[EM] Answering Markus once, to conclude the topic

[Markus Schulze]

Dear Mike,

you wrote (27 Jan 2004):
> Markus claimed that I had said that I've  never heard of the Floyd
> algorithm, when the passage that he posted said only that (at a
> specified earlier time) I _had_ never heard of the Floyd algorithm.
>
> When Markus said that I'd said that I'd never called Steve's algorithm the
> Floyd algorithm, when what I'd actually said, in the passage copied by
> Markus, was that I _don't_ call Steve's algotithm the Floyd algorithm (I'd
> said that in response to another false statement to the effect that I call
> Steve's algorithm the Floyd algorithm).

I suggest that you should use the term "anymore" more frequently.

When you say that you don't call your implementation "Floyd algorithm",
then, of course, I will point you to a concrete quotation where you call
your implementation "Floyd algorithm". However, when you say that you
don't call your implementation "Floyd algorithm" _anymore_, then I will
consider this to be a correction of your terms.

If you only wanted to say that you don't call your implementation "Floyd
algorithm" _anymore_ and not that you had never called it "Floyd algorithm",
then why did you spam this mailing list with tons of insulting mails instead
of just saying that you don't call your implementation "Floyd algorithm"
_anymore_?

The problem is: When you refuse to say that you don't call your implementation
"Floyd algorithm" _anymore_ and when you insult those people who mention that
you had called your implementation "Floyd algorithm", then you make the readers
mistakenly believe that you claim that you had never called it "Floyd algorithm".

Again: Why do you believe that it is of public interest why you mistakenly
believed what about the Floyd algorithm? I do not care why you mistakenly
believed that you had implemented the Floyd algorithm. But it is clear that
when we discuss how to calculate the strengths of the beatpaths it is
necessary for me to stress that your implementation is not the Floyd
algorithm. It is also clear that when I wrote on 15 Dec 2003 that you
mistakenly called your implementation "Floyd algorithm" this statement
was correct.

Actually, instead of being happy about the fact that you have been pointed
to the fact that the strengths of the beatpaths can be calculated with the
Floyd algorithm in a runtime O(N^3), you are upset about the fact that the
runtime to calculate the strengths of the beatpaths isn't prohibitive.
Otherwise you wouldn't have insulted people for pointing you to the
correct version of the Floyd algorithm. Your behaviour doesn't make any
sense unless you want election methods to be surrounded by some king of
mystery instead of being applicable tools.

Markus Schulze