On Tue, 22 May 2001, Tim Peters wrote:
> [Michel Pelletier, on http://www.lahey.com/float.htm]
> > I liked the tone too, but it really goes into a lot of detail, there's
> > this problem, and that one, oh and also *this* one and then there's
> > *that* and the other thing, and after a while you get the impression
> > that floating-point is for the insane.
>
> Using an unfamiliar power tool with sharp edges, and while blindfolded, is
> insane.
I should have been more clear, I liked the first couple of paragraphs for
their descriptions, and there is certainly nothing wrong with the document
as it stands, but such an explanation would be a bit too lengthly and
boring to a typical fifth grader or photoshop guru going through the
Tutorial and dabbling in programming for the very first time.
> [and on http://www.python.org/cgi-bin/moinmoin/RepresentationError]
>
> > I agree. Equations should not be needed to explain this.
>
> There's exactly one equation on that page, saying that one ratio of two
> integers is approximately equal to another ratio of two integers.
Who was it that said every equation will halve your audience? I agree
with that, the tutorial should try to be as broad and simple as possible.
> If that's
> too much for you, and you weren't satisfied with the *initial* hand-wavy
> explanation ("1/10 is not exactly representable as a binary fraction")
> either, then it's up to you to do better than the latter without actually
> saying anything useful <wink>:
The latter is fine, although I think the first document hand-waves better.
-Michel
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4