
Boost : 
From: jhrwalter (walter_at_[hidden])
Date: 20020124 05:38:03
 In boost_at_y..., Jeremy Siek <jsiek_at_c...> wrote:
> On Sat, 19 Jan 2002, jhrwalter wrote:
> walter>
> walter> The 'Teubner Taschenbuch der Mathematik' defines a vector
space and a
> walter> linear operator in the following way:
> walter>
> walter> Vector space
> walter>
> walter> F is a field. G is a additive Abelian group. * is a
function: (F, G) 
> walter> > G. The following laws hold:
> walter> a) distributive
> walter> For all alpha in F, a, b in G: alpha * (a + b) = alpha * a
+ alpha * b
> walter> For all alpha, beta in F, a in G: (alpha + beta) * a =
alpha * a +
> walter> beta * a
> walter> b) associative
> walter> For all alpha, beta in F, a in G: (alpha * beta) * a =
alpha * (beta
> walter> * a)
> walter> c) identity
> walter> For all a in G: 1 * a = a
> walter>
> walter> Sorry, I do not see any right multiplication with a scalar
here.
>
> I wouldn't worry to much about precisely following the above
definitions
> because there are slighly conflicting definitions in the literature.
>
> For example, in van der Waerden's classic "Algebra" volume 1, he
refers to
> the above as a left vector space. He also defines right vector
spaces
> (with multiplication on the right) and uses just plain "vector
space" as a
> synonym for right vector space, not left. Also, he says that when
the
> field is communtative, then there is no need to distinguish between
left
> and right vector spaces, which makes sense to me.
Agreed, it's a matter of definition (and habit ;).
> Since the fields we typically deal with (like floats) are
commutative,
> then I don't see a good reason for only providing left
multiplication. It
> is just going to confuse people (like it already has) when they get
a
> compiler error. The performance pessimization in some situations is
an
> issue, but experienced programmers can deal with this and learn how
to
> order things. I'd rather make sure that things work in an intuitive
way
> for beginners.
Since I'm starting to see that there could be nonstandard
applications of matrices too, we should consider to add right
multiplication with scalars.
Regards
Joerg
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk