Re: survey: psql syntax errors abort my transactions

On Thu, Jul 2, 2020 at 8:54 AM Jeremy Schneider <schnjere(at)amazon(dot)com> wrote:

>
> https://www.postgresql.org/message-id/flat/CABTbUpiAOKZ405uArt8cJFtC72RhzthmvWETQK_6Qw0Ad-HquQ%40mail.gmail.com
>
> This thread on hackers actually seemed kindof short to me. Not nearly
> enough bike-shedding to call it a closed case.
>

Seemed about right:

"We should do this"
Yes
No - with a reason
No - with a reason
No - improve user education
No - emphatically
Yes - but ends ups deferring to the majority

No one else chooses to voice an opinion

The status quo prevailed since no-one chose to contribute further arguments
for change and the original patch was retracted. What kind of
"bike-shedding" (which seems to be used incorrectly here) would you expect?

All I can speak for is personal usage but I don't find the current default
to be an issue. I'm also generally opposed to changing this kind of
default even when I disagree with its current value. If anything psql is a
bit too permissive by default IMO. Default should be as safe as possible
even at the cost of user inconvenience - so that unknowledgeable people get
the most protection. If options exist to trade safety for convenience that
is good - each user can make that trade-off for themselves and in the
process be aware of what exactly their decision entails.

David J.