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.

Q9 Answers

• How does one interact with cone trees?  

• For hyperbolic distortion: Can the user turn around the sphere to better see nodes?  

• Would it be useful to have two lenses (for zooming) in order to compare two regions in detail?  

• Nonlinear Magnification Fields: I am quite interested in the math background and I wonder whether figure b needs to have this trench around the plateau to be able to get the specififc magnification field? Are there methods so that there won't be any trench?  

• How hard is it to implement this stuff? I have no real idea how I could achieve these effects. In one of the previous lectures we were presented another network visualization tool that even offered GPU acceleration. How costly ( computational, memory ) are these methods in general?  

• I guess it was in the first lecture that we were looking at a visualization of former us president clintons excuse speech. I guess the DOITrees Revisited system would also be a quite fitting way to visualize a task like this ?  

• Is there a significant advantage why to use fisheye lenses in comparison to a normal magnifying glass. I mean you don't have that smooth transition of what's not important to the important data, but just speaking of an implementational view it should be much easier to just make everything bigger instead of setting the magnification factor in relation to the distance from the focus. Is it a benefit to show more of the surroundings in the lense?  

• Since this is our last reading, i'd like to use my last question to sum up some thoughts i came up with, while I was reading this book. First of all, sometimes it was quite hard to find some questions, because most of the book is more or less self-explaining. But on the other hand the writer is talking about Euclidean spaces and stuff like that without ever mentioning any mathematical explanations. Imho I think there should be more maths in it, but maybe thats just me. I really enjoyed the chapter about coloring, cause it's one of the topics I'm most interested in. It's really fascinating how people try to find models of wavelengths being reflected. As I just said, the book is quite straight foward, and not a lot of questions came up ( and i think, thats how a good scientific book should be) but to find 3 (askable) questions in every chapter, you somehow have to invent questions. so maybe it's enough if you have 2 questions, which are well-thought out. 

• 14.5. Distort: How would distortion with the help of a vector field look like?  

• wouldnt an embed system make a lot of sense in a map (e.g. google maps) especially when looking at unfamiliar places?  

• The 3D hyperbolic geometry example is really hard for me to understand, I cannot zoom into the sphere to look at higher level nodes, but rather select which are shown in what size. Is that correct?  

• Is it possible to use multiple focus + context idioms at the same time on a dataset and focus on multiple things?  

• Are there any studies about which distortion method is the easiest understandable for most of the users?  

• In fisheye distortion, is it possible to set up the lens' size manually according to the user's preferences?  

• why is the focus layer to a local region is limited in the suerimpose?  

• is there only a single region of focus, or does the idiom allow multiple foci?  

• why is there users that may not be aware of the distortion?  

• How would one actually apply the degree of interest function to a hierarchical data set?  

• The degree of interest function sounds really interesting, is there some more literature about it or something similar? Like algorithms that estimate/determine the importance of an element to the user based on certain criteria (like distance to the view port, distance to other elements, general importance of the element, etc).  

• In the book the originator of the fisheye view expresses doubts about the importance of how we display data and is of the opinion that it is more important what data we actually show. Is this an accepted design paradigm (that it is more important what is shown than how) or just an opinion?  

• The bring and go interaction is only affecting the one hop neighbors, but if the bring and go interaction isn't applied to the rest of the network isn't there a falsification of the whole visualization in the moment of the distortion ?  

• If it's a total no-go to visualize stuff in 3D ( at least if there are other ways to display that information in 2D) why would it - for example cone trees - be teached in books any longer. Are you also recommending to point out negative examples?  

• When is it good to use a fisheyelense instead of a normal magnifier?  

• If the fisheye lens is used, is there a way to show the user where the focused points are with respect to the axis? Given the background grid is hard to follow because of the distortion.  

• In 14.3 an “interest function” is mentioned; what would be an example and how are they created?  

• Would tooltips also fall under the “focus+context” idiom family?  

• Eliding and magnification/distortion increase the amount of pixels to display the details on, superimposing does not. What is the advantage to superimposing within a limited area instead of globally?