Re: [COMMITTERS] pgsql: Use a bitmask to represent role attributes

On Tue, Dec 23, 2014 at 11:44 AM, Stephen Frost [via PostgreSQL] <
ml-node+s1045698n5831875h49(at)n5(dot)nabble(dot)com> wrote:

>
> It would be great to figure out a way to get feedback like this earlier
> on in the development. This patch has been floating around for quite a
> while, with intentional breaks for feedback taken prior to incremental
> improvements and documentation additions. Clearly that gets back to the
> discussion around the commitfest situation.
>
>
​There four possible situations here:

Seen, agreeable
Seen, not agreeable
Seen, abstain
Not Seen

Tracking, for the committers in particular, ​the not seen and directly
soliciting their agree/disagree/abstain opinion is really the only way to
avoid this situation where Tom probably saw the subject lines but ended up
filtering them out since his focus was elsewhere. However, something
getting committed definitely gets his attention.

FWIW my initial reaction to this idea of introducing bitmaps was "why?" but
I didn't have anything to go on but the feeling that bitmaps are not the
most obvious API in modern coding. I didn't have anything else to support
that, including coding experience, so I didn't voice it and figured when no
one else did that I likely was missing something. I'm not sure how an
email to Tom saying: "hey, this doesn't smell right to me" would have been
taken but changing the underlying authorization mechanisms does seem like
something Tom should comment on before development gets to far along - and
that input should be prompted for if not seen.

That's my current feeling as strictly a monitor of these lists and
observing a subset of the threads and new features that are being currently
developed - take it as you will.

David J.

--
View this message in context: http://postgresql.nabble.com/Re-COMMITTERS-pgsql-Use-a-bitmask-to-represent-role-attributes-tp5831838p5831894.html
Sent from the PostgreSQL - hackers mailing list archive at Nabble.com.