On 23/04/2016 12:33, Mattia Rizzolo wrote:
some times when you're going through and creating a README.source, and
documenting, like 50 included R data files (explaining why the source
contains binary files), when you know that no-one will ever read it ... you
start thinking in terms of "what will get this over the line?", rather than
"i want to package this well" (perhaps this makes me a bad packager).
umh.. I'm having hard time understanding what you wrote here, tbh.
just saying that sometimes packaging isn't fun, and there's a lot on
your plate, and you reduce things to simpler heuristics like (all
warnings = bad). but it's worth taking a step back and seeing the big
picture - like the debian policy as you suggest, or that this software
is part of an eco-system, not an isolated transaction.
cheers
jonathon
