[Gzz-commits] fenfire/docs/pegboard/vocabprocess--tjl peg.rst

[Tuomas J. Lukka]

CVSROOT:        /cvsroot/fenfire
Module name:    fenfire
Changes by:     Tuomas J. Lukka <address@hidden>        03/05/12 06:31:47

Modified files:
        docs/pegboard/vocabprocess--tjl: peg.rst 

Log message:
        More of the peg

CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewcvs/fenfire/fenfire/docs/pegboard/vocabprocess--tjl/peg.rst.diff?tr1=1.3&tr2=1.4&r1=text&r2=text

Patches:
Index: fenfire/docs/pegboard/vocabprocess--tjl/peg.rst
diff -u fenfire/docs/pegboard/vocabprocess--tjl/peg.rst:1.3 
fenfire/docs/pegboard/vocabprocess--tjl/peg.rst:1.4
--- fenfire/docs/pegboard/vocabprocess--tjl/peg.rst:1.3 Mon May 12 00:54:27 2003
+++ fenfire/docs/pegboard/vocabprocess--tjl/peg.rst     Mon May 12 06:31:47 2003
@@ -3,8 +3,8 @@
 =============================================================
 
 :Author:   Tuomas J. Lukka
-:Last-Modified: $Date: 2003/05/12 04:54:27 $
-:Revision: $Revision: 1.3 $
+:Last-Modified: $Date: 2003/05/12 10:31:47 $
+:Revision: $Revision: 1.4 $
 :Status:   Current
 
 It seems that vocabularies are easy to create but difficult
@@ -55,8 +55,6 @@
 
     Also, if it's spatial, it should also have z ;)
 
-- Should we merge PP (directed links) into FF? 
-
 - What should be the namespace prefix? ``vocabulary``,
   ``rdf`` or ``terms`` or ``rdfv``?
 
@@ -83,14 +81,49 @@
     >
     > where ``id`` is chosen by the person creating the namespace.
 
+- Should we separate cLink from FF? A separate CLink namespace?
+
+    RESOLVED: Yes. It is sufficiently orthogonal to other structures.
+
+- Should we merge PP (directed links) into FF, or something else? 
+
+    RESOLUTION: Move to DLINK, Analogous to CLink and Canvas2D, 
+    this property creates its own orthogonal substructure. 
+
+- What should the types of coordx, coordy be? Float? Int? Numeric?
+
+- Should each vocabulary also define its "identity", a RDF node that 
+  is used to talk about that vocabulary. This could be useful for describing
+  the schema of some data, e.g. "This is spatial canvas data with dLinks".
+
+- Where should the RDF schemas the objects be defined?
 
 XXX Change changes to retained, + javadoc
 
 Changes
 =======
 
-General
--------
+Goals of the structure
+----------------------
+
+It would be Really Nice if we could structure the vocabularies so
+that each class contains a small, self-contained structural universe.
+This means that the code implementing access using the vocab. of a certain
+class should be relatively independent of the other vocabularies.
+
+For instance, a spatial canvas is a reasonable unit: there is a canvas, it 
contains
+certain nodes at certain locations. However, the PP links (now dLinks) or xu 
links (now CLinks)
+between different canvases do not actually belong in the same place; they are 
orthogonal
+to the spatial structure.
+
+Currently, our code is pretty well along this structure: CanvasView2D takes
+care of the spatial canvas, and PPConnector (name needs to change) of the 
dLinks.
+
+The more independent we can make the codes, the easier it will be to slot in 
new
+behaviours.
+
+Overall changes 
+---------------
 
 Create new package, ``org.fenfire.vocab.lava``.
 
@@ -108,8 +141,8 @@
 definitions there, in their javadocs. There shall be no members
 or classes without good documentation.
 
-Vocabularies
-------------
+Vocabulary changes, prior to freezing
+-------------------------------------
 
 ALPH
 """"
@@ -123,13 +156,47 @@
 
 Then, we have left ``xuLinkFrom``, ``xuLinkTo``, ``xuLinkType``.
 We should probably avoid 'xu' in the permanent names,
-just in case. These should be moved to FF.CLink, FF.cLinkFrom,
-FF.cLinkTo for clink, "content link", a term Ted at some point used.
+just in case. These should be moved to CLINK.CLink, CLINK.cLinkFrom,
+CLINK.cLinkTo for clink, "content link", a term Ted at some point used.
 
 FF
 ""
 
-Retain.
+Retain. Javadocs::
+
+    /** RDF Vocabulary of central concepts of Fenfire.
+     */
+    public class FF {
+       /** A property signifying fluid media "content" of a node.
+        * Used as  (node, FF.content, literal) where the literal is
+        * an enfilade
+        * parseable by alph.
+        * This is analogous to spreadsheet or zzStructure cell contents.
+        */
+       static public Object content;
+    }
+
+CLINK
+"""""
+
+A new namespace for Xanalogical content links.
+Javadoc::
+
+    /** RDF Vocabulary of content links.
+     */
+    public class FF {
+       /** The RDF type for content links. An node which is a content link
+        * must have both the cLinkFrom and cLinkTo properties.
+        */
+       static public Object CLink;
+       /** The Alph-parseable enfilade literal of a content link from-end.
+        */
+       static public Object cLinkFrom;
+       /** The Alph-parseable enfilade literal of a content link to-end.
+        */
+       static public Object cLinkTo;
+    }
+
 
 PAPER, SPATIAL
 """"""""""""""
@@ -139,17 +206,71 @@
 
 Rename coordX, coordY to x, y.
 
+Javadoc::
+
+    /** RDF Vocabulary of 2D spatial canvases.
+     */
+    public class CANVAS2D {
+       /** The RDF type of spatial 2D canvases.
+        * Canvases contain (with the "contains" property
+        * nodes, which shall have the "x" and "y" properties.
+        */
+       static public Object Canvas2D;
+       /** The property by which the canvas is connected to
+        * the nodes, as (canvas, contains, node).
+        */
+       static public Object contains;
+       /** The x and y coordinates of a node on a canvas.
+        * (node, x, literal), where the literal is parseable
+        * as an integer (XXX).
+        * Note that these are the <em>default</em> coordinate
+        * properties: later on, we might make it possible for a Canvas2D
+        * to define its own coordinate attributes, which would take
+        * use close to Ted's floating world ideas.
+        */
+       static public Object x, y;
+    }
 
 PP
 ""
 
-Remove everything except association.
+Remove everything except association. Move association to DLINK.
+Remove class. PP is really a special-case user interface for a subset of the 
full
+fenfire structure, so it's quite reasonable not to include a special 
vocabulary for it.
+
+DLINK
+"""""
+
+New class for what used to be called PP connections (directed, typeless 
connections). 
+Name comes from "Directed LINK".
+Javadoc::
+
+    /** RDF vocabulary for directed (so far typeless)
+     * one-to-one links.
+     */
+    public class DLINK {
+       /** The directed link association.
+        * A and B are linked by the tuple (A, DLINK.dLink, B)
+        */
+       static public Object dLink;
+    }
 
 
 RDF
 """
 
-Leave as is, javadoc properly.
+Leave as is, javadoc properly. Javadoc::
+
+    /** RDF vocabulary of central RDF URIs defined outside fenfire.
+     */
+    public class RDF {
+
+       /** The RDF type attribute. A node's type can be declared to be Foo 
+        * by a triple * (node, RDF.type, Foo).
+        */
+       static public Object type;
+    }
+
 
 RST
 """