Re: An exact simplification challenge - 57 (arctangents) - Proud

I'd like to note that the transformation that leads to the
simplification doesn't in fact take as long as my previous message
would suggest:

In[2]:= FullSimplify[expr, TimeConstraint -> .1] // Timing

During evaluation of In[2]:= FullSimplify::time: Time spent on a \
transformation exceeded 0.09999999999999999` seconds, and the \
transformation was aborted. Increasing the value of TimeConstraint \
option may improve the result of simplification. >>

During evaluation of In[2]:= FullSimplify::time: Time spent on a \
transformation exceeded 0.09999999999999999` seconds, and the \
transformation was aborted. Increasing the value of TimeConstraint \
option may improve the result of simplification. >>

During evaluation of In[2]:= FullSimplify::time: Time spent on a \
transformation exceeded 0.09999999999999999` seconds, and the \
transformation was aborted. Increasing the value of TimeConstraint \
option may improve the result of simplification. >>

During evaluation of In[2]:= General::stop: Further output of \
FullSimplify::time will be suppressed during this calculation. >>

Out[2]= {12.187, 1/30 (3 + 2 Sqrt[5] + Sqrt[15 + 6 Sqrt[5]]) \[Pi]}

This took only 12 seconds.  I wish there were a way to tell
FullSimplify:  "Stop, and just give me the simplest form you've found
so far!"  But maybe this wouldn't work.  Maybe the transformations
that take so long are performed _before_ the transformation that leads
to the simplification.