@rolfk wrote:
My guess about solving known bugs in respect to units is that it is all lumped together in one big Meta Bug that also includes fixing the entire LabVIEW unit system, and that is a really difficult thing to do, if at all possible. My experiments back then led me to believe, that it is basically an unsolvable problem if you talk about making units a fully featured attribute of all elements in LabVIEW.
I don't doubt that you're correct. However, as a programming tool and especially for LabVIEW as an engineering programming language, fully-implemented and working units are potentially very useful for code-time error checking, and ease of checking the implementation of a model. It's frustrating to see them there but to keep running into the issues you talk about - would it be better to remove them altogether?
@rolfk wrote:
... I abandoned units then and never looked back. And I know in fact nobody who really felt different about units then and since, and that also means it is never really tested, both in Betas and apparently not even in the LabVIEW unit tests.
I've used them perhaps half-a-dozen times, never for anything critical, and so probably should have checked them in the latest Beta but just didn't have time! I once wrote a simple model of oxygen diffusion using units, and it was very helpful to be able to check that all the components of the equations had correctly matching units, and also to be able to compute pressures in Pascals but display them for the user in mmHg. But it wouldn't be much more work to handle all that explicitly, especially as a couple of functions (convolutions) required converting out of units and back again.
So my take is that it would be great to have units working properly everywhere - for example, the Derivative function would differentiate a distance with a time and get a velocity. But if not, then they should probably be removed altogether, with a short footnote to mourn their loss, and wonder what could have been.
