretically
> work for. You know, rather than just having Alexey send you an email
> about the importance of not muting install commands, which everybody
> seems to ignore anyway.
> 
> 4) People need to un-knot their panties. Seriously. We're talking about
> "but *I* wanted to delete that port!" Someone did your work for you.
> Think about all the things you could do with the 30 seconds they just
> saved you.
> 
> But when faced with the choice between "let's all fix bugs and promote a
> consistent ports user experience" and "MINE!", we all know which one
> people will pick.


Except for the (unwarranted?) attack on Alexey, I'm pretty happy with
this post.

I want to address this statement from Mark, "While on portmgr I
maintained a deliberate bias in favor of existing maintainers.  My
reasoning is that contributing to FreeBSD is sometimes hard and
thankless, and if someone just hears too much negative feedback,
they may just find something else better to do."

I think this bias has backfired.  I also do not believe that a "regular"
maintainer, one that is responsible for 1-30 ports but no other
responsibilities, is not a creature that needs protecting.  If they are
doing the job they volunteered to do, they aren't going to get "negative
feedback".  If they have twisted knickers because somebody fixed an
obvious typo for them, then frankly they are too high-strung to be a
maintainer.

In many cases, no maintainer is better than an abusive maintainer, just
ask any user of vim.

Maintainers don't "own"  the port, they are simply the primary
caretaker.  Adam is completely correct that the main losers in this
"protection" of the maintainer are the ports users.

The other thing to consider is the trade-off.  Mark decided to trade off
the rights of active maintainers in order to protect maintainers that
may be long inactive and unresponsive.  This demotivates the active
volunteers!  Nothing is free, every rule costs something.  I don't know
if the effect on the other committers has been properly considered.

Please make it clear that MAINTAINER does not equal "owner" and
establish exactly when other can fix maintained ports.   I don't think
it is hard.  Is it broken on every platform?  ok.  Is it an obvious typo
causing a malfunction?  ok.   Does the proposed change alter the
original intent of the maintainer?  no means "ok".

I think we want those maintainers that severely neglect their duties to
"find something else to do" and we want to instill that all ports are
community property and not individually owned.

John