[ExI] Neither unique facts nor lies

[Mike Dougherty]

On Mon, Aug 15, 2011 at 11:26 AM, David Lubkin <lubkin at unreasonable.com> wrote:
> Unique is also considered binary, and "very unique" or "more
> unique" is mocked. I have no problem with treating unique as
> a fuzzy value. That is, instead of either 0 (false) or 1 (true), it
> can have any value from 0 to 1, inclusive.

If you have only seen 1 example of something should it be called rare
rather than unique?  Unless you have examined the complete universe of
things, then you have to conclude that you do not have enough
information to declare something unique.  Rare seems to have some
subjective measure such that the countable number of items in the
universe of rare is less than those that are common but more than
those that are extremely rare or candidate-unique.  I think the fuzzy
value you attribute to unique is a measure of confidence regarding the
yet-unexamined universe.

> Given the sequence (A, B, C), if I refer to the former, do I
> mean (A) or (A B), and does the latter mean (B C) or just (C)?

I would suggest "former" requires some reference to the current item.
If you have been discussing the ordered sequence A,B,C and you have
been talking about B then the former could be understood as A.  I
would also understand if you were talking about C that the "former"
would refer to {current minus one} or {immediately prior}.  While most
people would probably correct the reference from further context if
you made an alternate use of "former"  (discussing C then using
"former" as referent to A)  The parse rules for this kind of reference
would be needlessly complicated.  I would suggest "latter" be the
next-in-sequence opposite to "former."