Visual Information Design
From CS160 User Interfaces Sp09
Lecture on Apr 1, 2009
Readings
- Information Visualization. Readings in Information Visualization. Chap 1. Card, Mackinlay, Schneiderman.
Jeffrey Patzer - Mar 29, 2009 10:30:59 pm
I think that if I had to boil down this reading to its main point, that it would be this: organize the feedback of the system state and information it holds in a useful diagram that allows for effective viewing and understanding. This reading went into great detail about how humans understand visual diagrams of complex information. These diagrams provide a way to reduce search time, enhance pattern recognition, offer more resources, and perceptual inference. While diagrams can do this, the article cautioned that bad diagrams can offer none of those things and can slow the viewer down quite a bit. This reading seemed to combine the idea of direct manipulation interfaces and feedback, emphasizing their organization into effective diagrams. The later part of the article described the process that one goes through in creating these diagrams, offering a clear outline of how to determine variables and place them in your diagram. It was good to get a rigid framework for the diagrams that were being described in such detail. Whether or not I think this article will prove useful to us I think depends on the game we are building, but since most games have a stats page, it might be applicable there.
Saung Li - Mar 30, 2009 02:53:28 am
Information visualization increases our ability to think by providing external aids. It sort of relates to the concept of direct manipulation, since raw data is transformed into a data table and mapped to a visual structure and then visually transformed, as if the user is working directly with the data. Information visualization is important in that it allows the user to understand the information presented and reduces search time through usage of the principles such as the gestalt principles. Knowledge crystallization is the process of collecting data, identifying a schema relating the data, instantiating the schema with data, reorder the data to problem solve to reorder to trade off features, searching for a new schema that reduces the problem to a simpler trade off, and put together the patterns found in the output. This process allows the user to get insight into data that otherwise may not reveal any patterns. The concept of cost structure is pretty useful in that files that are not accessed often can be put away in a harder-to-reach place to reduce the clutter in the work space, while files that are accessed often can be immediately retrieved by the user so that the user can perform tasks most efficiently. For example, a user who is working on a word doc can save the word doc to his desktop, and files that he uses occasionally can be put in a folder on his desktop. Organizing data into effective diagrams and being able to transform them helps the user perform best since it would reduce the semantic and articulatory gulfs of execution and evaluation. Information visualization may be used in the wrong way and convey the wrong message, however, so the designer should make sure that the visualization suits the user's tasks. It seems like the article has a lot of important concepts, but how do you think we can apply them to games with a purpose?
Nalditya Kusuma - Mar 31, 2009 03:53:00 am
What an awesomely long reading! To me, the most important point of the reading is understanding what Information Visualization (IV) means and is capable of doing. IV is defined as the use of computer-supported, interactive, and visual representation of abstract data to amplify cognition. The basic example is the use of paper and pencil to accelerate multiplications. The advanced example is the use of diagrams and interactive animations in visualizing huge amounts of data that satellites send back to earth, and also in helping stock market people taking important decisions under time pressure given rapidly changing data. Even though the reading mentions the 6 ways IV can help amplify cognition, I think here are the most important 4 that need to be remembered: [1] Increasing memory/processing resources for users [2] Reducing the search of information [3] Enhancing the detection of patterns [4] Encoding information in a manipulable medium
Timothy Yung - Mar 31, 2009 01:00:54 pm
The article was an interesting overview of information visualization. Since much of the article made much sense and, if condensed into one sentence, would be: a picture is worth a thousand words. I felt like this article nailed the reason why that statement is true -- that visual representations allow humans to create many inferences. Up until now, we've discussed direct manipulation and how important it is in user interfaces, and I think the key concept from this article is that text is not necessarily present in the best representations.
David Burban - Mar 31, 2009 07:51:49 pm
This article is all about visualization, and the reasons to use visualization in favor of text. The article highlights the good and bad uses of visualization, such as blaming poor diagrams for the failure of space shuttle Challenger; by appropriately choosing axis to highlight the failure data, a much clearer picture could have been displayed. The article then goes not to talk about cognition, and how visualization helps amplify the cognition process.
Stephanie Shih - Mar 31, 2009 07:41:46 pm
This reading provided a pleasantly comprehensive explanation of information visualization and why it is important. Using a visualization, someone can zero in on the most pertinent information, and the visualization helps a person retain said information better as well. When the amount of data is too large to comprehend, information visualization helps someone process the data in an organized manner.
Shoeb Omar - Mar 31, 2009 09:26:34 pm
I liked this article, and basically it highlights the importance of presenting information in a format that a user can understand quickly, easily, and efficiently. That is, after all, the entire point of information visualization--to help users interpret and see information in a format that is easily accessible to them. I thought that the article for the most part used interesting and effective examples, with a really good intro with the multiplication example. It really highlighted what things have worked in different situations and what things haven't. However sometimes it did get a little frustrating having to learn and understand so much about an example before a point was made, and it would have been easier if there were more examples related to things readily within our realm of knowledge. however, that has it's downsides to in that many of the examples may have been contrived and thus the reading would not have been as interesting with its real life examples from all sorts of fields.
Dwij Garg - Mar 31, 2009 10:38:42 pm
This was an atrociously long, yet very comprehensive article on Information Visualization. Basically, this concept relates to the use of computational tools to interpret and read large amounts of abstract data. This is important for designers of user interfaces because it helps them in molding the information that needs to be presented in a way that calls for human interaction with the interface to understand and explore the data at hand. Moreover, using this, a designer can effectively present abstract data to users in a way that will also help them memorize and remember the data. The concept is similar to understanding hard scientific facts through simple illustrations that help a student not only learn the material better, but also remember it for future use. Information Visualization is thus very helpful to designers in making an efficient and effective interface for users to use.
Sean Hansen - Mar 31, 2009 10:53:18 pm
This is another article that provides interesting points that I hadn't ever thought of before; I never considered that having data visually available would enhance cognitive efficiency so much (although the example of a person taking a full 50 seconds to multiply a couple of two-digit numbers seems a bit of a stretch), although it doesn't take much thinking to see why. One of the previous readings about cognition mentioned that the size of working memory is pretty small, so having perceptive referents or actual data on hand decreases recall time tremendously and the better organized the data, the greater the efficiency. That being said, I have little use for the article beyond the overview; that is, I do not foresee myself having need for multi-dimensional graphing techniques.
Chunwei Lai - Mar 31, 2009 10:56:45 pm
One example, I liked in particular was the one on the cost structure (cost of driving & walking in San Francisco). The way it is presented visually was much more insightful than the paragraph they used to describe it. A good way of presenting the GUI would be to present it visually, let the user decide on how it should be used and then actually provide the instructions to cover any confusion. Had I read the paragraph on the cost structure prior to seeing the two visual cost structure, I believe I would have been more confused. It is easier to associate (recall) than it is to process (remember) certain types of information and this article made a strong presentation for those cases.
Alan Young - Mar 31, 2009 11:23:00 pm
“Information Visualization” was a very interesting article to me because I found the power of external aids to be more important than I gave notice to. It's very reminiscent of previous readings we had that touched on the importance of visual cues to enhance recognition over recall and other ideas of reducing the gulfs of execution and evaluation. I agree that the power of the unaided mind is overrated and that often times genius or ingenuity is really the result of some ability to fully visualize a given problem, even if it’s a non-spatial abstraction. I liked the example used of comparing mental multiplication with using paper and pencil and the explanation that mental multiplication is not difficult, it’s just that keeping intermediate results in memory is difficult. The paper and pencil merely extend a person’s working memory. Other examples of external aids given were navigation charts and diagrams. The important definitions summarized: external cognition is the interaction of cognitive representations and processes across the external/internal boundary in order to support thinking, information design is the explicit attempt to design external representations to amplify cognition, data graphics is the design of visual but abstract representations of data, and visualization uses the computer for data graphics. The last part of the article is concerned with the model of information visualization and the steps of translating raw data to a data table which can then be mapped to a visual structure, all while maintaining the Gestalt principles that were explained in last week’s “Organization and Visual Structure” article.
Kevin Huey - Apr 01, 2009 12:50:35 am
A reading intended for non-CS majors?? I guess, since they had to explain in one paragraph what "1" and "0" inside the data table examples meant. And they didn't say True/False in any part of the explanation...
I think this reading wanted to teach people that visuals are supposed to have a main underlying purpose (and it's not to look aesthetically pleasing). It talks about how we can manage our data, come up with good identifiers, and determine how to best convey our information through a visual representation. The aesthetics are secondary, as long as the user knows the purpose of the visual and how to read it. This reading is probably good for me personally because I tend to care about the look and feel of a visual more than the actual conveyance. Thus, I got to read about the arguments for the other method and maybe I can incorporate it more into what I do.
Mark Dhillon - Apr 01, 2009 01:15:46 am
Holy crap this is a massive article. One idea that I found intuitively explained was that of external cognition in regards to mental multiplication. I agree that the multiplication itself is not overly difficult, but trying to remember all of the previous calculations is a tricky task. I thought this tied in nicely with the memories discussed in the Human Model Processor article before the midterm. This concept also reminded me of the thumb and index finger method or right hand rule (for magnetic fields) in Physics. It's cool that when we combine the abilities of our senses that we can do things much more efficiently, instead of relying completely on one (like thought in this case in conjunction with sight helps improve the speed). Gawrsh I love this class.
Phiroath Chan - Apr 01, 2009 01:26:19 am
This article or paper mainly talked about how visualization can amplify cognition. There were a lot of definitions in the beginning of the paper was sorta unnecessary, but helpful i guess. the part that i thought was useful was the table on how visualization actually amplifies cognition. I agree that with the correction orientation of information one can expand their working memory. Less strain is also a plus with good interfaces since with reduce time in searching for items, more time can be put into the design. Consistency with interfaces can also dive into mental memory where one can perform an action without doing little or no thinking at all. Again that can also increase the efficiency of the project design. The last benefit i want to talked about is the benefit to manipulate the interface itself. This benefit was already described in detail in previous reading so no need to emphasize its importance any further.
Denise Ngai - Apr 01, 2009 02:03:07 am
I think visual information design is indeed an important aspect of UI design. A lot of the thing we are familiar with on the computer, we associate with their icon, or their visual representation. I agree with Timothy (above) about the saying, "A picture is worth a thousand words." This article really does stress the importance of visualizing information and how it can be more beneficial than textual representations of information.
Especially for those with photographic memories, visual aids are very helpful and are easier to retain. To stress this point, the reading even shows such visual aids as maps, charts, multiplication aids, etc. We witness visual aids on a daily basis and our minds are in tune to recognize them; this fact can be used to effectively design a UI that is user friendly and user intuitive.
Carolchen - Apr 01, 2009 02:02:05 am
While this article describes the theory behind information visualization in quite some detail, I am doubtful that it equipped me to make information visualizations myself. I did appreciate the points that were made about how visualization reduces the effort expended searching for or recognizing patterns in information, and frees up resources that would otherwise be tied up in intermediate storage of values to do other tasks. The practices of grouping certain information together that is used together relates back to some of the graphic design concepts we just learned, such as grouping or aligning by similarity or proximity. The main process the article described was that of taking raw data, converting it to a data table, mapping it to a visual structure, which is then manipulated with through interaction with a user. I found the descriptions of the conversions to be very tedious and sometimes overly wordy, though there were insightful, interesting tidbits thrown in here and there, such as the relative effectiveness of different retinal properties; for example, greyscale is useful for depicting the pertinent relationships between ordinal data but not normative data.
Moonway Lin - Apr 01, 2009 02:09:56 am
This article expands upon the previous lecture and our Human Brain Processor reading in emphasizing the importance of visuals to aid the processing of information. The previous lecture on Gestalt, groupings, etc. emphasized more of the aesthetics side of visuals, while this article emphasized the technical reasoning behind using visuals. The command line interfaces of the 1980s giving way to GUIs with information presented visually marked a great revolution in the development of user interfaces.
Jason Lo - Apr 01, 2009 12:57:39 am
I liked how this article explained how the human brain can not process information that effectively. Our cognition is limited by the tools we have at our disposable. It makes a good point when I do multiplication in my head, the most difficult part is remembering the part that is no longer the center of my attention. I watched a video where someone had learned to multiply large numbers in his head very quickly through the use of words that encoded numbers that he could easily remember and still perform operations on. I felt the article started really well and went kind of downhill from there, the tail end seemed to stretch to infinity (like a limit). Still the section on visualizations and looking at data tables at the end helped us digest how to better design interfaces and be more productive.
Sean Ahrens - Apr 01, 2009 01:58:32 am
I found this paper very interesting. The definitions of visualization and the related terms were very helpful. I actually wish I would have read this article at least a semester prior -- it makes the CS Visualization class seem really appealing. I am really impressed by the idea that the human mind is little without the aid of visualization. It actually is a pretty profound thought -- this means that human evolution is not necessarily so much the process of mental development, but instead the process of cognitive tool development. That pretty much just like blows my mind.
Also, the way the paper breaks down the knowledge crystalization process is very helpful. It makes it clear that data visualizations help the user discover schema of importance, and new insights to understanding the data. The Home Finder example makes this point clear. And the Challenger O-Ring example makes it clear that, unless the data is presented properly, coming upon insights may not be easy or even possible. I think this will have applicability to my group's project as well, as we will be reporting on user's performance on news trivia questions broken down into many categories. We want this data to have an insightful impact on our users.
Chao Michael Zhang - Apr 01, 2009 02:28:15 am
I found it interesting that the reason why paper and pencil multiplication is so much faster for humans than mental multiplication is simply due to the difficulty of storing small numbers temporarily to do easy calculations with. Nothing is inherently harder about the problem, but it is merely a problem of temporary memory storage.
A second thing I found particularly interesting was the Figure 1.30 on page 26, which showed a processed photograph that tries to demonstrate the level of detail available at different parts of our visual field. It's interesting that of all that we "see" at one time, we only focus on a single small area that is clear to us, while the rest of our massive visual field is simply shapes, locations, and colors. It's amazing that our sense of sight comes from this fact coupled with our eye's ability to move in tiny movements up to 70 times per second.
Derek Liu - Apr 01, 2009 02:33:35 am
One point this article addressed that I thought was quite important, even if a little bit obvious, was the fact that visualization amplifies cognition. This point is very important in interface design and human computer interaction since visual information must be presented in such a way that users can benefit the most. It is also important to realize that information must be presented in an intuitive way, as we have been studying all semester. If visual information is presented in some non-standard or confusing way, the effect is much more detrimental and the user would probably be better off without what visuals they are presented. However, good visual presentation and layout is most definitely better than some simple text directives.
Timofey Titov - Apr 01, 2009 02:42:00 am
This is another reading that points out the highly visual inner workings of the human mind. In my opinion, visualization has the same importance as data structures in programming. You can't do without them and a good one will make life a lot easier. The primary goal of software is to help humans. While the previous reading dealt with reducing articulatory distance, this one reduces semantic distance. This goes hand in hand with task-oriented design - it helps people be more productive by getting things done faster.
Matthew Can - Apr 01, 2009 02:40:09 am
The reading did a nice job of explaining the importance of information visualization. Primarily, it is a way to improve human cognition. One benefit of visualization is that it extends the working memory of humans. I found this point interesting because I had never before thought of visualizations in that light. I think their most important functions are enhancing the detection of patterns and reducing the search for information.
I thought it was helpful that the reading went through a model for visualization. It showed the process of going from raw data to data tables to visual structures and finally to views. I enjoyed reading about visual structures the most. The reading mentioned some interesting tips on how to employ them effectively. For example, grayscale is good for showing the comparison between ordinal variables but not the values of quantitative variables. Distortion is a useful technique for placing the focus on one part of the information while still maintaining the context.
Alexei Baboulevitch - Apr 01, 2009 02:36:03 am
As strange as this may sound, it was revealing to me that all cognition occurs within the mind, and that visual cues are used predominantly as a form of external memory. When we do multiplication on paper, we feel like we're transferring the calculation onto the paper, in some sense; mental multiplication just doesn't feel the same. With this in mind, visualization is in large part designing an efficient mapping from human memory of a particular task to visual elements.
I've found that I often use the exact steps of knowledge crystallization in order to comparatively shop, but I thought my method was at least someone unique to me. It's interesting to know that all humans do this the same way.
Sean Kim - Apr 01, 2009 03:25:42 am
The main idea in this article is to explain how visualization amplifies cognition. The use of computer-supported, interactive, visual representations of data is mainly used in these days. The author suggested the useful techniques that can be used to get he cognitive amplification; increased resources, reduced search, enhanced recognition of patterns, perceptual inference, perceptual monitoring, manipulable medium, and mapping data to visual form. And as we learned the previous lecture, he also claimed the importance of gestalt principles of organization; pragnanz, proximity, similarity, closure, good continuation, common fate, familiarity. Through this article, I could get the acknowledge about how to use visual information for better result, that is, for amplifying the user's cognition.
Cuong Ngo - Apr 01, 2009 03:42:11 am
Even though the chapter was pretty lengthy, I did find the discussion on View Transformations interesting. There are three common view transformations: location probes, viewpoint controls and distortions. Location probes reveal additional properties/information pertaining to a particular object. I think this probe is somewhat similar to magic lenses, which give the user an alternative view of the interface. Viewpoint controls come in handy when it comes to guiding the viewer's focus. They also make the UI elements more visible. The problem with viewpoint coltrols is that "the surrounding area disappears as the details are zoomed." Distortion combines overview and detail into a single visual structure. This view transformation can help the user "perceive the larger undistorted Visual Structure." I like the Perspective wall example. It's so much better than a tabular structure to display information in chronological order.
Raymond Young - Apr 01, 2009 03:29:08 am
This was a cool reading because it went over how different types of visual presentations can convey such different information. Different visual representations emphasize different patterns in the data, and some are more useful than others at different times depending on what one wishes to see. Some representations display data in a way that is hardly useful in any situation.
There are many points this paper went over that aid in UI design, since UIs are visual representations of the user-mutable system, and should emphasize the most desirable/useful possibilities that the user can execute in his/her actions. Many of the concepts were quite simple, but highly valuable in practice. For instance, grouping together information that is used together. This seems simple enough, but it really should be considered in every aspect of an interface design. The idea of grouping together information can be expanded in many ways. If there is a lot of data being presented to the user, then perhaps grouping like elements is not enough since there still may be an overwhelming amount of data on the screen. Considering how grouping information throughout different levels of abstraction is one way the idea of grouping can be expanded; Representing a group of data with an icon can be useful idea when trying to display an otherwise overwhelming amount of display data. It aids in crystallization since the user will be able to understand how the data in the group is related, not specifically but as an idea. The idea is meant to allow the user to successfully guess what could be within each icon's data, but not have to consider the amount of truth the categorization has upon the actual data. And when the user finally decides to look at the data, he/she can reference the icons to easily navigate throughout the entire set of data while only having to deal with a subset at a time. Such a display definitely extends the user's memory capabilities.
Meiying Li - Apr 01, 2009 03:24:41 am
This chapter introduce how a information visualization model is developed. It first persuades us using information visualization will help making the mental information process faster and easier. It amplifies cognition by:
- grouping the information, which reduces search time and working memory in human mind
- supporting many perceptual inferences, which are easier for human.
After this the chapter talks about how to map data to visual form. We first translate the raw data into data tables, then we map the data tables to visual structures, then it gets to human perception, which is affected by the view transformation.
Rohan Dhaimade - Apr 01, 2009 03:31:22 am
The multiplication example is definately something I've encountered before. I've tried multiplication in my head before and it's so much easier to work on paper. I just tend to forget things and having working memory extension is so useful. Visualization is definetly key to cognition. I've noticed that doing technical puzzles and taking notes is also the same way. Reading words is a bit different from listening to words.
I think the entire article seems to be one giant example of how to visualize things. The paper covers many different ways to visualize data and import aspects of visualization. The article though does a good job and is well informative. I got many different ideas about visualization and how to visualize data itself. One thing that I"m wondering is if fonts affect visualization and cognition. Reading itself is a visual exercise, and I wonder how fonts/styles affect cognitive learning.
Eric Hernandez - Apr 01, 2009 04:56:42 am
This article was refreshing; I've always had only a vague understanding of what people in Information Visualization are trying to do. I suppose that since 30-50% of the human brain is devoted to vision, choosing the proper visual representation for data is key in problem solving and understanding. For most of the examples given in this paper, the good representation clearly showed the patterns that emerged from the data. It is also very true, as discussed extensively throughout this article, that visual aids are a way to extend working memory and make processing faster. Why not auditory or other types of aids the article asks? My immediate answer, once again, was that humans simply specialize in visual processing (that 30-50% again). Specific details were also interesting, such as which colors and shapes stand out the most in scenes.
William Cho - Apr 01, 2009 03:20:37 am
Information visualization is a relatively straightforward and relevant concept. What I found interesting were the ways of translating raw data into views, and the examples about how visual aids enhanced cognitive abilities, like the one about multiplication of numbers mentally versus using pen and paper. Sure, multiplication is easy; it's just keeping track of the intermediary steps that is the difficult part. This idea seems simple, yet extends to our domain of user interface design, of how to present abstract data visually in a useful way that users can easily take in and find insight in. Note the distinction between visualization and information visualization; the key word 'abstract.' The reading got me thinking about several programs I often use, and the ways in which their visual cues vastly aid cognition.
Sum Sum Wong - Apr 01, 2009 05:01:30 am
This week's reading is a interesting and inspiring one, although it is quite long. It provides us with some basic information about visualization, like, the importance of visualization and how it is related to human brain (how it amplifys cognition). The "time to multiply" example on the first few pages is a nice daily example to demonstrate why visualization is so important. The most interesting part, to me, is the visual structures. In my opinion, a good visual structure should be the one that can maximize the amplification of cognition. The sine curve example (fig 1.26) is undoubtedly a nice example. But when things get more complicated, how can we make the best choice among all those reasonable and usable visual structures? (P.S. fig 1.23 is very helpful)
Adit Dalvi - Apr 01, 2009 05:16:07 am
This was a really long reading! What I got out of it was that information visualization is extremely important in user interfaces, which is kind of the same message that we got from the earlier readings on the visual store, the perceptual processor, the cognitive processor and short-term memory. The multiplication example follows from that too. A large portion of the reading covers (in detail) different ways to visualize data and does a pretty good job at it.
Salman Rahman - Apr 01, 2009 01:30:14 am
This article was so ridiculously long that I lost patience and ended up skimming it. I think the main point is about how visual information when presented properly to the user allows the user a much more useful and understandable way to internalize information. The article goes into a lot of detail about different aspects of this and different examples (i.e. how a paper pencil can help do multiplication or cost structure of walking through downtown SF). This concept of IV seems to be at the heart of this class. We are designing interfaces that allow the user to perform tasks in an easy-to-understand manner. Presenting information visually to the user such that this can occur is what this reading is all about. The two go hand in hand.
Bernardo de Seabra - Apr 01, 2009 06:01:28 am
The chapter on information visualization presents us with the idea that raw data, even when analyzed might not intuitively and readily offer the user a quick interpretation. With this said, the authors present the process of translating data into graphical visualizations in order for a better user experience in understanding the data presented by an interface. The presented model is described in a set of steps: first the raw data needs to be translated into a data table; second, the data table can then be used to display a visual structure representing the data. The view transformations also presented in the chapter allow an infinite amount of data to be readily available through the interface. All of this constitute an information workspace for visual interpretation of data. The chapter then proposes a few different techniques to handle different situations and provides real-life examples of known implementations. I personally feel that the reading is a little out-of-date. A lot of times there are limitations on data visualization (like in cell phones, web pages, etc.) that make trivial data presentation techniques a lot more interesting and challenging. It would be great to spend some time on these.
Prahalika Reddy - Apr 01, 2009 05:54:57 am
One of the main ideas I got from this reading was that visualizing something reduces the time needed to complete that task or action and increases the understanding gained from it. I agree with this idea because it usually works for me. When I read something, in order to help me understand what I'm reading, I create a mental image and as I keep reading, the image evolves.
In the article, there's a section about the difference between multiplication done mentally versus on paper. It was concluded that it takes much less time to compute the answer when doing it on paper. What I found interesting was that even when I try to do multiplication in my head, there's an image of the long-hand calculations in my head, as if I were doing it on paper. I think it reduces the time it takes for me to do the mental calculations as opposed to if I were to just try to keep track of all the numbers without any structure.
Overall, the reading was very interesting and insightful. There were many concepts that I liked and haven't thought about before.
Ling Chen - Apr 01, 2009 06:27:24 am
This article presented many concepts that i have not really thought about before. It's not hard to believe that information organization and visualization was an important part of UI. I mean, it makes sense. We need visual cues and aids to help us better identify more important areas of our interfaces. However, this is the first time I have ever heard of the purpose of information visualization as being use perception to amplify cognition. We also want all these knowledge to crystallize within the user. Being able to map data to visual forms can also help users navigate greatly. Also, like what Prahalika said, the example of the mental multiplication was interesting because I also found myself "seeing" a picture of the multiplication process in my head to help me keep track of numbers. The three common view transformations (location probes, viewpoint controls, and distortions) were quite interesting. I have never heard of such things before.
Alexander Cho - Apr 01, 2009 08:57:28 am
The beginning example of trying to do long-multiplication problems in your head makes it quite clear how important and vital visualization is. I never thought of solving problems in terms of memory storage. We take so much time trying to store intermediate values and that is very analogous to how programs work. With more RAM we can do processes much faster. Visualization is like having more RAM in this sense. Also it is quite clear how much more effective visualization is for large amounts of data. Everyone would rather see a chart or graph of plotted out data points compared to just some list of values. Visualization is obviously key to conveying data, especially in our UI's. There is a reason why GUI's are preferred so much more. Think of UNIX vs WINDOWS. You have to know the commands on UNIX, but in Windows there are buttons that show you what you can do.
Anjana Dasu - Apr 01, 2009 09:52:51 am
At the heart of this reading is the concept that visualization can amplify cognition in six ways:
- by increasing the memory and processing resources available to the user-- effectively providing a source of external memory
- by reducing the search for information-- imposing visual groupings on data speeds up search, also allowing for manipulation (see #6) often allows a user to search in the most convenient matter
- by making it easier to recognize patterns-- visualizations create structural relationships that allow the user to see patterns
- by enabling personal inference-- graphs/figures/etc. allow the user to have the core information in front of his face (think external RAM again) and this enables the user to change his focus to forming connections and making inferences
- by having perceptual mechanisms for monitoring-- visualization allows for the highlighting of certain more important aspects of a system
- by encoding information in a manipulable medium-- data becomes interactive, which in turn allows for easier searching, recognition of patterns, etc.
Andrew Chen - Apr 01, 2009 09:12:11 am
Something which I found interesting in this reading was that the same information presented in different visual formats yields different results. The Challenger example was particularly important: the same data presented in a different way would have aided the engineers' cognitive process in the right direction, instead of pushing them toward unnecessary knowledge (where the damage occurred or what the rockets looked like). This calls attention to keeping track of how we present information in UI design; it must focus on the primary information it is meant to convey, in the format which emphasizes that information.
Michael Cohen - Apr 01, 2009 10:31:15 am
Although quite long, I found this reading to be one of the better ones in the class. Information visualization is a pretty important concept. It seems to me that on a general level, most UIs serve exactly two purposes: to convey information, and allow the user to manipulate it. That being said, thinking about how best to visualize often abstract data will be pretty central to all UIs. I was most interested in the parts of the article that was pointing out that different displays of the same information lead to the same results.
Colin Downs-Razouk - Apr 01, 2009 10:22:46 am
I had a difficult time with this reading. The reading seemed to simply supply a series of definitions without giving any sort of explanation or example. This would be alright if the reading did this only a couple times, but there were many definitions, and few of them were fully explained. I thought the section on Data Transformations was the clearest and most useful section. Data Transformations seem to be an important aspect to any sort of visualization, because no matter how you design your graph or visual system, the data you use must be designed and ordered in a way that naturally brings out trends and associations. By the way, the date on this comment is incorrect. I am writing this at 8:50am.
Szu-Chun Mao - Apr 01, 2009 10:53:57 am
This article is quite long, however it is very interesting. It shows how visual representation connects with our cognitive activities. The first example in the reading, doing multiplication in one’s head versus on paper, gives us an idea how visual information can amplify our active working memory. This concept is not difficult to understand; however, I have never thought about this topic in depth. This reading use lots of visual examples to illustrate its points, making it easy to grasp the concept of visual information design. In short, this article is trying to tell us what we already know- a picture is worth a thousand words. The visual aids help us learn things faster and easier than words.
Chris Thompson - Apr 01, 2009 10:54:26 am
Seeing is believing, and such. Vision is so important to us that there are dozens of metaphors about it. It aids us in almost everything we do, even if it's just used as a reminder (such as temporarily storing the digits for multiplication). Sailors have used charts to navigate the vast and dangerous seas for hundreds upon hundreds of years. Diagrams can convey all kinds of useful information at a glance. After giving off a few examples of visual aids, the article begins to study the theory of information visualization. For instance, you can use things such as color, size, or radius to indicate quantity, quality, or time. A lot of the article just covers basic and somewhat obvious information. But I suppose it would make a good reference for the basic information when one wants to create a new type of graph, chart, or otherwise convey information visually.
Aaron Hong - Apr 01, 2009 10:56:50 am
Like they said Visualization is about insight and not pictures. It's pretty interesting how much more we can get just with good visualization of data. We can detect patterns that would have never been seen before (as in fraud), and so many other things. There is just so much stuff about external cognition that it's amazing how much more we can get from data by just using maybe a bit of orientation in a different direction or different shape. I'm also taking a database class and there is a lot of table about how you take table data and map it to visual structures... it might make an interesting ruby on rails project to do a visualization of some data.
Victor Lum - Apr 01, 2009 10:57:49 am
I found it interesting that different forms of visualization are better depending on what you're trying to do. Seeing something one way may allow us to spot patterns immediately, while seeing it another way may actually hinder us. The Challenger example really got me, because I never thought that something like how the data is presented can affect what people get out of it so much. I always figured that if something is presented in a different way, you'll still arrive at the same conclusion, only it might take a little longer. Guess that isn't the case.
Kevin Nakahara - Apr 01, 2009 10:39:33 am
Two discussions in the readings that I found interesting were the ones on diagrams and, later on, active diagrams. These are not manipulated by the user, but still require extensive mental interactivity. The example of the Challenger and the temperature diagram was another of those life-death problems presented by UI that we've faced in class, and I thought it was poignant to think that simple design can make important knowledge much more attractive for people to understand. The other example given was the periodic table. The discussion made me realize how comprehensively it is structured, since it takes alot of information into account and presents it in a educative and informative way. Users actually learn quite a bit just by looking at it for a few minutes.
Gregory leshner - Apr 01, 2009 03:32:33 pm
An extremely long article ... and one that defined a lot of terms which I doubt I will have much use for in my life. The basic premise of the paper is that visualizations aid in cognition. They do so by improving the amount of working memory one has, by providing a quick reference to dense amounts of data, and providing easily accessible and searchable structures for data. I found the discussion on the crystalization methodology a basic walk through of how we all approach tasks, organize information and slowly exclude input and reorganize information somewhat useful ... at least I know I am not alone in how my mind works. Knowing that most people think this way will aide me in designing interfaces that leverage this process.
I found the discussion of how the eye actually works pretty fascinating. I am not sure I have a direct application to UI ... but I found that I actually paid a lot of attention to that section of the reading. I never thought about the stabilization process or how we make many quick adjustments in view to create a full picture. The driving analogy shows that we prioritize our attention on objects that have more and relevant information to our tasks and build a picture of information based on these areas of attention.
The last section of the paper pulled together how Raw Data is transformed into Data Tables which are mapped into Visual Structures which are then transformed into Views. This is a useful model to remember and reflects most models found in computer science that deal with the transformation of data into visualized information .
Ian Hildreth - Apr 01, 2009 10:23:53 pm
This article was way to long, I definitely liked Monday's readings better. They spent too much time explaining trivial things, like how data is stored in binary for example. It is interesting to think that the way data is visually laid out and what information you actually provide increases cognitive efficiency, but 40 pages was not needed to explain the simple ideals behind this.
