Explanation for use of Dimensionless exponent

qudt / qudt-public-repo

QUDT -Quantities, Units, Dimensions and dataTypes - public repository

Other

116 stars 73 forks source link

Explanation for use of Dimensionless exponent #270

Open dr-shorthair opened 4 years ago

dr-shorthair commented 4 years ago

I'm unable to find clear explanation of usage for the Dimensionless exponent. Should it be set to 1 if the other exponents in a dimension vector cancel out - e.g. MilliMOL-PER-MOL ? i.e. the dimension vector can never be A0E0L0I0M0H0T0D0

dr-shorthair commented 4 years ago

And what about unit:NanoMOL-PER-MicroMOL-DAY - would that be A0E0L0I0M0H0T-1D1 or A0E0L0I0M0H0T-1D0 ?

steveraysteveray commented 4 years ago

We handle the simple case of ratios as described here in our wiki. More general cases, such as equations where various units partially cancel out, will need to wait until we support a more general expression capability.

dr-shorthair commented 4 years ago

What about unit:PER-SR - would that be A0E0L0I0M0H0T0D-1 ?

(I'm trying to fill in the Dimension Vectors for that large set of units needed by NVS P06. Good progress, and a good way to do another check through. But uncovering some places where I'm a bit unsure ...)

dr-shorthair commented 4 years ago

And what about all the 'number per XXX' units - e.g. NUM-PER-L ? A0E0L-3I0M0H0T0D1 or just A0E0L-3I0M0H0T0D0?

steveraysteveray commented 3 years ago

What about unit:PER-SR - would that be A0E0L0I0M0H0T0D-1 ?

(I'm trying to fill in the Dimension Vectors for that large set of units needed by NVS P06. Good progress, and a good way to do another check through. But uncovering some places where I'm a bit unsure ...)

--

I would say this one would be A0E0L0I0M0H0T0D1. In fact, I think any vector having D-1 is an error. I see that there are 4 such instances, but nothing points to them.

I will give this some time for responses, but then I plan to delete the 4 D-1 vectors. Also, I'd like some feedback on some creative use of "D1" vectors that I defined back at the beginning of COVID-19, specifically http://qudt.org/vocab/quantitykind/MorbidityRate, and http://qudt.org/vocab/quantitykind/MortalityRate

I'm not entirely convinced, myself, that this is a good way to use the numerator and denominator vectors. It might just be better to just have the regular dimension vector and not have the numerator and denominator ones...

steveraysteveray commented 3 years ago

And what about all the 'number per XXX' units - e.g. NUM-PER-L ? A0E0L-3I0M0H0T0D1 or just A0E0L-3I0M0H0T0D0?

I believe it should be the second one, L-3 and D0 in your example.

dr-shorthair commented 3 years ago

OK - so to confirm:

D can be either 0 or 1 only
if D=1 then everything else is 0, and if [AELIMHT]≠0 then D=0

I'll fix up #268 to conform

steveraysteveray commented 3 years ago

Basically, yes. ...plus the possible "creative use" I mentioned above. But again, I'm not sure that was a good idea.