Stream: implementers
Topic: Significant figures
Harrison Tarr (Jun 30 2020 at 21:54):
I have a question/note about the FHIR definition of significant figures. In the examples of number search, specifically for the first and third examples (Values that equal 100, to 3 significant figures precision, so this is actually searching for values in the range [99.5 ... 100.5)) and (Values that equal 100, to 1 significant figures precision, so this is actually searching for values in the range [95 ... 105)) seem to be contrary to the definition of significant figures.
For #1, In my experience, leading zeroes in a decimal and trailing zeroes in an integer do not count as significant (ie 100 is 1 significant figure, not 3)
For #3, all the other examples of sig figs in the FHIR Number Search examples seem to take the sig figs, add one significant digit of 5 in either direction. However, the third example adds two significant digits and then 5. (ie i think the range should be [50, 150) )
Lloyd McKenzie (Jul 01 2020 at 14:45):
@Grahame Grieve
Grahame Grieve (Jul 03 2020 at 06:30):
so:
For #1, In my experience, leading zeroes in a decimal and trailing zeroes in an integer do not count as significant (ie 100 is 1 significant figure, not 3)
The trouble is that 100 is ambiguous - are the zero's significant? After spritely argument about this, we fixed to say that they are
#3. I think you're right, and it's supposed to say 1.0e2. Brain fart on my part when I wrote that. Can you create a task to fix that?
Last updated: Apr 12 2022 at 19:14 UTC
