[EM] Floyd algorithm?

[Markus Schulze]

Dear Mike,

you wrote (1 Feb 2004):
> So I'd suggest only changing that order if the runtime
> would be prohibitive without the Floyd order-change.

On the other side, when you use a faster algorithm you
can implement additional features. For example, you wrote
that you agree to my suggestion that the strength of a
pairwise defeat should be measured secondarily by its
margin.

*********

You wrote (31 Jan 2004):
> Markus'd said:
> 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".
>
> I'd replied:
>
> No, _your_ problem is: ....

I wrote (31 Jan 2004):
> No, it's _your_ problem. It is _your_ problem that you don't
> understand that you hurt yourself when you describe your favorite
> method in such a manner that the readers mistakenly believe that
> already for small numbers of candidates a computer is needed to
> calculate the winner and that for large numbers of candidates the
> runtime to calculate the winner is prohibitive. It is _your_
> problem that you don't understand that a discussion about the
> merits of the different implementations is _not_ a discussion
> about grammar.

You wrote (1 Feb 2004):
> A computer isn't needed when there are just a few candidates. If
> there are so many that one would want to use the Floyd algorithm,
> that's already enough work that one would want to use a computer.

On the other side, with the Dijkstra algorithm even examples with
dozens of candidates can be calculated by hand.

You wrote (1 Feb 2004):
> I don't know that that's true. I don't know if any reasonable
> number of candidates would cause a prohibitively long runtime,
> without Floyd's index-order change, using a modern computer.
> I suspect that it would take very many candidates to make the
> runtime prohibitive. That's something that could be determined
> by experiment.

On the other side, when you use a faster algorithm you
can implement additional features.

You wrote (1 Feb 2004):
> So it isn't clear why you're still having a problem about it.

It is your problem that you understand neither the Floyd algorithm
nor the Dijkstra algorithm. I don't understand why you believe that
this was my problem. I am not your babysitter.

*********

You wrote (31 Jan 2004):
> Markus continued:
>
> However, I haven't made a mistake since when I wrote on
> 15 Dec 2003 that you called your implementation "Floyd algorithm"
> my observation was correct.
>
> I reply:
>
> Can you post a quote in which I said that your first statement
> that I'd called our implementation the Floyd algorithm was
> incorrect?
>
> "I don't call that the Floyd algorithm" doesn't mean that you were
> incorrect a long time ago when you first said that I'd previously
> called it the Floyd algorithm.

I wrote (31 Jan 2004):
> On 18 Dec 2003, you wrote: "Wrong. I don't call that the Floyd
> algorithm." "Wrong" means "incorrect". Doesn't it?

You wrote (1 Feb 2004):
> Yes it does. And you were incorrect, which was why I said "Wrong".
> After I'd said that it wasn't the Floyd algorithm, you kept saying
> that I call it the Floyd algorithm. Yes, that's wrong and incorrect.
>
> As quoted above, I asked you for a quote in which I said that your
> first statement that I'd called our algorithm the Floyd algorithm
> was incorrect. You found a passage in which I said you were wrong
> because I don't call that the Floyd algorithm. I'd said that many
> times in reply to your continually-repeated claim, after I'd said
> that it isn't the Floyd algorilthm,  that I call it the Floyd
> algorithm.
>
> You've posted a quote in which I said you were incorrect to say that
> I call it the Floyd algorithm.
>
> You've previously indicated that you value honor. But that quote by
> me was said in reply to your statements made after I'd said it wasn't
> the Floyd algorithm. I've posted a copy of my words immediately after
> you said it isn't the Floyd algorithm, in which I acknowledged I'd
> called it the Floyd algorithm. So when you post that quote from a
> later time, representing it as a quote of my saying that your
> _initial_ statement that I called it the Floyd algorithm was wrong,
> that shows that you are without any honor. You've shown that the
> truthfulness of what you say isn't reliable. Liar is the usual word.

In my 15 Dec 2003 mail, I complained that you mistakenly called a
different algorithm "Floyd algorithm".

Then in your 18 Dec 2003 mail, you wrote: "I do call a certain strongest
beatpaths algorithm the Floyd algorithm, only because someone on this
list told us that that's what that algorithm is called." Later on the
same day you wrote: "I don't know if our strongest beatpaths algorithm
is the Floyd algorithm." Later on the same day you wrote: "Whether the
version that I wrote is or is not the Floyd algorithm, it certainly
finds the strongest beatpaths." Then in your 19 Dec 2003 mail, you wrote
that you decided not to call your implementation "Floyd algorithm"
anymore. You wrote: "I don't call my implementation the Floyd algorithm.
I no longer call it that."

Then in my 20 Dec 2003 mail, I acknowledged that you don't call your
implementation "Floyd algorithm" anymore. I wrote: "I don't claim that
you are continuing to claim that your implementation is the Floyd
algorithm." But this didn't hinder you from continuing to claim that
I continued to claim that you continued to claim that your algorithm
was the Floyd algorithm. You are the idiot because you are unable to
admit that you have made a mistake without bombarding with insults
that person who pointed you to this mistake. You are the liar because
although I acknowledged that you don't call your algorithm "Floyd
algorithm" anymore, you continued to claim that I continued to claim
that you continued to claim that your algorithm was the Floyd algorithm.

On the other side, although you have updated other parts of your
website recently, in your website you still call your implementation
"Floyd algorithm" (http://www.electionmethods.org/CondorcetSSD.py):
"Determine 'beatpath' magnitudes array using the Floyd Algorithm."

*********

You wrote (1 Feb 2004):
> And yet you were continuing to say that I call it that, even after
> I'd said that it isn't the Floyd algorithm, because, at some time
> in the past, I had called it that. Yes, I admit that you were
> contradicting yourself, or showing that you don't know the difference
> between "call" and "called". There's a parrot that has shown that it
> knows the distinction between past and present tenses. That parrot
> is one-up on Markus.  Maybe Markus should take lessons from it.
> If you're now through saying that I call it the Floyd algorithm, then
> I won't continue to claim that you continue to claim that I call it
> the Floyd algorithm--then I'll only say that you were incorrectly
> making that claim during the entire discussion up to this point.

Of course, I could acknowledge that you don't call your implementation
"Floyd algorithm" anymore only after you stopped calling it "Floyd
algorithm".

*********

You wrote (1 Feb 2004):
> It's a common feature of elementrary school repartee, to repeat what
> the other person has said about you, about the other person. No,
> there's no similiarity in what I've been saying and what you've been
> saying. False statements that I call it the Floyd algorithm, after
> I've said that it isn't the Floyd algorithm, because I'd called it
> that before I'd acknowldedged it wasn't that and before you pointed
> out that it isn't. Confusion about the difference between past and
> present verb tenses. Endlessly repeated replies to things that hadn't
> been said. No, I don't say anything resembling what the confused
> wording-Nazi idiot has been saying.

When I acknowledged in my 20 Dec 2003 mail that you don't call your
implementation "Floyd algorithm" anymore, this didn't hinder you from
continuing to claim that I continued to claim that you continued to
claim that your algorithm was the Floyd algorithm. You are the confused
wording-Nazi idiot because you don't see that you do exactly that thing
that you call "a common feature of elementrary school repartee".

*********

You wrote (29 Jan 2004):
> Look, this mailing list is about voting sytems, and it just isn't the place
> for you to find out about verb grammar. There must be grammatical discussion
> mailing lists. Couldn't you take your questions there instsead of here?
> you're off topic, and people don't appreciate your off-topic spamming about
> your grammatical misundestandings.

I wrote (29 Jan 2004):
> Actually, it is you who spams this mailing list with lessons in English
> grammar. I agree with David Gamble: "One of your rhetorical techniques
> is to mock and highlight unintentional errors of grammar and spelling."

You wrote (31 Jan 2004):
> But you're showing that you need the lessons. But I ask you to go somewhere
> else for them, because here the subject is off-topic. In this case, your
> errors of grammar lead you to endlessly spam the list with incorrect
> statements about another list-member. You're not posting about voting
> systems. You're off-topic. Aside from that, you're incorrect too.

I wrote (31 Jan 2004):
> Apparently there's no way to help Mike to understand that a discussion about
> the merits of the different implementations to calculate the strengths of the
> strongest paths is not a discussion about English grammar. The fact that Mike
> calls mails on such implementations "off-topic" questions whether he is
> really willing to discuss election methods.

You wrote (1 Feb 2004):
> You're right: I don't agree that this discussion has been about the merits
> of the different implementations to calculate the strength of the strongest
> paths.

The fact that you want to discuss English grammar doesn't change the fact
that this is a mailing list on election methods.

*********

I wrote (29 Jan 2004):
> When I wrote that your use of the term "Floyd algorithm" was
> incorrect, then I didn't point to a _grammatical_ error, I
> pointed to a _mathematical_ error.

You wrote (31 Jan 2004):
> No one said that that statement pointed to a grammatical error. But
> in other statements, when you're telling me that "anymore" is needed
> in order to indicate the present tense, you're making an incorrect
> grammatical claim.

I wrote (31 Jan 2004):
> I didn't say that the term "anymore" is needed. I said that it would have
> been better if you had used the term "anymore" to stress that you had
> corrected your terminology. Do you really believe that whenever someone
> makes a suggestion about how you could word things he talks about
> grammatical errors?

You wrote (1 Feb 2004):
> And I told you that it would be just as well if you didn't presume to tell
> me how to say things. You say it would be better if I used "anymore" to
> emphasize that I had corrected my terminology. But where were you when I
> corrected my terminology? I did so clearly and fully.  I said that our
> implementation isn't the Floyd algorithm. I said that immediately after
> you told me about Floyd's order-change. If you forgot that I "corrected
> the terminology", that's your fault, not mine.

On 19 Dec 2003, you wrote that you don't call your implementation "Floyd
algorithm" anymore. On 20 Dec 2003, I acknowledged that you don't call
your implementation "Floyd algorithm" anymore. So I acknowledged that
you don't call your implementation "Floyd algorithm" anymore immediately
after you wrote that you don't call your implementation "Floyd algorithm"
anymore.

*********

I wrote (31 Jan 2004):
> The large number of annoying and insulting mails from you suggests
> that you are upset about the fact that the strengths of the strongest
> paths can be calculated in a polynomial runtime. On the other side,
> several members indicated that they are interested in efficient
> algorithms to calculate the strengths of the strongest paths. If you
> are not interested then I suggest that, instead of insulting those
> who are interested in such algorithms, you simply shouldn't read
> those mails.

You wrote (1 Feb 2004):
> My annoyance with your confused wording-Nazi idiot postings results
> from your repeated mis-statements about what I said, what is meant
> by somethng I said, and your repeated replies to things that were
> never said.

Please stop insulting those who are interested in discussing efficient
algorithms! When in your opinion those who are interested in efficient
algorithms are confused wording-Nazi idiots then you have to live with
the fact that most people are confused wording-Nazi idiots.

Markus Schulze