RE: scope of id() (was: recalculating several models)

>3/ I'm affraid this also means that you can't have any kind of dependencies
>between instances, which is a must IMHO.
>
>Is this correct ?

There is still the attribute model as part of the binding attributes.

<model id="m1">
  <instance>...

<model id="m2">
  <instance>...

<input model="m1" ref="a/b/c">...

<input model="m2" ref="d/e/f">...


Additionally, any bind element (on any model) can contain an id, which is
then referenced by

<input bind="bind1">...

So the question of cross-model computational dependencies still applies!

.micah