example comparing method weights

thesis-pics/dsu-weight-bad-example.c

@@ -0,0 +1,19 @@
+#include <ds.h>
+
+const int n = 4096;
+const int x = 10;
+
+int main(int argc, char **argv)
+{
+	ds d = init_d();
+
+	for(int i = 0; i < n; i++)
+		insert_d(d, i, 0);
+
+	delete_max_d(d, 1);
+	delete_min_d(d, 1);
+
+	for(int i = 0; i < x; i++)
+		search_d(d, i, x);
+
+}

thesis.tex

@@ -531,8 +531,17 @@ \section{Extensions of the idea}
             the operation for that data structure identity. That's not a very precise solution, because a lot of
             less-weighted heavy operations can overpower a crucial bottleneck fast operation.
 
-            \todo{example}
+            \begin{figure}[!h]
+                \lstinputlisting{thesis-pics/dsu-weight-bad-example.c}
+
+                \caption{}
+
+                \label{fig:dsu-weight-bad-example}
+            \end{figure}
 
+            \begin{verbatim}
+            \texttt{http://www.wolframalpha.com/input/?i=x+\%2B+2+*n+\%3E+\%28x+\%2B+2\%29*+log\_2\%28n\%29\%2C+n+between+0+and+1000\%2C+x+between+0+and+1000&dataset=&equal=Submit}
+            \end{verbatim}
             A better approach would be to approximate the current element count $N$ and actually evaluate the complexity
             function on the element count, multiplied by the weight, then the sum of those would be our metric of
             profitability of a data structure implementation. That would more accurately describe the cost and would be