The domains of each verb is key here, but
whose descriptions are non-existent.
Otherwise meanings of primitive trains
are too abstract to make sense.
E.g. not x verb y,
but (Nx4 index matrix) verb (list of ...)
or
Univ=. Univ verb Choices
with legend table
Univ N,4$ Matrix of current universe values
Choices 4>=$ Current choice
The output domains are also important, as they
make up the input domain if chained verbs.
So starting with the top verb, domains can be
deduced along the data pipeline paths.
http://www.jsoftware.com/jwiki/Essays/Tacit_Expressions
Also helpful is execution/data flow graph with
major operations on vertices and data on arcs.
http://www.jsoftware.com/jwiki/Addons/stats/dendrite
Conversely, the flow can be shown in commutative
diagrams as in maths with verbs arcs:
http://www.jmilne.org/not/ (Guide to commutative diagram packages)
http://www.ctan.org/tex-archive/macros/generic/diagrams/taylor/
Noun "aliases" can be semantical or structural collected
in a common table of name, structure (shape), description.
Note, the "moo" essay has such table for verbs, but
no such table for noun-domains.
With proper approach this excercise should be
literate reading of J sentences as opposed to
reverse engineering.
--- "Sherlock, Ric" <[EMAIL PROTECTED]> wrote:
> ---Raul Miller wrote:
> > On 1/8/08, Björn Helgason wrote:
> > > It would be really interesting to see Rogers moo
> > functionality explained
> > <SNIP>
> >
> > Hmm... I do not remember this code, and do not remember
> > its purpose, nor its domain. And, purpose and domain are
> > crucial for understanding code.
> <SNIP>
> >
> > Anyways, I am reluctant to spend more time reverse engineering
> > this program, I would prefer a few senteces of documentation.
> >
>
> Roger's essay http://www.jsoftware.com/jwiki/Essays/moo is a start.
____________________________________________________________________________________
Never miss a thing. Make Yahoo your home page.
http://www.yahoo.com/r/hs
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
