[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Exactness

This page is part of the web mail archives of SRFI 77 from before July 7th, 2015. The new archives for SRFI 77 contain all messages, not just those from before July 7th, 2015.

Marcin 'Qrczak' Kowalczyk <qrczak@xxxxxxxxxx> writes:

> I already listed them among the 10 properties. 

Most of what you gave weren't requirements you think we should have,
but questions you were asking about implementation.

> Of course most numeric
> programs don't need all of them, but I don't want to care about each
> aspect separately each time. 

Of course, this kind of tedium is what computers are for.  Care about
it once, and then make one declaration around your whole programe
(with-my-numeric-conventions ....)

> I trust IEEE designers, I know that
> common pitfalls are taught, that methods for estimating errors are
> known, that the knowledge about which algorithms are numerically
> stable exists independently of the language, and that people who
> implement numeric algorithms know how to apply floating point to them.
> I don't want each language to introduce a distinct set of pitfalls.
> It's a too delicate territory to be creative without a good reason.

Great!  So trust IEEE, and wrap your program with

(with-ieee-numeric-conventions ...)

You haven't said what *wrong* with that?