On 01/14/2014 12:40 AM, Merlin Moncure wrote:
> On Sun, Jan 12, 2014 at 4:38 AM, Craig Ringer <craig@2ndquadrant.com> wrote:
>> Implicit casts to text, anybody?
>
> This backward compatibility break orphaned the company I work for on
> 8.1 until last year and very nearly caused postgres to be summarily
> extirpated (only rescued at the last minute by my arrival).
You're far from the only one, too. Until last year I was still seeing
people saying they "can't" upgrade because of this. OTOH, that was a
sudden and drastic change, with no BC switch like the removal of
implicit joins had, that affected wide swaths of code. Lets not do that
again.
Removal of lower bounds for arrays is unlikely to even get noticed by
the vast majority of users, and can be done progressively with BC features.
The real issue IMO is how to get those few to stop using it so it can be
truly removed. Past experience has shown that people just turn the
compatibility flag on and forget they're using the deprecated feature.
If there are warnings they'll silence them in their application and
still forget they're using the deprecated feature. If there are log
messages, they'll probably turn logging detail down to hide them and
STILL forget they're using the deprecated feature.
Then whine about it three years later when it gets removed.
So I guess the question is: Is it worth all that hassle to remove a
misfeature you have to go out of your way to use? Is support for non-1
lower bounds stopping us from doing something useful and important? Or
is it just an irritation that it exists?
-- Craig Ringer http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training & Services