On Tue, 2 Aug 2011 14:52:46 -0400, Ray Denenberg, Library of Congress wrote
> Ok, I think I have a solution to all of this.
>
> 'u' clearly has a much different usage when used internally than when it
> replaces one or more consecutive rightmost digits. I don't think we have
yet
> confronted this.

The semantics for u restricted to application as one or more consecutive
rightmost digits can have but need not have the semantics as a placeholder
but does not do well to express measurement precision.

199u where u is a placeholder has unknown year but expresses a date with a
precision of year. 199x by contrast has a precision of decade. The u as a
rightmost placeholder can be used to express precision via a semantics of
uncertainty but then it would not be a placeholder..

This is, I think, a problem with our current specification for u and not
really x.

The solution, I think, is to call u a digit placeholder:
unknown/unconfirmed/untested/unspecified/unmeasured ....

Its use and semantics are simple. One can use u in place of a digit anywhere
in a date/time expression.

>
> Thus 196u means:
> "a date in the 1960s"

Yes. But with year precision... what year? We reserve the digit place.. The
whole point of u is, I think, the explicit intent "I hope to fill in the
blanks".

> and expresses imprecision, where 1u66 might be said to express uncertainty.

I think we should avoid the term "uncertainty" with u and keep to
placeholder.. Uncertainty to century with year 66 is I think something
else.. Instead its saying, I think, I know its year 66 and second 1000
years.. I need to check the century and so I'll leave it as u and come back
sometime to fill it in.. That I always thought was the reason we got them in
MARC.. In MARC the u is called "unknown" but its not really unknown.. only
not available at the moment the record is authored.. Its saying "to be
filled in later"... Precision is not about filling in later.. A sample that
weights 2 grams on a balance that reads only to gram is a weight of 2 grams
with gramm precision.. I might never have call to measure it with a more
precise balance..

>
> In other words I suspect the solution to this problem is not to refer to
> "rightmost" 'u' as "uncertainly" and instead refer to it as "imprecision".
>
> And since 'u' means different things when used internally vs. rightmost,
and
> perhaps 'u' shouldn't be used for both.
>
> I therefore propose that 'x' be used instead of 'u' for level 1
uncertainty,
> and that we rename "uncertainly" as "imprecision" for level 1 (and continue
> to call it 'uncertainty' for level 2 - internal).
>
> --Ray

--

Edward C. Zimmermann, NONMONOTONIC LAB
http://www.nonmonotonic.net
Umsatz-St-ID: DE130492967