lilypond-devel
[Top][All Lists]
Advanced

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

Re: how many patches can we review?


From: Mike Solomon
Subject: Re: how many patches can we review?
Date: Tue, 16 Aug 2011 10:24:56 +0200

On Aug 16, 2011, at 10:08 AM, Graham Percival wrote:

> We're up to 59 patches now, plus 17 patches on the "plz review; no
> known problems" list.  We're losing ground quickly, and this will
> soon become a serious problem for developers and contributors (if
> it isn't already).
> 
> This is not an official proposal, so don't worry about trying to
> "get it right".  If we try something and it sucks, we can change
> it instantly.
> 
> 
> I see three options, none of which fill me with joy.
> 
> 1) encourage more pushing without reviews.

I like this the least.

> 
> 2) assign certain people to be in charge of certain areas (e.g.
> Carl has final say over anything beaming-related), and encourage
> that person to review+push patches in his area ASAP.
> 

I also dislike this.

> 3) double or triple the rate of patches in the countdown.
> 
> 

I don't mind this at all.  I think that if people police themselves and are 
willing to take fast responsibility for after-patch cleanup if needed, this is 
not at all problematic.

> Out of those options, I dislike #3 the least.  I'm thinking of
> having 10 patches per countdown, 48 hours for each countdown.  I
> know that's a lot of reviewing, but if we want to keep the
> opportunity for reviews, that's the kind of workload we need to
> handle.
> 
> It may be a bit worse right now due to people in academia having
> less to do in August instead of September/October (when the
> academic year begins in the northern hemisphere)... but I'd really
> like to be able to tell people that their patch will get onto a
> countdown within at least a week.  In order to do that, we need to
> get through a lot of backlog first.
> 

This is not untrue.  That said, I find that there is a positive correlation 
between the amount of my music that gets played / paid / enjoyed and the amount 
of development I do for LilyPond (I have no clue how this works, as in theory 
this means I have less time, but oh well!).  So, I don't anticipate stopping my 
quest to eliminate all collisions[1] from LilyPond in the near future.

Cheers,
MS

[1] Within reason.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]