Now. I'm rolling my own OO framework (isn't almost everyone?)
Now, the problem I see with this (and I've seen many others do it - even Pear are/have) is that arrays ARE basically double linked lists - and they operate at a compile level, so are generally faster.
ASfter much benchmarking, the only place I've seen a linked list class faster, is where new nodes/entries are placed at the head of the list (or unshifted into an array version).
For all all other uses, an array is quicker. So. Instead of creating a linked list class (single or double) in the traditional manner - it seems a better approach is to do it using an array within the class.
Unless you are moving material to the head. Even a stack can be simulated better with an array.
Is there something I'm not seeing here? ANyone else have similar experiences?