[Haskell-cafe] vector-space and standard API for vectors

[Stephen Tetley]

On 30 October 2010 11:07, Henning Thielemann
<schlepptop at henning-thielemann.de> wrote:

> Looks like you are about to re-implement numeric-prelude. :-)


Ah, but Numeric-Prelude is huge though[*].

DavidA complains in the recent Cafe thread "Decoupling type classes
(e.g. Applicative)?" that the Num hierarchy can't be replaced due to
inertia. My own feeling is it can't be replaced because no-one can
define an acceptable middle ground between the two extremes:

miminal - Haskell currently (for discussion, we'll elide some of the
"problems" such as the Show superclass constriant).

huge - numeric-prelude or a "mathematics" system such as Aldor.


[*] This is not a criticism - just an observation.