[seqfan] Re: A024812

[Maximilian Hasler]

[off list]

On Wed, Mar 20, 2013 at 2:27 AM, Charles Greathouse
<charles.greathouse at case.edu> wrote:
> Brackets are already much too overloaded:

Gauss defined the notation [...] *ALREADY* centuries before any of the
other cited meanings.

> grouping symbols, floor,

yes

> round-to-nearest,

never saw s/o use [] for that.

> Iverson,

never understood why (a=b) would not be as good as [a=b]

> function invocation

using [.] ?
You mean that absurd Mathematica-only aberrance ?
does not count.

>, etc.,

so far  I admit only the use for "grouping" as 1 alternate uses,
but this is common to (), {}, []....

> alongside their
> notionally distinct use in matrices, Stirling numbers of the first kind,
> commutators, (half-)closed intervals, and even (half-)open intervals in

closed, you mean... ;-)

> Bourbaki. The half-brackets are much better for floor than brackets.

But no-one has them on the keyboard.

> In their absence I prefer floor(...) to [...] because of the multiplicity of
> alternate meanings.

I know, that's why my battle is lost in advance, and also why I prefix
it with the idempotent "floor" operator when better notation is not
available.
(none of your arguments makes floor(x) any better than floor[x] ...)

[No offense intended.]
Best wishes,
Maximilian