[Dbix-class] Let's call things what they are 2/2

[Matt S Trout]

On Tue, Oct 11, 2016 at 11:55:42AM -0700, Darren Duncan wrote:
> Whereas, if I am wrong and the primary reason against releasing your
> existing work isn't about the risk of breaking things for users,
> then I don't understand why the future DBIC governance has any
> bearing on what you can do right now.

Right now, I still consider riba to be chainsaw delegate on technical
matters, and he has every right to complete and ship this work if he
believes that to be the best thing for the code and user base.

> Regardless, I agree with what you said in "1/5 What is stability"

The thing is ... mostly, so do I, and in general, over the years, the vast
majority of stability arguments that he and I have been involved in have been
ones where both of us have been on the same side - fighting for more attention
to be paid to stability than people seemed to be intending to - and it's
been really nice to have the company.

The examples given of my being against stability are, well, the most visible
times we've disagreed. But at least up until a couple of years ago, my
memory says we've been on the same side a lot more often than not.

Where we diverge is where you get to -

https://twitter.com/ribasushi/status/727539149822132224

'There is "general FOSS" and there is "platform FOSS" In face of
uncertainty In the latter case the imperative is to do nothing'

which I agree is a good *default*, but not always universally correct.

Some times something is a big enough leap forwards that it justifies taking
the risk of shipping it and having to clean up a few small fires that ensue
as a result. To pick an example dear to my heart, we went through six
dev releases before shipping 0.05 (and had enough people using DBIC against
their production database that I was already moderately paranoid), but
then still had to do a couple of releases very quickly afterwards to fix
bugs people found.

On the other hand, that release is when ResultSet acquired a search() method
and became chainable.

So, basically: from where I'm standing I hold 90% of riba's stance to be
self evident, but prefer to be utilitarian rather than dogmatic about the
remaining cases.

I'd hope that the combination of a core team and a public decision making
process that consults the userbase about priorities will be sufficient to
ensure that if people thinking I'm weighting said utility function wrongly,
I'll find out before applying the results.

-- 
Matt S Trout - Shadowcat Systems - Perl consulting with a commit bit and a clue

http://shadowcat.co.uk/blog/matt-s-trout/   http://twitter.com/shadowcat_mst/

Email me now on mst (at) shadowcat.co.uk and let's chat about how our CPAN
commercial support, training and consultancy packages could help your team.