Visualization - guest lecture by Maneesh Agrawala

From CS 160 Fall 2008

Jump to: navigation, search

Lecture on Nov 24, 2008

Readings

Discussions

Please post your critiques/commments on the required readings below. To do that, first login by using your user name and password, then click the "edit" tab on the top part of this page (between the "discussion" page and the "history" page), New to wikis? Read the Wiki editing guide. . Hint - Please put a whole line == ~~~~ == (literally) at the beginning of your submitted critique, so the wiki system will index, sign and date your submission automatically.

Contents


Vedran Pogacnik 22:44, 21 November 2008 (UTC)

The paper explains how transforming (visualizing) a structured piece of information helps humans think. It touches upon some other factors, like metadata, that help as well. The paper also talks about the ways of compiling and then categorizing the data. I guess finally it depends on the individual, how well he is able to work with a particular mode of presentation; for example, one might prefer looking at data in ordinal over nominal modes. However, some situations dictate the type of data used. The specific design of the visualization tool is highly dependant on the objective one wants to accomplish and the type of data collectable.

Jacekmw 19:56, 23 November 2008 (UTC)

The article seems to me to me to be essentially an extension or expansion of Nielsen's heuristic on minimalism and ease of use. In this case, the focus of the article is mostly on consolidating information into a format that makes the meaning of the data presented as clear as possible. The article suggests eliminating any focus on data that is not very relevant and making sure the presentation is applicable - for instance, in the example of the O-rings in rockets and temperature of launch, the correct informational diagram was the graph simply correlated failed launches due to o-ring issues against launch temperature. The other chart showed local rocket damage, it appeared, and had a condensed chunk of text explaining the effect of temperature.

This is also very applicable to serious (or not) games. In many games, there is a HUD, or a heads-up-display, which shows the character's life points, ammunition, or whatever else is important to the game. The issue at hand is to show only what is relevant and in a manner that tells the player what he actually wants to know. For instance, if the game operates on a health point system, you would want to give the player a meter or a numerical basis for how much health the character has, not, for example, a diagram of where the player is injured.

Mike Kendall 20:29, 23 November 2008 (UTC)

This paper deals with data in storage, organization and visualization modes. Most of the visualization schema presented follow our previously known policies for proximity and alignment. Not only that, most of these methods for presenting data are dependent almost solely on pre-existing standards for data (a calendar is a grid, for example). I felt like this paper was, more than anything else, presenting us with definitions rather than methods. Maybe we're reading this just in preparation for tomorrow's lecture which would use these ideas and definitions in a way that would help interface design? Then again, when it comes to visualization, every problem is unique. Palantir focuses on writing software for solving very specific problems, visualizing large quantities of data in ways that are not useful for most other problems. This is because the solution requires a good understanding of the problem in order to make associations and conclusions suggestive.

Kai Lin Huang 02:16, 24 November 2008 (UTC)

"The purpose of visualization is insight, not picture." quoted from Information Visualization. Once a set of complex data represented with graph and/or plot, a wider range of people become able to interpret, conclude and develop new ideas based on their understanding of the graph/plot. The idea behind its success is very basic. Human lives in a world that consists of more tangible interfaces than abstract representations, especially in a person’s early stage of life. However, the science of how to create and what kind of visual representation for some data so that it can convey the most useful abstraction for individual purpose takes some time to try out and revise; this is analogous to developing user interface for a program. For example, plot and statistical graph summarize the overall aspect of data, but a table with categorized rows and columns can give a little bit more details without losing the summarizing function. The reason why computers can be widely used has to appreciate the invention of a graphic user interface. A side note, I was also astonished by how the analysis of visualization with human cognition system because of its relevancy and involvement in this topic.

Juanpadilla 03:25, 24 November 2008 (UTC)

I think most of us know that having a visualization of something will greatly increase the understanding of it, especially if it is something abstract. This article, although often quite dense, gives an interesting insight to why this is the case while presenting interesting examples. Since our eyes are the fastest sense processors, it only makes sense to utilize them by creating something visual to use as memory storage. It was particularly interesting to note the comparisons in the graphs of the O rings used in the space shuttle and to know that something this simple could have saved peoples lives. But I think this article also gives rise to question about how much visualization should be done on the data. Even though they talk about mapping the data to visual form, many of the example projects were, in my opinion, overly complicated; for example the stock graph on page 9. The one useful tool that I think I did take away from this article is knowledge crystallization. It is a process that can be applied to many applications and I think will be interesting to experiment with in future projects.

Perry Lee 04:02, 24 November 2008 (UTC)

As Jacekmw mentioned, this article essentially seems to be an extension of Nielsen's heuristic on minimalism and ease of use. The concepts it set forth are not all that surprising; it makes sense that we find information visualization so powerful and useful. Of the five traditional sense, it seems like most people rely on their sense of vision the most. The article takes the time to list the six major ways in which visualizations can amplify cognition: by increasing the memory and processing resources available to the users; by reducing the search for information; etc. I don't disagree with any of the points they bring up. Using techniques such as knowledge crystallization certainly work.

Karen Tran 05:37, 24 November 2008 (UTC)

Information visualization is the representation of data graphically that gives the readers a understanding and insight into the data. Information visualization will compress a lot of data into one picture. It can display relationship between different quantities, for example, both in space and time, position and names and also new space-like structures. It opens up the possibility to view the data selectively. The rapid increase of today computer speed and memory provide an environment to develop more complex and more realistic simulations. However, this will bring with us huge amounts of data. We need more graphics power to visualize these multi-dimensional data sets. The article introduces a few tools which we can use such as visual tales, graphs and various kinds of plots. However i think that visual cues can also be very misleading if its not used correctly. An example scenario is if you look at a 3D graph from only 1 axis, then you will only see a plane and not the whole picture. Many visualization tools which I've used tends to have a similiar effect because it uses some forms of lossy algorithm to compute the final visualized data. I think that at best, visualization should abstraction, it should not simplify the data in any way that cause a loss in information.

Jordan Berk 05:54, 24 November 2008 (UTC)

Visual information design is clearly very important in many ways, as this chapter made clear. The space launch example in the "Diagrams" section is a good representation of this. By using an appropriate graph model, the information is more accessible and appropriate. Often a graph/chart/diagram can change the data accuracy or reduce readability of certain aspects, intentionally or not, with different techniques. For example, pie charts that have one category as "Other." Back to the O-ring chart though, it's kind of sad (as Juanpadialla pointed out) that something as simple as a diagramming choice affected people's lives. I personally have a lot of trouble making out the top chart, even after a minute of looking at it, where the second chart is instantly accessible and puts the pertinent information right at the forefront. This is also a good example of the value of empty white space, and how it is not always an evil thing in visual design.

Buda Chiou 06:23, 24 November 2008 (UTC)

Sometimes presenting data by graph might be a good way to make information more understandable, but sometimes it makes things more confused, especially the 3D graph. I admit that the information visualization may clearly present the information, but it's definitely not making the information more comprehensible like the author said in last paragraph. I think knowing when to use what technique to visualize the information is also a hard task when we use the information visualization.

Alan McCreary 06:32, 24 November 2008 (UTC)

I like the author's example of mental multiplication early in the reading. We like to think of the mind and the outside world as two separate entities, but this example shows that the two are in fact connected, something that isn't obvious. The author shows - as does my experience with solving things like computer science problems - that using visual aids greatly helps in solving problems, since visual diagrams can put together large amounts of information in a well-organized way that is native to humans. As Jacekmw mentioned, the ideas presented in this papers are very relevant in game design, since a well-designed game will present important information in such a way that the player can take it in without having to waste time deciphering what it means.

Cynthia T. Hsu 07:03, 24 November 2008 (UTC)

I agree with Mike Kendall's statement, that "when it comes to visualization, every problem is unique." For example, the reading cites how the proper representation of the correlation with temperature and O-ring fracture could have prevented the Challenger explosion - whether or not temperature was an important factor could have been known without proper visualization of course, but technology back then was probably not suited for analyzing different factors. The possible causes of an explosion could have been anything - temperature, weight of the shuttle, age of the equipment, experience of the pilots; there's a long list of things that they could have looked at as a preventative measure, without possibly knowing whether or not they were assessing the correct variables. I can see why an initial diagram showing things by historical date makes sense - chronology is an intuitively sensible way of making sense of a variety of events that have no other organization schema otherwise. Of course, this fails for analyzing correlative data, but without a previously requested need for it, there is no way of knowing. Contextual Analysis strikes again.

I also found the remark by Tufte on the matter rather interesting: "There are right ways and wrong ways to show the data; there are displays that reveal the truth and displays that do not." There are also ways to display data such that it reveals something to be false. I remember my CS70 professor gave us the example of a case study in which smokers and nonsmokers were tallied, then tracked twenty years later - they found that a higher percentage of non-smokers died than smokers. This discrepancy was caused by the average non-smoker being of an older age than the smoker, something that could easily be hidden if a diagram is designed carefully or carelessly enough.

Billy Grissom 07:45, 24 November 2008 (UTC)

Interesting. I really never thought about how much our visual perception influences our cognitive abilities. I think the multiplication example that was shown in the reading was a great example of this. I mean I can totally relate to the other's conclusions. Even in division, it's often far easier for me to calculate the problems on a sheet of paper rather than do it in my head. This isn't because a problem is difficult but moreso because, as the article shows, visually seeing it makes it easier to sort thoughts. Visualization provides an area of storage that seems to relive a lot of the extra pressure on our mind. this lets us spend less time organizing and more time actually solving the problems at hand. i also really liked the example on maps and he o-rings. I think although visual elements do make thinking easier, it is important that we develop ways for the actual visual element to make sense. As the o-ring example pointed out failing to do so can cause the visual information to be more of an obstacle rather than a tool for support.

Jonathan Fong 07:51, 24 November 2008 (UTC)

This was a very interesting reading. I never knew or considered that visualization of information was much of a user interface topic. I guess this is because I don't think of interacting with information as much as I should. In this age of overwhelming/massive amounts of data and information, it is becoming more and more crucial to be able to see data in meaningful and efficient ways.

A big trend in this area is the use of dashboards. These are designed to show information to executives; their top priority is to see things quickly, and they don't need to manipulate the data as much. On the other hand, Palantir Technologies makes interfaces for people to see and manipulate data in unique ways. Their customers are finance and government, both of which are known for massive amounts of data.

Lastly, it was interesting to see my workspace layout be presented in a different light. Like the office worker in Figure 1.19. I use the "hot-cold-warm" layout, which I always thought was analogous to a computer.

  • Hot area: information/supplies needed often and/or quickly are placed nearby (on the desk, where space is commodity), just like the CPU cache
  • Warm areas: less often needed info/supplies are a bit further away where there is more space, but is still relatively quick to access, just like RAM
  • Cold areas: rarely accessed supplies/info are placed on a bookshelf much further away, where there is much more space, like the computer hard drive.

Now I can say that I understand and use the "Cost-of-Knowledge Characteristic Function." =)

Wenda Zhao 07:52, 24 November 2008 (UTC)

The reading is pretty interesting. It seems that today 3D interface is still not widely used and 2D interface is still dominating. I think one of the reasons is that users are more used to simple easy visual info and layouts than more complicated 3D interface. Users are overwhelmed when developers put too much info and fancy stuff on the screen. Speaking of simplicity, google has done a really good job. Their homepage only has a input field, so people immetiately know what to do after they go to the site. In conclusion, the simpler the better in terms of visual layouts and contents.

Trinhvo 08:10, 24 November 2008 (UTC)

"A picture is worth thousand words." That's what this article mainly talks about. This article begins with an introduction to many types external aids, and then their applications and shows how powerful these external aids are to amplify human cognition. Because external aids are so powerful to crystallize human knowledge, with modern powerful technologies, there are many (educational) applications nowadays use 3D graphics to help users have better visualizations.

Yuta Morimoto 08:11, 24 November 2008 (UTC)

The reading focuses on general knowledge of visualization. It is very useful to consider user interfaces from more conceptual and theoretical view. I think that one of the important things described in reading is "Mappng data to visual form". It gives me simple model to figure out the sequence of visualization. The sequence may look pretty general and not be useful specific case. However, when I made visualization application, the framework of mapping data to visual form is very useful to separate a procedure of visualization.

Frank Yang 08:51, 24 November 2008 (UTC)

I would never have thought that visual perception had so much of an impact of how we perceive things. However, after reading the article, it makes sense. We typically view the world through what we see, with all the other senses merely enhancing the image. It was interesting to read about the multiplication experiment. I thought about it myself as soon as I read it, and I realized that if I had tried to multiply the same numbers in my head, I would try to picture the numbers floating as if I were doing it on paper. As Wenda pointed out, 3D interfaces are still not widely used. I think another issue with 3d interfaces is that the majority of the time, a 3d interface is presented on a 2d screen, which forces the user to attempt to navigate through the interface in a non-intuitive way. Therefore, while data could be displayed in a more interesting way, if the same effect can be achieved with a 2d implementation, the 2d one will still be preferred.


Shyam Vijayakumar 09:00, 24 November 2008 (UTC)

The reading stresses that visualizing helps us understand things better. The multiplication aid experiment definitely supported this because people who tried to do the problem with a pencil and paper reduced the time to solve the problem by a factor of 5. However, that's because people, by now, have a clear and simple way of doing the math on paper. If this method was convoluted in some way that was not clear and simple, then it would turn into getting wrong answers. This can be translated to the example that the reading talked about with the Challenger. Since the engineers saw the diagram that was misleading, they recommended the launch. In our prototype, we had trouble deciding how to represent the weekly statistics of driving behavior. Eventually, we decided to go with a bar graph to make it easy for the users to visualize the various stats like avg. miles driven versus miles driven for the current week.

KevinFriedheim 09:07, 24 November 2008 (UTC)

In CMS's (Card, Mackinlay,and Shneiderman) "Information Visualization - Using Vision to Think," a clarification is made about what is involved in how one goes about making a visualization what it should be. One point that struck me as important was when CMS describes how visualization helps facilitate information finding and understanding as with Mendeleyev's Period Table of Elements in Chemistry. He used "highlighting" as a method for organizing data dynamically -- the example that CMS gives is: "... a user can indicate interest in ionic radii between 93 and 206 and instantly those values will be highlighted on the table." I though this simple technique used by Mendeleyev was extraordinarily easy to employ but just as extraordinarily difficult to derive.

Its amazing how much a simple visualization scheme can affect how something is viewed and this reading really touched up on this fact and did so with some pretty obvious examples.

Jimmy Nguyen 09:18, 24 November 2008 (UTC)

This was an interesting article because in a broader scope, your learning effectiveness is really dependent upont your environment. I remember learning from high school that all factors that may contribute are your scenario, the music you may be listening to, and also the way you associate yourself with visuals. My teacher before taught me that even the slightest coloring of notes and pictorals will help one learn. I help that even when coding in computer science, something simple like syntax highlighting is very constructive. One of the best things developed that is now universally used, originally by Microsoft Word, is the red underline spell checking. It helps people learning and is a very useful tool in general.

Xuexin Zhang 09:37, 24 November 2008 (UTC)

Information visualization is a very important topic of Human Computer Interaction; it covers the ideas of how to effectively represent data into certain forms of piece which human could understand easily and find the data itself meaningful. Information visualization could usually separate into the forms of graph, chart, and diagram. I agree with Jacekmw’s idea of applying the concept of minimalism and ease of use to information visualization. By having a minimal and ease representation format of information, human could understand the information easily and effectively.

Gary Wu

This article reminded me of Nielson's heuristic of matching the system to the real world. Having an icon that accurately and simply represents its corresponding action makes using the interface a lot easier and manageable. The article really stresses the importance of visual perception and how they aid in processing information. Another heuristic that this article touches is aesthetics and minimalistic design. With the multiplication aid as an example, I can see how the placement and design of the "interface" gives rise to efficiency from the user. Ultimately, many of the ideas and examples this article discusses, relates heavily to what this course is all about.

Witton Chou 10:00, 24 November 2008 (UTC)

Visualizations are a very important part of any tool we use. It is a field that covers a wide variety of techniques to organize and convey concepts that may not be easily extracted as raw data. Card et al brings up the well discussed topic of the Challenger O-ring problem (a severe visualization impairment analyzed by Tufte). Whereas the original visualization of the o-ring behaviors under certain temperatures tried to picture the state of the o-ring, card's recreation with a scattergraph of damage better demonstrated the severity of the conditions of the launch day. Had effective visualizations been utilized with less ink clutter, the correct decision to delay the shuttle launch would more likely have been made.

While graphs, charts, and drawings are very helpful to us, visualization is much more than just static representations of data. The dynamic homefinder is an example of an interactive visualization tool that allows the user to dynamically update search results through sliders and scroll bars to easily compare and contrast the data the user is interested in. Something I made when playing around with the Flare, a visualization toolkit developed by Berkeley's Jeffrey Heer, is a sort of movie searching [utility] that categorizes, sorts, and filters movies from a database of the past two years.

I think a very important topic in the paper is also the categories of different types of variables. It's hard to imagine that any variable can be classified as one of three types: Nominal, Ordinal, or Quantitative type. And these properties lend themselves very well into different types of encodings such as color, shape, size, position, etc.

Information Visualization is a very important piece for anyone interested in visualization. There are a lot of topics to explore in visualization and this chapter is a great overview of what this field is about.

Volodymyr Kalish 10:46, 24 November 2008 (UTC)

With a continuous stream of incomming data our brain stores all the data in schemas, so later the data can be easily retrieved. Visualizing data is a very important concept used by many scientists and engineers. (That's how Moore came up with his law, since all he did to see the tendency of doubling transistors was to plot the number of transistors versus time in years) All people think and interpret the same data differently, so what some people see in some data visualization, others don't (discovery of Fibonacci sequence). That's why it is important to be able to represent the same data visually in many different ways, so it would make more sense and be more helpful in drawing conclusions to most people looking at the data visualization.

Hao Luo 11:12, 24 November 2008 (UTC)

The reading is somewhat related to the Heuristics mentioned earlier in class. It talks about data structures and the organization and visualization of data. Really this is what this course is all about. User Interface is all about taking the data and making it easy to see, easy to understand, easy to use. The field of Computer Science, in fact, is largely about taking the hardware from the Electrical Engineers of the world, and building layers and layers of abstraction so you create a product that is interface-friendly to the average user. Thus, most of the computer science world has to do with organizing the data and visualizing it.

nathanyan 13:19, 24 November 2008 (UTC)

I found the section on "Knowledge Crystallization" (page 11) to be particularly interesting - oftentimes one deals with a fairly concrete task/objective, but without a certain way to approach how to solve it, whether that be what questions to ask/data to research, or after amassing data, how to organize it into information. With raw data, one often becomes pigeon-holed into spitting it out in generic table or chart format - formats that often are easy to read and retrieve any particular data from (the value of X for Y and Z circumstance), but in terms of summarizing data and returning information that conclusions can be made from, are often useless. Thinking in terms of information visualization for this schema thus forces one to come up withe a schema that actually conveys information, rather than pure data.

Geoffrey Lee 13:37, 24 November 2008 (UTC)

Ultimately, visualizing information is about generalizing the data so that you can wrap your mind around it. It's kind of like the difference between a list of a hundred numbers and a scatter plot of those numbers. The human mind simply isn't designed to keep track of that many things at once.

Kumar Garapaty 15:25, 24 November 2008 (UTC)

The chapter talks about how to increase visualization for a specific piece of data. Visualization obviously benefits the understanding of that piece of information for faster comprehension and use of that information. I was previously familiar with the word visualization but never with the phrase information visualization. The chapter discusses various ways to show this data, one of which is information chromatography where data is just laid out on a visual substrate.

Bing Wang 16:09, 24 November 2008 (UTC)

I find this reading pretty interesting. It is logical that the author divides the visual perception into two big categories: one being the explanation part where he mentioned "a picture is worth a thousand words" and the other part being the central point of the paper which was visual thinking. To me however, these two big areas have a lot in common. With the picture worth a thousand words, ultimately, the human mind which the author said have very limited memory can digest those thoughts of thousand words into a simple picture. This in turn is like visual thinking as the user uses the picture to facilitate the thinking process.

How we use visual aids and moving object is very important. As one of the most delicate organs in human body, our eye really provide the visualization that we need. In terms of games, this is also the case. We use visual aid to provide us with feedbacks and interactions with the game system. We can also make intelligent decisions based on the visual perception we observe. A game with visual and graphics is probably much more fun to play than a game simply based on text.

James Yeh 17:02, 24 November 2008 (UTC)

I think this reading is interesting in that it shows how what see and in what form we see it can greatly affect how we perceive data. I was reminded of the Model Human Processor when the scenarios such as the multiplication example were brought up; the fact that I could not store and keep track of intermediate values very accurately or for very long was evidence of the limitations of short term working memory. In addition, the diagram presented for the process of knowledge crystallization represents a process that a human is constantly looping through to make sense of visual information. The many examples provided of visual cues also relate to the reading on patterns, since in some of the examples had visual organizational characteristics that could easily be derived into pattern form. The fact that our brain reacts to visual arrangements in certain ways means that there are specific patterns that humans can just interpret and understand more quickly, while other, less intuitive visualizations might result in decision making errors.

Antony Setiawan 17:32, 24 November 2008 (UTC)

I found this reading both interesting and informative. Visualize information is indeed amplifies cognition and acquisition. Schneiderman explains how information visualization using diagram helps: reduce the amount of search by grouping them together, the needs of math symbolic labels was avoided, and it supports large number of perceptual inferences.

Saliem Than 17:54, 24 November 2008 (UTC)

  1. The arrangement of office spaces
  2. The walking routes for a city
  3. The site design and ideas behind turning search engines into visualization search engines - kartoo - displays clouds of the areas for a search term - a search for javascript shows me the groupings of information currently available to me online - other sites like quintura makes uses of word clouds - other sites show screenshots of the site results - kartoo, quintura, pixsy, grokker, searchme, viewzi
  4. The palantir project recently showcased on FB about user activities on FB
  5. Geolocation + visual search
  6. Visualization libraries and apis in javascript
  7. site trends google search - shows the amount of traffic a site gets compared to other sites, or shows the amount of queries being done for one term versus another
  8. iphone mobile visual search?? different level of visualization - Evolution Robotics' Visual Pattern Recognition technology. (ViPR®)
  9. How to best use/improve visualization for finances
  10. The idea of trying to turn text search engines into visual search engines is interesting. I wonder if you can extend the idea of arranging an office for better access to information as an analogy of arranging information online.
  11. wonder if there's a way to apply visualization to tasks to optimize tasks.
  • Using visuals to increase cognition
  • A method for transforming data into visual structures
  • Knowledge visualization - for making patterns become more apparent

Stuart Bottom 18:03, 24 November 2008 (UTC)

I think I have to respectfully disagree with my classmates who commented that this reading is a lot like Nielsen's heuristics guidelines; it seems to me this reading presents a recommended framework (the authors call it a "reference model") for data visualization, but does _not_ present any sort of "requirements" for mapping data to visual form. There is, in fact, no definitive list of heuristics presented for a "good visualization," simply (I would argue) because the field is too new. To me, this reading reminds us that there is still plenty of room for innovation in computer science; there is no standard set of "best practices" for data visualization because (as mentioned previously in this discussion) the field is still far too problem-specific. You can't always directly apply the visualization for one discipline to another, simply because it doesn't always fit the same goals - and as the reading mentioned, there hasn't been enough study in the field to define which methods of visual mapping are more effective than others.

As far as how to apply this reading to our semester topic of serious games, I definitely have more questions than answers. It might be nice to see if there are any readings on data visualization specifically within computer games. One of the aspects we've had to think about in our game is the speed with which the user can access information - namely whether music "tracks" are on or off - and we've redone our interface to support faster saccades over the relevant data display area by changing the indicators to use binary color instead of position as the means of communicating their status.

Paul Im 18:19, 24 November 2008 (UTC)

This article by Card, Mackinley, and Shneiderman presents a very computer science-y approach to visualizing data on an interface. The concrete methodology presented here involves taking the raw data and creating data tables, which then develop into visual structures. The article provides many, many examples of different visual structures that can be used to represent data. Different types of visualization models accommodate better depending on what type of data is provided. Overall, it seemed like a very informative article.

There seems to be a number of applications in relation to serious games. For example, we need to create as simple of a representation as possible in order to keep the user from being bombarded with all sorts of data. As we have seen in our project (even in Flash), creating variables to represent our data in our code has made our lives much easier. In turn, the visual representations are much cleaner as well. One gripe that I have about the article is that most of the examples provided refer to huge, complex data sets and their representations. In this way, it seems a little less applicable to our current project.

JoshuaKwan 18:20, 24 November 2008 (UTC)

Visualization has been essential since the beginning of time to give a "big picture" of ANY data to those who use it. We used charts on Nov. 4 to see how our favorite candidate was doing. We use charts when using glookup to try to visualize how everyone did on a midterm. At the lowest level, e.g. the multiplication example in the reading, you are simply using your visual aid as a state machine for how far along you've gotten with the operation.

I agree with Stuart's viewpoint on the chapter as a whole: It does not specify any hard guidelines as to what data visualization needs to fulfill. It is simply a whirlwind tour of visualization with many good examples, and I think the point is to illustrate what kinds of data sets are appropriate for certain visualizations - not hard and fast rules.

Mikeboulos 18:23, 24 November 2008 (UTC)

I found it very interesting how the author defined External cog, Information design, Data graphics, Visualization. all of these amplify cognition, except for external cognition which is the use of external world to accomplish cognition. This article totally relates to our game, since we tried to make the scoring to be as visual as possible, and hence we fell under the following definition without even knowing it: "Human interaction with information workspace reveals properties of information that lead to new choices". Our game exactly followed this sentence, when the user hovers over a piece of food, they can then see a projection of how this food will affect their body, which then will change their choice or keep it depending on the information perceived from the visualization.

Haosi Chen 19:21, 24 November 2008 (UTC)

Information visualization is the representation of data graphically that gives the readers a understanding and insight into the data. Information visualization will compress a lot of data into one picture. It can display relationship between different quantities. It opens up the possibility to view the data selectively. I think the reading is quite interesting. I really never thought about how much our visual perception influences our cognitive abilities. I think the multiplication example that was shown in the reading was a great example of this. I can totally relate to the other's conclusions.

Retrieved from "http://bid.berkeley.edu/cs160-fall08/index.php/Visualization_-_guest_lecture_by_Maneesh_Agrawala"
Personal tools