Zhongshan Zhu
I have completed the graded portion BEFORE looking at anyone else's work on this assignment. Zhongshan Zhu.
1) Good Visualization
This is an example of a computational fluid dynamics visualization. This visualization uses several perspective properties including 3D, shapes, and color. The combination of 3D and color work together particularly well in this visualization. The combination of two properties can help us clearly visualize pressure (and air stream velocity) on the surfaces of the 3D model (of the race car). Color is perhaps the best option to represent the continuous data sets (pressure/velocity). The decision to make red be the highest (pressure/velocity) and blue be the lowest (pressure/velocity) is quite logical as we generally associate red with high and blue with low. The 3D model allows us to view multiple surfaces of the race car at once. This is quite useful as it allows us to view a large amount of data at once. In the actual simulation model you are able to rotate the visual in various directions allowing you even more control over the visualization. The wind stream lines not only demonstrate the the velocity of the air flow, but also help show the different textural features of the race car. Overall, This visualization does a good job of of representing fluid dynamics (in this case applied to wind streams on a race car).
2) bad visualization #1
apple/swift repos (repurchase agreements) report
https://rocketgraph.com/s/6WNiog_X8un
Though this site visualization tried to use 3D to their advantage it just didn't work out. First of all, the visualization didn't really need to be 3D. The visualization decided to plot the months and days in the months on different axis when the the two could easily have been combined on one axis. The 3D bars show the number of commits on a particular day. This value is hard to compare. Not only is there not a visible axis for the number of commits, but the large amount of bars blocking each other's ways makes it hard to compare short bars with long bars. Even the color scheme makes it hard to perceive differences when it comes to similar value items. Overall, this visualization seems to only be able to give people a sense of which months have the highest productivity rather than being able to show all the specific points.
bad visualization #2
http://www.weforum.org/agenda/2016/02/how-safe-are-robot-drivers?utm_content=buffer734b2&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer
In this visualization the creators of the visualization seemed to be aiming to show the percentage of the whole. However, they chose a poor way to represent this. It is very difficult for humans to compare values when a pie chart is used ( in this case a variation of a pie chart). In this case, it it would almost be impossible for us to perceive percentage of the whole or compare the values if they didn't give the data figures to use. It is generally quite hard for us to compare items in pie charts (especially this curved shapes variation of a pie chart; the regular slice pie chart can at least show percentage of the whole better). Positives of this visualization would be that it's able to accurately show area and that it chose two distinctly different colors to represent the two options. For this particular visualization, I would recommend a stacked bar chart. Overall, this visualization is just really difficult to read for humans. The shapes make it difficult to compare and make the data hard to read.
3)
I actually do like these botanical models. They take advantage of 3D to help allow us to get a sense of a large amount of data at once. The figure also uses certain perceptive aspects such as color, shapes, and zoom. Though this may be confusing at first, the visualization should be understandable with a proper key. Using colors on the branches to differentiate between different levels of the hierarchy is a good use of color to distinguish different aspects of the visualization. The idea of using different shapes on branches and on the phi balls (ex. shapes on the phi ball to represent files) does seem to work. There is a distinct rule to every shape and color. The ability to zoom in and view distinct aspects of the visualization (for example a phi-ball so that you can count the files on it) can be extremely helpful. One thing that comes to mind is the aspect of red/blue or red/green color blindness when looking at this visualization. This could be solved by a simple change in the color scheme. Perhaps a major negative to this visualization is that it is difficult to just take 1 look at the figure and remember everything about the structure. There seems to be just a huge amount of information in one place. I do however, appreciate that there is an attempt to show this massive amount of information at once. Overall, I think this visualization can be quite useful for looking at the structure of massive systems. These are very accurate models that can have pieces broken down and analyzed. I think that botanical models do a good job of taking advantage of perceptive properties in order to convey data.