...
> Since I didn't see an immediate batch of squawks, I think I will go
> ahead and commit what I have; we can always revisit the implicit-allowed
> flag settings later.
Squawk. But I haven't had time to look at the full ramifications of your
proposed change, so it is inappropriate to comment, right?
We have never been in complete agreement on the optimal behavior for
type coersion, but it seems that most users are blissfully ignorant of
the potential downsides of the current behavior. Another way to phrase
that would be to say that it actually does the right thing in the vast
majority of cases out in the field.
We'll probably both agree that it would be nice to avoid *hard coded*
rules of any kind for this, but do you share my concern that moving this
to a database table-driven set of rules will affect performance too
much?
- Thomas