The behavior is generic at the level you're referring -- restricting
position of an embedded morph should be handled by a subclass.


On Tue, Oct 2, 2012 at 12:51 PM, Christopher Oliver
<current.input.port at gmail.com> wrote:
> When one embeds a morph via direct manipulation into a descendent of a tranform morph, the embedded morph
> moves according to the transformation rather than stays in place.  While I understand what's going on, this seems
> very counterintuitive.  The morph can get moved outside of the Transform's clipping rectangle appearing to vanish.
> Should embedding invert the coordinates of the morph to embed first?  I'm not saying this is easy, but the current
> behavior doesn't seem like "the right thing"(tm).  I'm using a 4.3 image for the record.
>
> --
> Christopher Oliver <current.input.port at gmail.com>
>