[EM] 2nd reply to Markus on 12/20/03

[MIKE OSSIPOFF]

I'd said:

>What started this discussion was when Markus said that my BeatpathWinner
>algorithm wouldn't work, because it isn't the Floyd algorithm, whatever
>that is.

Markus replied:

Craig Carey claimed that my implementation of my method doesn't work
(presumably because it makes only one pass through the triple-loop).
I explained to Craig that it is true that when I had considered the
possible short cuts in that order that has been proposed by you then
my implementation would not have worked.

I reply:

Actually, you posted a fragment of our Python program and said that my 
algorithm won't work.

But if you're now backing down from that claim, that's good.

Markus quoted me:

You wrote (28 Feb 2001):
>This is the Floyd algorithm for making an array of greatest beatpath
>magnitudes between each pair of options:

I reply:

You see, Markus, this is why I refer to you as an idiot. Did I deny that, in 
Feb 2001, I was calling Steve's algorithm the Floyd algorithm.? No. I agreed 
that I'd formerly called that algorithm the Floyd algorithm, but that I no 
longer do. But you kept repeating that I continue to claim that my 
implementation is the Floyd algorilthm, though I kept trying to tell you 
that I no longer make any such claim.

And now you post a quote from 2001, apparently believing that it shows that 
you're right to say that, during this current discussion, I claim that 
Steve's algorithm is the Floyd algorithm.

Seriously, Markus, all namecalling aside,there really is something wrong 
with you.

Ernie:

Can you really blame me for eventually beginning to make unflattering 
comments about this twit?

Markus continued:

You wrote (17 Dec 2003):
>I re-emphasize that I didn't get our strongest beatpaths algorithm from
>you [= Markus] or Floyd, or anyone but Steve Eppley, who suggested it.

If you really got your strongest beatpaths algorithm from Steve Eppley
and not from Floyd or me then why did you call it "Floyd algorithm"?

I reply:

How many times is it necessary to explain that to you. If I explain it to 
you now, one more time, are you just going to forget again?

Ok, Steve wrote the algorithm. Some time later, you posted something here 
that resembled Steve's algorithm, but which only made one pass through the 
3-candidate permutations. You called it the Floyd algorithm. I assumed that 
you must have accidentallly miscopied the algorithm, accidentally writing so 
that it only make one pass through the permutations. So, believing that 
"Floyd algorithm" is the name of the algorithm that you'd miscopied, and 
believing that it must make several passes, then it looked as if "Floyd 
algorithm" was the name of Steve's algorithm.

So I got the algorithm from Steve, and I got the name from you, because I 
believed that what you'd tried to copy was something that was the same as 
Steve's algorithm. Now, did it sink in this time, or are you going to ask 
the question again in a few days?

Markus continued:

I have explained the Floyd algorithm in a private mail (30 April 2000)
to David Catchpole, Blake Cretney, Steve Eppley, Rob Lanphier, Norman Petry,
and you. I don't remember that Steve Eppley called his implementation
"Floyd algorithm". Therefore, I guess that you have got your algorithm
from me

I reply:

That certainly follows if the fact that I got the name from you means that I 
got the algorithm from you. But it doesn't mean that. I got the name from 
you, and, believing that you'd miscopied something that was the same as 
Steve's algorithm led me to believe that "Floyd algorithm" was a name for 
Steve's algotithm, that they were the same.

Markus continued:

, but that you have never understood this algorithm sufficiently
to implement it in such a manner that it has a runtime of O(N^3).

I reply:

As I said, it seemed to me that you must have miscopied it, and that you'd 
tried to copy something that was the same as Steve's algorithm.

Mike Ossipoff

_________________________________________________________________
Worried about inbox overload? Get MSN Extra Storage now!  
http://join.msn.com/?PAGE=features/es