There is not an imperative to *do* anything about
inheritance. To deprecate it (or rip it out) because
it does not seem to be a "selling point" or easily
explainable is absurd.
As long as it is there and does not cost people who
do not use it anything, and there are people who
find it useful, then just leave it alone.
elein
On Wed, Jun 30, 2004 at 09:53:53AM +0200, Thomas Hallgren wrote:
> elein wrote:
> >I strongly suggest a more cautious approach than "ripping
> >things out." Ripping out time travel got us row space
> >but lost point in time recovery. And is pushing us to
> >implement more traditionally logging. It was an OK trade
> >in the long run, but it took us a long time to get around to pitr.
> >
> The approach could perhaps be as simple as changing the terminology.
>
> Kind regards,
>
> Thomas Hallgren
>
> ---------------------------(end of broadcast)---------------------------
> TIP 8: explain analyze is your friend
