Please disable Adblockers and enable JavaScript for domain CEWebS.cs.univie.ac.at! We have NO ADS, but they may interfere with some of our course material.

Q5 Answers

• Whats the best way to order streamgraphs?  

• How does matrix reordering work? Which algorithms are used here? How is the order decided?  

• How can a good order of the axis for parallel coordinates be found?  

• How usable are radial bar charts?  

• How do counter trees work and how are they red correctly?  

• What is meant by 'size-coded' in scatterplots?  

• I don't really get that space-filling, it says: "…designer cannot make use of white space in the layout. " But isn't space-filling all about, filling the white spaces?  

• Would you use Parallel Coordinates or SPLOM? Which one is easier to realize?  

• Are there good ways to visualize time-oriented vector fields?  

• Streamlines and ellipsoid tensor glyphs. Which one would be better suited to show multiple tensor fields in the same view?  

• How should one compare tensor fields?  

• Munzner Chapter 7 - Streamgraphs: I don't see any point in using it, I find it hard to read, especially if you have much data. Is there a good way to get information out of it?  

• 7.5.: Separate, Order, and Align: Categorical Regions. when the categorical data has an order for some users but now for others, how do you decide whether to use that or not(e.g. colors), as it may confuse those who havent learned that specific order? 

• 7.6.3.: Radial Layouts: when using a radial layout instead of a rectilinear one, should we consider placing close values next to each other because its even harder to differnetiate between those values when they are on different sides of the circle? 

• 8.1.: The Big Picture: For both vector and tensor fields, with multiple attributes at each cell, there are four families of encoding idioms: geometric approaches that compute derived geometry from a sparse set of seed points. I am not sure what that means when it states before that  
  the spatial channel cannot be used to express attributes?  

• Why is in the scatterplot in fig. 7.2 (and commonly) the size of the circles used as a dimension when it is hard for humans to compare sizes of circles?  

• In fig 7.9 there are bar charts and line charts each showing two discrete values. Isn't a line chart in this case misleading? We don't know if the 11-year-olds are like linear in between.  

• Would the best visualisation for access for memory or gates in computers in general be a cluster heatmap? 

• How can be the most efficent Ordering mechanisms can be find out. Like in Barchart 7.4 what makes b for Users Perspektive the most efficent way than a?  

• Is Matrix Allignment available for every kind of Charts?  

• Is Spatial navigation not possible while isovalues scannt or does it require full dataset to function properly?  

• Eine Frage die mir kam, zum Thema, Barchart zeigt eine Dimension, Scatterplot zeigt 2 Dimensionen, gibt es sinnhafte Ideen 3D Diagramme zu machen, ich könnte mir vorstellen, dass man sowas vielleicht auch wenn noch statisch mit 3D-Druck realisieren kann, ist zwar überhaupt nicht kompatibel mit variablen Daten und es besteht keine Möglichkeit zur manipulation, aber man kann vielleicht leichter mit einem Objekt umgehen als mit einer 2D Simulation.  

• Was für einen Vorteil bietet das Radial Bar Chart? In meinen Augen wirkt es eher unpraktisch, da man einzelne Bars nicht mehr so leicht vergleichen kann.  

• In Kapitel 7.7.1 wird beschrieben, wie Sourcecode eingefärbt wird je nachdem wie oft ein Test darauf zugreift, oder erfolgreich ist, kann man sich Dense generell so vorstellen, dass man einen Text hat und diesen analysiert, zum Beispiel mit: zeichne mir alle "e" rot und alle "i" gelb, oder braucht man, eben wie beim Dense Software Overviews ein komplexeres Datenset im Hintergrund?  

• Ich persönlich merke ich kein Unterschied zwischen ein Adjazenzmatrix und ein cluster heatmap. Sind die beide gleiche Sachen, oder gibt es schon ein Untershied, welche ich nicht merke?