A very good question, Saaha. To look at it more generally, the question is, if a number is said to have _n_ significant figures, what do nonzero digits after the first _n_ mean, if anything? It seems like they should mean something! So, for example, 1920P2 should _not_ mean the same thing as 1950P2. But I don't recall ever seeing any discussion of this question. One reason is surely that an explicit statement of how many significant figures a value has is rare. You almost always have to infer the number of significant figures from the value itself. So we'd assume 1920 has three significant figures -- but if we're told it has only two, then what?
On Nov 6, 2015, at 6:40 AM, Saaha Metsärantala <[log in to unmask]> wrote:
>> Saaha asks: "What would "1950P2" have as boundaries: (1900 and 1999) or (1950 and 2050)?"
>> Based on the example in the draft: "y17101e4p3 Some year between 171000000 and 171999999, estimated to be 171010000 ('p3' indicates a precision of 3 significant digits.)"
>> So '1950P2' means "a date estimated to be 1950, with two significant digits".
> Thanks for this clarification. But then, What would "1900P2" have as boundaries: (1900 and 1999) or (1850 and 1950)? My point is: How much does the semantics of "P" overlap the semantics of "X"? Is 19XX a short form for 1950P2 or for 1900P2? Does 1920P2 mean the same thing as 1950P2 and 19XX?
Woodrow Wilson Indiana Teaching Fellow
Adjunct Associate Professor of Informatics
Visiting Scientist, Research Technologies
Indiana University Bloomington