Thanks.
Winson
On Mon, 9 Apr 2001, Cheng Tai Suan Donavan wrote:
>
> It seems that the test cases for the Depth First graph traversals are
> based on the recursive algorithm for DFS. If you trace the graph using
> the non-recursive stack algorithm and then use the stack algorithm, you
> would find that the results are different - even though the stack
> algoritm for DFS is a "bona fide" DFS graph traversal.
>
> How then can you expect the students to write a stack variant of DFS and
> expect them to pass all the autotest answers which were based on a
> recursive variant of the same traversal? Pardon my impudence, but that
> hardly seems to be fair.
>
> We hand in Graph.cc's with recursive DFS's - get full marks in autotest
> but get docked by the TAs.
> On the other hand, hand in Graph.cc's with Stack variants of DFS's - get
> no marks in autotest and get pity marks from the TAs.
>
> It seems that we're caught between the devil and the deep blue sea here -
> what should we do?
>
>
>
>