[Gzz-commits] manuscripts/xupdf article.rst

[Tuomas J. Lukka]

CVSROOT:        /cvsroot/gzz
Module name:    manuscripts
Changes by:     Tuomas J. Lukka <address@hidden>        03/02/14 12:08:03

Modified files:
        xupdf          : article.rst 

Log message:
        Buoy leftright

CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewcvs/gzz/manuscripts/xupdf/article.rst.diff?tr1=1.105&tr2=1.106&r1=text&r2=text

Patches:
Index: manuscripts/xupdf/article.rst
diff -u manuscripts/xupdf/article.rst:1.105 manuscripts/xupdf/article.rst:1.106
--- manuscripts/xupdf/article.rst:1.105 Fri Feb 14 11:56:59 2003
+++ manuscripts/xupdf/article.rst       Fri Feb 14 12:08:03 2003
@@ -229,6 +229,9 @@
   to go back, without a "back button" [implies bidirectional links] 
   "once you get there, you should see how you can get back".
 
+There is one scrollable focus node ("current document"),
+and links 
+
 To be able to show all the link targets near the focus,
 only the *relevant fragments* (the immediate surroundings of the other
 end of the link) of the target nodes can be shown.
@@ -291,39 +294,43 @@
 - the view should animate continuously when the focus moves
 - the user should be able to understand and predict the motion
   of the buoys.
+- there should be little or no hysteresis
 - buoys should be placed close to their anchors
+- time coherence: when traversing a link, i.e. animating a buoy into focus,
+  the former focused node should have a clear relationship.
 
-Furthermore,
-it is important to maintain orientation locally,
-that is, going in one direction should result in arriving from 
-the opposite direction, so that the local spatial structure
-does not change too much.
+The apparent conflict between "no hysteresis" and "time coherence"
+can be resolved by maintaining a local spatial structure
+by giving each link a specific left-right orientation 
+so that a right-end node and left-end node retain their relative
+locations w.r.t. to the link
+independent of either of them being the main node.
 This matches the way the brain understands space
 as globally distorted, segmented, locally Euclidian views
 (see, e.g. [wang01mindsviews]_).
 
+..  it is important to maintain orientation locally,
+    that is, going in one direction should result in arriving from 
+    the opposite direction, so that the local spatial structure
+    does not change too much.
+
 .. http://acad88.sahs.uth.tmc.edu/research/publications/iccs2001spatial.pdf
 .. http://citeseer.nj.nec.com/hunt99orientation.html
 
-Because of this, each link has a specific left-right orientation 
-so that a right-end buoy is drawn on the right side of the view
-and a left-end buoy on the left side.
-
 More than two opposing directions could also be used, but
 is is not required for the local spatial coherence and it would also 
 limit the layout of a large number of buoys.
 Furthermore, there is usually no meaningful global 2D layout for 
 a network of nodes, so the extra directions would not help much
 in perceiving the position in the global structure.
-In our application, the structure has simply directed links, 
-visualized as the left-right direction of the connection lines.
 
-As for the horizontal orientation of the link direction, 
-it is more natural than vertial, 
-because the visual field is wider than it is tall
-(as exemplified by the usual screen aspect ratios of 4:3 and 16:9) 
-leaving more space at the left and right sides than on the top and bottom.
-XXX: left and right hand???
+Orienting the link direction horizontally
+rather than vertically is more natural 
+because the visual field is wider than it is tall ---
+consider the usual screen aspect ratios of 4:3 and 16:9.
+
+..  leaving more space at the left and right sides than on the top and bottom.
+    XXX: left and right hand???
 
 From the above principles, we have chosen a simple geometry
 depicted in Fig. [ref-figbuoygeom]_.