[Dbix-class] IMPORTANT: A discussion of DBIC governance and future development

[Darren Duncan]

On 2016-10-06 8:43 AM, Matt S Trout wrote:
> On Tue, Oct 04, 2016 at 08:17:49PM -0700, Darren Duncan wrote:
>> That would be good, also in light with how that sentence continues:
>>
>> "I suspect what we need to try and achieve is to get DBIC a bit more
>> decentralised - have it be a specific framework build atop a
>> more-like-Plack-for-DB-stuff - but you already know that's what I
>> have in mind and we both already know it's going to be a big-ass job
>> and we'll see if it pans out or not."
>>
>> My own near term planned contributions to DBIC are precisely what
>> you said above.  They would constitute a more-like-Plack-for-DB
>> ecosystem and in particular they should benefit DBIC by optimizing
>> it more for maintainability, so it is easier for others to add
>> features or make changes or otherwise just be more confident that it
>> works properly.  If things go as I hope, this should start to land
>> in about a month.
>
> I think anything that's in that sort of vicinity is likely to be a big
> enough set of changes that it'll (a) need to wait until whatever new
> administration we end up with is settled in (b) need a decent RFC process
> and advance discussion of the risk/reward trade-offs involved.
>
> After all, when I say "big-ass job" I'm generally not kidding about that,
> and at this point it looks like it's not only going to be a big-ass job,
> but a big-ass job we're going to have to conduct without the help of the
> person I was relying on being the other half of the architecture team for
> it.
>
> So, I mean, "cool" but also "this is going to need serious discussion" and
> especially "please don't get your hopes up about 'near term'".

Not to worry, and I agree.

The first version of the thing that I was intending to land in the short term 
was only intended to be, on its own, classified as a green field experiment or 
proof of concept.  It would NOT by any means be intended for production as is.

Basically I am working on a Plack-for-DB as an independent project, and I was 
going to use an experimental fork of DBIC (on GitHub) as an initial test case by 
roughly replacing DBIC guts to use that project while using the fact that DBIC's 
pristine automated test suite as a validation that DBIC still behaves correctly 
with the changed internals.

The new administration of DBIC can then use this working proof of concept in 
their discussions on how they want to formally evolve DBIC.  My experiment would 
constitute an RFC for how would you like to use my Plack-for-DB, or adapt its 
design, to implement DBIC features that were long desired but not provided.

-- Darren Duncan