Re: [Gzz] Structure proposal: RDF (+Xu)

[Alatalo Toni]

On Wed, 19 Feb 2003, Benja Fallenstein wrote:

> > to me (GUID-(contains)-enfilade). so it's very similar to RDF?
> Structurally, it's a superset of RDF (in a similar way that RDF is a
> superset of zzstructure). Of course, you can express everything in any

ok.

> > 1. catering for Xu content: why 'should probably define a type of literal'?
> We're talking not about singe spans, but enfilades (lists of spans).
(...)
> Nope, I meant a literal which is some representation of an enfilade,

ok. i still don't see all of this but must just study more, this is
probably enough basis for that -- thanks.

> >    trees a problems with zzstructure, or only with the views that did
> >    not really support working n-dimensionally, i.e. typing the links?
> It's that the fundamental zzstructure views work well with trees, but
> not very well with m:n links. You can make them, of course, but they

i seemed to be missing the point a bit in the comment -- wasn't actually
thinking of trees/not-trees, but the dimensionality: at least what i did
was mostly just 2-dimensional, was it so with you as well? and this i
thought was not good use of zz. (did Ted discuss this somewhere?). i guess
that could be accounted for the fundamental zzstructure views, as well?

with RDF we obviously wouldn't be making just two kinds of associations
(1&2). or hm, for simple note-taking a really simple structure (very few
link types, as people might not understand/use several -- Harri has made
some empirical studies on this) still might do ok. i am quite happy with
those trees and the fact that the system has the other dimensions for
other things, like clones, and that they can be used at will for anything
is good to have, even if the end-user only sees a few. e.g. when using
d.an and d.org i had d.private on z when no one else was watching :)

but this was just a sidenote anyhow, i guess. and like i said having
straightforward m:n links seems like a good idea to me.

> >    person you would always edit the name ..
> If you accurse the person. The properties are own 'nodes,' even if one

oh of course.

> >    independent, structures. missed e.g. tables with 0.6.
> You had them through the row/col views...

actually i was going to mention those -- never actually used them like
that, might try one day (i still occasionally use 0.6).

> I think structures need to be discussed with their 'fundamental' views.

ok.

> -b.

~Toni