>That's assuming that weak references are slower than
>strong references, which I thought I read somewhere...

They are slower because they need to do a hashtable lookup instead of a pointer 
dereference.

>Agreed.  Though this discussion has piqued my interest about renewed
>discussions in ListenerList/WeakReference.

I don't think we want to go there. Yes, there are issues with maintaining hard 
references to objects in memory, but, as noted, they are not particularly 
common and easily manageable.

G


Reply via email to