> OTOH, my reasoning is that, *I* (so it's not as if EM is doing
> something I didn't requested) did select the prototype and thus I
> expected it to switch my attribute to using the prototype even if
> that means resetting values. In fact I actually expected it to reset
> the attribute so that it exactly matched the prototype. But in this
> case it did exactly *nothing* (?) which is totally misleading and
> puzzling.
>
>
> When you speak about "overriding" settings I would expect that this
> is again something *I* would do *after* specifying the prototype.
Well, for instance, attribute name is defined in the prototype.
Should changing prototypes override your attribute name?
> Also, specifically for the external type, I'm wondering what would
> the benefit of using a prototype with an external type different
> from the one defined in the prototype? For the other settings the
> it's obvious that preserving their value may prove useful.
Even though you and I think this is strange, never underestimate the
confusing choices people make. EOF allows this, so we have to also.
I'm also not a fan of force-overriding the value only for that one
setting -- That's more confusing that not doing it for any.
I think maybe it could bring up dialog when you change prototypes on a
non-blank attribute asking if you want to replace the attribute's
settings, and it has a "Don't ask again" checkbox.
ms
This archive was generated by hypermail 2.0.0 : Sat Mar 08 2008 - 12:49:59 EST