svn commit: r422114 - head/misc/fortune_strfile

[Adam Weinberger]

> On 14 Sep, 2016, at 10:49, John Marino <freebsd.contact at marino.st> wrote:
> 
> On 9/14/2016 11:40, Kurt Jaeger wrote:
>> Hi!
>> 
>>> My own opinion is that 4814 is way too many.  And, I don't buy the
>>> argument that some have made that "unmaintained ports are better
>>> maintained than some maintained ports".
>> 
>> We have the data to go from opinion to knowledge by analyzing
>> the commit logs etc. Analyzing it is difficult, but maybe it helps
>> to find out where we stand.
> 
> 
> I don't think that's necessary to get the actual numbers.
> While it's clearly true that some unmaintained ports (aka maintained by ports at FreeBSD.org) are better maintained than a significant number of maintained ports, I think most people would agree having a maintainer is the best situation.
> 
> In this particular case:
> 1) I don't maintain any fortune ports
> 2) I don't use any fortune ports
> 3) I don't care if all the fortune ports are deleted
> 4) The port is about as trivial as they come.
> 
> I was only fixing a problem that I identified that should have been fixed long before.
> 
> If there was a rule that said I had to maintain the port for 1-week or 1-month or even 6-months, then I'm just dropping the port the next day after the expiration period.  It's better to give somebody that actually does care a chance to adopt it (the most likely being one of the fortune port maintainers).
> 
> Finally, most of the games ports are intentionally unmaintained.  Since strfile has its origin in games, I really didn't see a distinction with fortune_strfile and any of those games.

Hi,

This whole argument is kindof silly. John is being told that ports should be maintained for a minimum arbitrary amount of time that's up to the committer, but that he didn't pick the right minimum arbitrary amount of time that was up to him.

I agree with Mark that the PHB doesn't need to be an endless collection of rules that cover every possibility, but I disagree that this particular issue doesn't belong in there.

I will agree with Mat that 1 day is not enough time, because somebody needs to have ownership if users report problems that the committer didn't think of. That said, John is very responsive and if users reported problems I have no doubt that he'd address it quickly and properly.

The PHB should say:

The right amount of time that a new port should be maintained is at least long enough to verify that there are no build failures on any platforms, and that end-users have had a chance to report failures that they encounter in real-world use.

# Adam


-- 
Adam Weinberger
adamw at adamw.org
http://www.adamw.org