Re: [Gzz] REPOST: PEG: Alter Canvas2D vocab to support transclusions

[Benja Fallenstein]

Asko Soukka wrote:
> Hi,
>
> Sorry, that I'm a bit late on that conversation. So, the structure after 
> Benja's PEG would look something like:
>
>              1:n                                 1:1
> (Canvas) --contains-- (NodeOnCanvas)(x)(y) --transcludes-- (Node)

No, no, the second cardinality is n:1! There can be an arbitrary number 
of NodeOnCanvas (maybe better called Coordinates) resources for each 
individual Node!

> Basically, we would have made connection into a node. Did I understand the
> PEG right?

Yes.

> As Tuomas told me, this could be done alternatively using reification. The 
> term is still quite abstract for me (though, tried to learn it), so, 
> probably you can tell me how it works :)

That's not a good idea for this, but FWIW, it would be like this--

reification --subject--> Canvas
reification --property--> contains
reification --object--> Node
reification --type--> Statement

I.e., the "reification" resource represents the statement which has 
'Canvas' as the subject; 'contains' as the property; and 'Node' as the 
object.

> In my MindMap vision, connections between nodes and canvases could be 
> shown/created/removed also on 2D structure view (Loom alike). Of course, 
> the view should abstract the structure in these both solutions...

Yes.

I don't like the term "2D structure view" btw: it's not a 2-dimensional 
plane, although it is shown on a 2-dimensional screen (which is IMHO the 
only sense it is '2D' -- in that sense, Quake is '2D' also...).

- Benja