[EM] Bolson reply, 5/23/04

[MIKE OSSIPOFF]

I'd said:

>You or Ken Johnson can define anything you want, but please understand that 
>defining something a certain way doesn't mean that others should go by that 
>definition, if they're already using a different definition.

Brian Olson replied:

Same goes for you, Mike. No one has a monopoly on definitions here.


I reply:

Be sure to let me know if I say that other are incorrect if they use an old 
term in a way different from a new definition of mine for that term.

You continued:

Of
course in the interest of productive dialogue, it would be good for us
all to have some common language.

I reply:

Obviously we had a common definition that we were using.

You continued:

I think we all understand what  a 'ranking' is and what a 'rating' is
(whatever the scale).

I reply:

But Gervase was re-defining "rating", and implying that others were 
incorrect because they were using the term differently. That's all I was 
objecting to; that's all I meant. It needn't be a fight issue.

You continued:

I think it's reasonable to claim that straight Cardinal Rankings (no
shifting, no scaling, simple summing) is equivalent
(psychological/anthropological effects aside) no matter what the scale.

I reply:

Rankings are ordinal. You mean cardinal ratings.  I didn't object to the 
re-scaling for different permitted minimum & maximum ratings. The ratings 
are equivalent even if the permitted minimum rating is negative. What I was 
saying wasn' t that the +/- rating system isn't equivalent to other systems, 
only that a +/- system makes it less obvious where, percentage-wise, in the 
permitted range, a voter rates a candidate. The midpoint of a +/- rating 
system, its 0 or origin, doesn't really have any meaning, and it only 
obfuscates the ratings. That's all I was saying.

You continued:

Some rated systems behave differently if signed numbers are used or if
positive-only numbers are used.

I reply:

I dare say that two methods that use the same balloting, but a different 
count method can certainly behave differently.

But CR behaves the same if the ratings are linearly transformed to a system 
with different minimum and maximum ratings, even if the minimum is negative.

You continued:

OK so far?

I reply:

So far, so good.

Mike Ossipoff

_________________________________________________________________
Watch LIVE baseball games on your computer with MLB.TV, included with MSN 
Premium! http://join.msn.click-url.com/go/onm00200439ave/direct/01/