On 12 Apr 2009 at 21:53, Jay West wrote:
>>
I've run into "CS" graduates from the university who supposedly
>> have learned about really "good" sort algorithms (usually
>> "Shell-Metzner"), without understanding which situations an
>> optimized "bubble" is better for.
What? No mention of the n-way polyphase sort/merge?
CS grads don't do external sorts. ;)
Not that the experts are particularly easy to work with. I recall
one very memorable episode where the compiler front-end writer and
the optimizer writer couldn't agree on the type of tree (binary or n-
ary) to use for the internal representation of a compilation unit.
The argument went on for more than a month, while both parties were
writing their code trying to be the first done so that their method
would be the one adopted. Neither would give an inch.
I finally ended up writing the code to transform a n-ary tree into
it's binary representation. Utter insanity just to keep the peace.
Both participants were CS PhDs, who should have known better.
--Chuck