Date: Fri, 21 Jun 2024 08:45:09 +0200
Hi,
> On 21 Jun 2024, at 00:35, Tiago Freire via Std-Proposals <std-proposals_at_[hidden]> wrote:
>
> They are not SI units, science and engineers mostly don’t work with them in practice, they just convert everything to kelvin to do the math and their problems are over.
As soon as you expand this to finance, a main user of strong numerical types / unit libraries, you’ll see that most yield and risk metrics fall under that category. Yet, we need to accomodate them. It’s the automation that needs to adjust to conventions (and sometimes even convenience!) of the user, not the other way around.
Also, never forget rounding. Even C and K might differ in that for very large / small numbers, but esspecially F / K. Either you keep the unit the user specified, or introduce things like decimal(8,2) a’la COBOL (or both).
Bests,
-lorro
> On 21 Jun 2024, at 00:35, Tiago Freire via Std-Proposals <std-proposals_at_[hidden]> wrote:
>
> They are not SI units, science and engineers mostly don’t work with them in practice, they just convert everything to kelvin to do the math and their problems are over.
As soon as you expand this to finance, a main user of strong numerical types / unit libraries, you’ll see that most yield and risk metrics fall under that category. Yet, we need to accomodate them. It’s the automation that needs to adjust to conventions (and sometimes even convenience!) of the user, not the other way around.
Also, never forget rounding. Even C and K might differ in that for very large / small numbers, but esspecially F / K. Either you keep the unit the user specified, or introduce things like decimal(8,2) a’la COBOL (or both).
Bests,
-lorro
Received on 2024-06-21 06:45:26