svn commit: r422505 - head/archivers/snappy-java

[Tijl Coosemans]

On Wed, 21 Sep 2016 09:11:51 -0500 John Marino <freebsd.contact at marino.st> wrote:
> On 9/21/2016 09:05, Mathieu Arnold wrote:
>> Le 21/09/2016 à 15:56, John Marino a écrit :  
>>> Maybe it's time for portmgr to research this and publish a list of
>>> forbidden upstreams if they find it really is a cache and find they
>>> really don't want it.  
>>
>> No. Like I already told you on another subject, we are not a law firm.
>> We are not turning the Porter's Handbook into a 35k pages and 42 annexes
>> ISO9000 monstruosity.
>> You have to use common sense.
>> A cache is a cache, and it is not suitable as an upstream.
> 
> It's not about law, it's about rules and procedure.
> I come from an operational environment where everything is CONCISELY 
> documented (the handbook is not so different).  Currently, what I did is 
> allowed by the rules and IMO it's not underhanded at all.  It was done 
> with the best of intentions.
> 
> You, as portmgr, have no right at all to either accuse me of being 
> underhanded or getting angry about my legitimate solution which, IMO, 
> makes perfect sense.  If the criteria is common sense, I met that 
> criteria with flying colors and you've got no standing other than your 
> own "common" sense that I'm wrong.
> 
> Stop slighting the legal profession.  This is something that requires 
> more guidelines than is available, and as a member of both documentation 
> and portmgr, you have the ability to improve it.

I would say using Fedora's cache is leeching and not very nice.  On the
other hand I wouldn't have marked the ports broken.  Use of our distcache
for 14 days should be acceptable.  After that you can mark it broken.

I also think that broken ports shouldn't be fixed until a user shows up
and we know somebody actually cares.  Spending committer time and build
time on something nobody uses is a waste.