On Wed, Jul 16, 2008 at 10:45 AM, Stephan Richter <srichter@cosmos.phy.tufts.edu> wrote:
But I agree. The system supports simple type and that is one of the use cases. While I originally wanted to add at least 2 of the events to the IAnnotations adapter itself, I agree with you that the overhead penalty might be too severe.
I don't think I said anything about a penalty on this, though that could certainly be an issue. I doubt it really is, though.
With this in mind, I am modifying my proposal to add 2 event notifications to the factory only to produce a function like this (it is the result of calling the annotation factory):
My preference would be that these events would not be specific to the factory, keeping the model clean.
I am torn whether the created event should be a annotation specific or not, because an object is not an annotation until it has been added as one.
The created event should not be annotation-specific, for the reason you cite -Fred -- Fred L. Drake, Jr. <fdrake at gmail.com> "Chaos is the score upon which reality is written." --Henry Miller