Re: [zzdev] Re: [zzdev] Re: [zzdev] Re: [zzdev] Re: [zzdev] More about dimensions (lists)

[Tuomas Lukka <lukka@xxxxxxxxxxx>]

> Hm, are there any plans to introduce (generic) lazy structures in ZZ?
> As in, assign to the current cell a procedure that produces a ZZ structure
> starting from current cell poswards on d.foo; then move your cursor
> poswards on d.foo, and the procedure is run until it produces a cell
> there (with the new cell possibly containing a procedure to generate more
> cells as needed); and as you traverse the structure, more of it will be
> generated on demand? (just a crazy thought, but it could be useful in
> modeling countably infinite structures)

At the moment, no concrete plans until you make them ;) ;) ;)
 
> > In a sense, then, wouldn't the proposal of explicificating many-to-one
> > relations go agains this sense, too: the ZZ structure is a simple and
> > powerful foundation; does it really *need* the standardized many-to-one 
> > thing?
> 
> Then again, *if* it needs a many-to-one thing, a standardized one would
> probably be a good thing.

Well, thereare the standardized schemas:

1. rank, with special headcell (like clone)
2. corner list, with empty headcel
3. corner list with no special headcell, possibly looping.

> > > How do d.1, d.2 and d.3 have any meaning at all?  Why not call them d.foo,
> > > d.bar and d.baz?  They'd be about as descriptive.
> > 
> > But longer and harder to remember for non-CSers. 1-2-3 is pretty easy
> > to live with.
> 
> I was trying to make a point.  I have difficulty understanding why we
> need d.1, d.2 and d.3.  Why not use some dimensions with names that *say*
> something about the purpose of the dimensions?  Such as d.list etc.

All dimensions are "d.list" in that sense. See Benjamin's email for a 
very accurate interpretation of those dimensions. 

> > Oops, I was unclear. By "free" I was referring to the three visible
> > dimensions: the more of them somehing uses, the less easy navigation becomes,
> > like Benjamin was saying.
> 
> Hm, BTW, couldn't we have a fourth dimension going from UR corner to
> LL corner?

Ted mailed the spec for having five at some point; on the task list - want
to have a go?

	Tuomas