Discussion


Note: Images may be inserted into your messages by uploading a file attachment (see "Manage Attachments"). Even though it doesn't appear when previewed, the image will appear at the end of your message once it is posted.
Register Latest Topics
 
 
 


Reply
  Author   Comment   Page 4 of 5      Prev   1   2   3   4   5   Next
benbendc

Registered:
Posts: 3
Reply with quote  #46 
I don't recall the incident you describe so can't comment, except to say that I'm known for pushing researchers to do stronger work:  : Ben Shneiderman beat us all up for wussy correlational studies.  
 
There are several active discussions about InfoVis research, so my future posts will be to those discussions.

As for your comparing yourself to Jesus, I can only note that this week is the Festival of Lights so I wish you Happy Hanukah!

       --Ben Shneiderman

__________________
benbendc
sfew

Moderator
Registered:
Posts: 803
Reply with quote  #47 
Ben,

Don't shut yourself off from the infovis community that exists outside of academia. We are far greater in number than the academic community and more intimately in touch with the importance of the work. Participating only in private discussions will not solve these problems. I remember quite fondly many discussions that you and I have had about the infovis research community's need to connect with the world of practitioners. At one time we were even working together to create an event for a large business intelligence conference. Please don't lose sight of this goal by retreating into the hallowed halls.

Happy Hanukah. (Perhaps, rather than Jesus, I should have used the analogy of Moses coming down from the mountain top and breaking the stone tablets in frustration. That's how I feel. By the way, referring to Jesus' action in the temple is not the same as comparing myself to Jesus.)

__________________
Stephen Few
sfew

Moderator
Registered:
Posts: 803
Reply with quote  #48 

What’s our real goal in this discussion? I’ll tell you what mine is. It is not to do anyone harm. It is not to win an argument. It is not stroke my ego. There are more direct paths to these goals than the one I’ve taken. My goal is simple and it’s precisely what I’ve been saying from the beginning: I want to help the infovis research community produce better solutions to real-world problems. That’s it. Infovis research could and should be much better than it is. In my article, I tried to help by exposing flaws in a specific research paper to illustrate systemic problems that are currently undermining this goal. I don’t get paid for doing this and I get no pleasure from conflict. I do this because it needs to be done.

Isn’t this a worthy goal? If it is, then let’s quit squabbling and work together to make it happen. Think of me whatever you wish (e.g., a mean-spirited candle-snuffer with a Messiah complex), but open your minds to fresh perspectives and put your house in order. You can offer the world more and better than you currently are.

Here are a few suggestions:
  1. Develop standard curricula for students involved in infovis. Make sure that this includes training in scientific method, knowledge of infovis best practices, and actual experience in using infovis to do real work in the world.
  2. Develop robust guidelines for the peer review process, and enforce them.
  3. Create a process for providing people who do flawed research with the help that they need to do better.
  4. Root out the fear of recrimination that prevents people in the community from identifying errors in the work of their peers, even after those errors have been published.
  5. Change the path to advancement from the number of publications and citations to the production of good work.
  6. Build a bridge to the world of people who could benefit from your work. This will require some humility. You must recognize that people exist apart from you who know as much or more about infovis than you do, even if they don’t have PhDs following their names. This will also require that you open yourself to feedback, even when it isn’t complimentary.
  7. If someone who cares deeply about infovis tries to contribute to your efforts, don’t slam the door in their faces. Find a way to work with them.
  8. If you don’t want feedback from those outside your community, for God’s sake don’t issue a press release.
  9. And finally, here’s one that will be a tough nut to crack. Find ways to dissuade professors from teaching subject matter or supervising work that they don’t understand. There are many aspects to infovis and no one understands them all. Let’s stick to what we know while always pushing the boundaries of that knowledge.

What we do matters. Let’s do what we can to produce better outcomes.


__________________
Stephen Few
tamara_munzner

Registered:
Posts: 2
Reply with quote  #49 

I've now read both of the memorability papers and both of your
critiques closely and am writing up my thoughts in detail. It may take
some time before that's ready, it's a busy time of year.

In the meantime, I'd like to bring the attention of readers of this
forum to the thoughtful article that Ben Jones has posted in his blog,
since I note there's no link posted here yet. See

http://dataremixed.com/2015/12/when-memorability-matters-another-practitioners-view/

Cheers, Tamara

sfew

Moderator
Registered:
Posts: 803
Reply with quote  #50 
Tamara,

Do you really think that Ben's blog post is thoughtful? In his example, he enhanced a chart with information that made it more comprehensible. Its usefulness was based on this, not on its memorability. He also set up a strawman when he suggested that I would label his enhancements as "chartjunk." I would not. An enhancement that supports understanding is not chartjunk. Despite its claims, Ben's blog post does not support the research paper.

Ben also employed a bit of misleading rhetoric to counter my statement that most decisions that are based on charts are made within seconds. He suggested that we would be like robots if we made decisions that quickly. Actually, as anyone who has studied cognition knows, most decisions made by humans are made in seconds. They are not based on memories of charts that were seen in the past. It is true, however, that charts do not always lead to immediate decisions. A chart may provide a piece of information that is later combined with other information to support a decision. In that case, however, it is the information that must be remembered, not visual elements of the chart.

__________________
Stephen Few
ebertini

Registered:
Posts: 3
Reply with quote  #51 
I am not sure whether memorability has a role in vis or not. But I agree with Stephen that in the example the additional marks make the chart more comprehensible (this is what I had written previously in my comment in Ben's blog). I think another aspect is relevance of information. In the example, the emphasis on what is missing through the additional (missing) bar and the note has a lot of relevance for the business. My hunch is that this is what makes it memorable, not the graphical format.

A more interesting aspect for me is how some graphical elements direct or redirect attention to what really matters and how these lead to 1) increased comprehension and 2) engagement: that is, leading people to spend more time with the graph. This does not seem to be studied/understood well enough. Ben's example is a really good one under this light. The annotation layer of a graph can carry a lot of useful information.

Stephen, it would be useful from your side to hear if you think there is a major difference between presentation and analysis. When data is presented rather than analyzed there is a much stronger need in my opinion to "guide" the reader and to "explain". This is where additional layers beyond "pure" data encoding have a role.

Similarly, I think it would be fruitful to discuss what parameters a visualization designer should maximize when building a visualization for presentation purposes. While I agree that comprehension is a key one, I am pretty open to considering other possible important goals. What would be an ideal list? You seem to suggest memorability does and should not make the list. Fine. What else should go there in addition to comprehension? How about "relevance"? And "engagement"? Many people have the practical problem of making their graphics attractive in the first place and then engaging, so that people do not just fly away after a few seconds.

This is a practical problem designers do have in practice and that often does lead to some trade-offs.

All the best,

__________________
Enrico Bertini
jlbriggs

Registered:
Posts: 191
Reply with quote  #52 
@stephen there is also the issue that, were to dress up all of our charts with imagery that made them more memorable, that threshold of memorability would soon be exhausted, and we would need to engage in an arms race of increasingly ridiculous levels of memorable items that we must add to a chart in order to make it memorable compared to the hundreds or thousands of other "memorable" charts that the user has now seen.

This would quickly defeat the purpose altogether.

I have always advocated for adding illustration and contextual information around the chart to aid in both understanding, and in grabbing the user's attention, in such circumstances as it may be required or helpful.

These things seem to be what Ben's post promotes, and do not in any way refute or contend with the issues at hand.

I think it's clear that if I look at 100 bar charts, and then I see a chart with dinosaurs, I am going to remember the dinosaurs more than I will remember any individual bar chart.

But so what?



sfew

Moderator
Registered:
Posts: 803
Reply with quote  #53 
Enrico,

I've actually written extensively about data sensemaking versus data communication. In fact, my book Now You See It focuses exclusively on the former and Show Me the Numbers focuses exclusively on the latter. When we have a particular message to communicate, we do whatever it takes to get that message across. Nothing that supports the communication of that message is chartjunk. Communication can be facilitated in many ways.

In addition to understanding, when we communicate information to people we certainly strive for other goals as well. For example, we strive for truthfulness. I don't think of engagement as a goal in and of itself. I think of it as a means to facilitate understanding. In other words, I want readers of graphs to become engaged in thinking meaningfully about the data, resulting in understanding. I don't want readers to become engaged in aspects of the graph that don't lead to understanding, such as the pretty colors in and of themselves. You mentioned "relevance" as a goal, which it certainly is. We should communicate information that is relevant to people's needs and interests. We should present information that matters to them.

I don't agree that "many people have the practical problem of making their graphics attractive in the first place and then engaging, so that people do not just fly away after a few seconds." This is sometimes a concern, but not nearly as much or often as one might assume based on the amount of attention that it gets. The vast majority of data visualization by far is done in the workplace to provide information that people need to do their jobs. Their attention does not need to be grabbed. It is in the context of web-based infographics, meetings in which people are being bored to death by a series of slides, etc, when the goal of attention-getting comes into play. Most people who write about data visualization seem to think that everyone who produces charts are professional graphic journalists when, in fact, very few use graphs in this way. Most data visualization practitioners need to focus on clear, accurate, and truthful communication, not on eye-catching graphics.

__________________
Stephen Few
sfew

Moderator
Registered:
Posts: 803
Reply with quote  #54 

The straw man fallacy in argumentation occurs when someone cannot defeat his opponent’s position, so he claims his opponent’s position is something other than it is—a position that is easy to topple—and then proceeds to claim victory when he discredits that erroneous position. I pointed out earlier that in his response to my article, Ben Jones built a straw man by claiming that I label useful enhancements in a chart, such as annotations, as chartjunk. Jones claimed that he toppled my position by showing that his enhancements were indeed useful. As he knows, however, I do not label anything that supports useful communication as chartjunk. I wrote to him, attached an article that clearly explains my position on chartjunk, and politely asked him to correct the misinformation that he was spreading about me. He countered that he would not correct what he wrote unless I admit that I was wrong when I wrote the following in the article:

Quote:
Visualizations don’t need to be designed for memorability—they need to be designed for comprehension. For most visualizations, the comprehension that they provide need only last until the decision that it informs is made. Usually, that is only a matter of seconds.

Jones insists that only robots make decisions this quickly. I responded to Jones that I could not admit that I was wrong because I’m not. Just in case others question about my assertion that most decisions are made within a few seconds, I’ll elaborate.

The typical human makes hundreds, perhaps thousands of decisions each day. The vast majority of these decisions are examples of what many psychologists call System 1 thinking, which is fast, intuitive, and unconscious. For example, when you drive your car or even walk through your house, you are constantly making decisions that result in actions, such as subtle changes in direction or speed. These are examples of System 1 thinking. Most of them are made in microseconds.

So what happens when you view a chart? Actually, several things occur. At first, you decide if the chart is worth your time. Ideally, you base this on a quick assessment of its contents and their relevance to you. This usually takes seconds. If you decide to read it further, you then make many decisions during the process of viewing it, directing your attention here and there to make sense of it. Most of these decisions are unconscious and immediate. During the viewing process we assign meanings to the data, usually through immediate acts of pattern recognition, connecting what we see with knowledge that we already hold in memory. If we are using the chart for information related to our jobs and become aware of something that demands action, we make an immediate decision. If we learn something from the chart that doesn’t drive an immediate decision but strikes us as potentially useful in the future, we decide in the moment to hold onto that information. We either record it in some manner (e.g., by jotting down a note) or we attempt to store it in memory. That decision to retain the information is sometimes based on System 1 thinking and sometimes on System 2 thinking, but in each case the decision is usually made quickly. When long-term memory is consciously brought into play, it is a memory of the information that we seek to store, not a memory of some visual aspect of the chart. Everything that I’ve described so far illustrates decisions that are typically made when we view charts, all of which are made in seconds.

Does this mean that all decisions are made in seconds? Of course not. Decisions that require deeper System 2 thinking might take hours, days, or years to complete. These bigger-than-usual decisions, however, are typically made up of countless micro-decisions, usually made in seconds, that serve as stepping stones to the eventual big decision.

So, Ben Jones, you have my response. Fast decisions are not the exclusive realm of robots. Now, I would appreciate it if you would remove your false statements about my position from your blog.


__________________
Stephen Few
jheer

Registered:
Posts: 3
Reply with quote  #55 
Stephen,

As promised earlier, my research group (the UW Interactive Data Lab) has written a short article on possible steps for further developing research methods and training in data visualization. Here it is: https://medium.com/@uwdata/next-steps-for-data-visualization-research-3ef5e1a5e349

Regards,
Jeffrey Heer
ebertini

Registered:
Posts: 3
Reply with quote  #56 
Stephen,


Quote:
I've actually written extensively about data sensemaking versus data communication. In fact, my book Now You See It focuses exclusively on the former and Show Me the Numbers focuses exclusively on the latter. When we have a particular message to communicate, we do whatever it takes to get that message across. Nothing that supports the communication of that message is chartjunk. Communication can be facilitated in many ways. 


I know your work Stephen. I am a big fan of it. What I meant is more that when we study vis for communication vs. for analysis I believe we want to optimize for somewhat different goals. Some are common, some are different. I just wanted to know if you agree and also to hear from you what these differences are.

Quote:
In addition to understanding, when we communicate information to people we certainly strive for other goals as well. For example, we strive for truthfulness. I don't think of engagement as a goal in and of itself. I think of it as a means to facilitate understanding. In other words, I want readers of graphs to become engaged in thinking meaningfully about the data, resulting in understanding. I don't want readers to become engaged in aspects of the graph that don't lead to understanding, such as the pretty colors in and of themselves. You mentioned "relevance" as a goal, which it certainly is. We should communicate information that is relevant to people's needs and interests. We should present information that matters to them. 


Agreed. But there are ways to engage without inserting silly nonsensical graphical elements in a visualization! This is were "graphical excellence" happens in my opinion.

Truthfulness is a goal but many readers, I suspect, are just not knowledgeable enough to judge whether a given chart is truthful or not (by the way a major statistical/visual literacy challenge we have in front of us!). This connects to the problem of trust and credibility which is also not very well understood in my opinion.

Therefore, I believe that when we talk about comprehension, we need to distinguish between comprehension of the data and its meaning as well as being able to parse the chart in the first place. Barbara Tversky has done some fantastic work in this area.

Anyway ... Getting back to engagement, I believe it can be a major concern under specific circumstances. A little anecdote: I went to the nysci science museum with my kids a few days back. They have fantastic scientific explanatory pieces over there. If something does not look interesting, even before being informative, my kids would not even go there. They would just skip it. But once something caught their attention, they would listen to the story, try out things, ask questions, etc. I think it's the same in vis in some contexts.

Let me state this again: shall we then use useless embellishment to attract people to the degree of harming the quality of the content? No. But we do need to take this aspect into account and encourage visualization designers to seek "graphical excellence", which includes being truthful but also beautiful and attractive sometimes.

Quote:
I don't agree that "many people have the practical problem of making their graphics attractive in the first place and then engaging, so that people do not just fly away after a few seconds." This is sometimes a concern, but not nearly as much or often as one might assume based on the amount of attention that it gets. The vast majority of data visualization by far is done in the workplace to provide information that people need to do their jobs. Their attention does not need to be grabbed. It is in the context of web-based infographics, meetings in which people are being bored to death by a series of slides, etc, when the goal of attention-getting comes into play. Most people who write about data visualization seem to think that everyone who produces charts are professional graphic journalists when, in fact, very few use graphs in this way. Most data visualization practitioners need to focus on clear, accurate, and truthful communication, not on eye-catching graphics.


I disagree with you here. Let's face it: one of the biggest problems of our times is the huge amount of information we are bombarded with every single moment of our life. Most people split their attention among a myriad of little data sources (anyone attending a talk and tweeting or checking emails?). In an ideal world this should not happen, but in reality it does happen. All the time. I believe this is one of the biggest challenges of our times by the way.

This happens of course to people surfing the web but also in many other contexts where it should not happen. It happens in the classroom, it happens in business meetings, it happens everywhere. People's attention does need to be grabbed and sustained because otherwise they would fly away.

Again, does this mean we should hang flowers to every x-axis? No! But it is a practical issue many people do have. I see this in class all the time. I have to force my students to close their laptop and listen to me, otherwise they would just get lost.

I think a very good example is the work of Hans Rosling. His charts are accurate and truthful. He uses "proper" encoding. And his message is very relevant. But he also knows how to grab people's attention and make a show out of a few scatter plots. The way he moves, the way he talks and, yes, the way the charts look like are all really good engagement devices. This is fantastic work! Which is beautifully integrated and does consider all the aspects of visual communication; engagement included.

I agree. Not in all contexts this is needed. But in many it is. And many of them are indeed very relevant, even beyond journalism (which is of course also relevant).

One last note: I am not trying to win an argument, I am just trying as honestly as I can, to express my ideas and better understand yours. Hope this is going to be productive.

All the best.

__________________
Enrico Bertini
sfew

Moderator
Registered:
Posts: 803
Reply with quote  #57 
Jeff,

Thanks for sharing the link to your recommendations for improvements to infovis research. I think this is a great start.

__________________
Stephen Few
sfew

Moderator
Registered:
Posts: 803
Reply with quote  #58 
Enrico,

You and I are basically on the same page. What you call "engagement" however, is what I call "attention getting." I think of engagement as something that happens when someone reads or interacts with a data visualization. Getting someone to look at a data visualization happens before engagement can occur. Another slight difference in our perspective is the degree to which we think "attention getting" is necessary. This difference is probably related to the differing nature of our work and experience. I primarily work with people who use data visualization either for data sensemaking, which is part of their job to support the needs of their organization, or for data communication, to inform others in their organization about data-based evidence of what's going on. This is not a context in which "attention grabbing" is usually needed. People are naturally interested in the data that they need to do their jobs--or at least should be.

Here's another concern of mine. Too much emphasis is placed on "attention getting" in web-based content about data visualization. Most people who write about data visualization are not particularly concerned or even aware of the ways that most people use data visualization. They are more focused on web-based infographics, which makes up a small percentage of data visualizations. By focusing to this degree on attention getting, many people who need not be concerned with this are encouraged to spend their time dressing up their data visualizations in ways that aren't necessary. Because they lack the considerable skills that are necessary to make something eye-catching without undermining the integrity of the data, they end up doing harm to their data. Like you, I love the work of Hans Rosling, but few people possess his skill or have access to a BBC video crew.

Here's another concern. You mentioned that you must do something to get your students to pay attention in class. I understand that, but I'm concerned that we should focus on really effective teaching to grab and keep their attention rather than giving into their desire to be entertained. I'm not suggesting that you personally engage in senseless entertainment. I'm merely pointing out that by giving in more and more to the tactic of entertaining people to grab their attention, we are not helping them become good learners. There is a line that we shouldn't cross if we want our students to develop good learning habits. The same is true of people in businesses. If you spend time making your charts entertaining, you are engaging people in useless activity rather than encouraging them to become engaged with data in meaningful ways. Our organizations will not become more analytically savvy by giving into the desire of people to be entertained as opposed to informed.

__________________
Stephen Few
sfew

Moderator
Registered:
Posts: 803
Reply with quote  #59 

Robert Kosara of Tableau Software and the blog EagerEyes.com wrote a response to my article defending the infovis research community. I would encourage you to read it because it vividly illustrates my concerns. Robert previously taught infovis for years at the University of North Carolina and has done many research studies himself, yet his response to my article is not rational. Read for yourself what he wrote. Notice the many rhetorical fallacies he committed. In particular, note the number of straw men that he constructed, misrepresenting my position, so that he could strike them down and claim victory without actually addressing the issues. Also notice the way that he casts aspersions upon my character rather than responding to my arguments.

Once you’ve read it for yourself, go ahead and read my observations below.

Robert countered my statement, “I suspect that her [Borkin’s] studies of memorability were dysfunctional because she lacked the experience and training required to do this type of research” by pointing out that she attended Harvard. This is not a rational response. This is a rhetorical fallacy. The fact that she attended Harvard for a portion of her studies provides no evidence that she was trained in the scientific method or that she applied it properly in her work.

Robert stated that my criticism of Borkin’s paper was “odd” because I previously praised a paper that she did in 2011. Yes, I pointed this out myself in the article. I also pointed out that the paper she did in 2011 was of a different type, which she was apparently qualified to do and do well.

It’s interesting that, after pointing out that I praised Borkin’s 2011 paper, he suggested that I have singled out women in my negative reviews. It is easy to survey my work to see that this isn’t true. I have written negative critiques about the work of several men and I have praised the work of several women, including one of Robert’s former students when he asked me to write a letter of recommendation for her, which I was delighted to do. Robert is well aware of the fact that I am not discriminating against women, yet he felt compelled to make this accusation. For what purpose? This is a rhetorical trick. When you cannot oppose someone’s position rationally, you can always attack their character.

Robert joins others in arguing that “the tone and personal style” of my critiques “are way over the line.” He does not provide examples of this, however. I have invited those who make this claim to provide examples, which they have not done. Isn’t it interesting that he considers my style aggressive yet is quite comfortable making inflammatory remarks about me?

Robert suggests that I insist on inappropriately applying the scientific method to what I call “Software Development Studies” and he calls “Technique Papers.” Studies of this type propose and usually test the effectiveness of new visualizations or visualization techniques. Robert’s claim is a straw man. I don’t insist on this if the paper does not include experimental tests. A paper that proposes something new without testing it contains nothing to which the scientific method applies. If experiments are included, however, they should confirm to the scientific method.

Robert countered my assertion that “visualizations don’t need to be designed for memorability—they need to be designed for comprehension” by stating that the authors of the research were correct to focus on memorability alone. He implies that I suggested the study should address both memorability and comprehension. This is a straw man. I agree that a single study should not try to measure both because it would be too difficult to control.

Robert constructed another straw man when he suggested that I claimed that “visualizations are disposable: look at it once, make your decision, throw it away.” Rather, I stated that it is the information contained in visualizations, not the visualizations themselves, that are useful, and that in most cases that usefulness only lasts until the decision that it informs has been made. He failed to point out that I also explained that information in visualizations can have lasting value, but even then it is not the visualization that must be remembered, but the information that it contains. He went on to suggest that I don’t value visual memories as useful, which is nonsense.

Robert endorses Borkin’s studies of memorability by calling them “trailblazing.” He doesn’t explain how they have broken any new ground, however. It would be difficult to explain this because the two studies have not contributed useful knowledge.

Robert summarizes his response by constructing another straw man: “Instead of insisting on particular arbitrary demarcation lines, why not let a thousand flowers bloom and see what happens?” I have not insisted on arbitrary demarcation lines. I would love nothing more than to see a thousand flowers bloom. It is for this very reason that I have expressed my concern about poorly done research. Flowers cannot grow from infertile soil. Similar to Ben Shneiderman’s claim that I snuff out candles, Robert’s final words characterize me as “shooting the bird” of infovis research that is “struggling to emerge from the nest.” So now I'm a killer of baby birds. This is a false analogy.

I’m more than a little concerned that leaders in the field of infovis research are resorting to rhetorical tricks to counter my arguments rather than engaging in rational discourse. Jeff Heer is the only member of the community who has taken the time to respond to the content of my article in a rational manner. What Robert has written underscores my concerns about the infovis research community. This is not the level of clear thinking and rational discourse that is needed to advance the work.

Let me say once again what I’ve said several times already: I care about infovis research. As someone whose work relies on the research community to find ways to extend our reach and improve the efficiency and effectiveness of our work, I care deeply about it. This research will not improve, however, if the community responds to valid criticism by circling the wagons and cutting itself off from the outside world where data visualization is applied every day to solve real problems. Research papers that make invalid claims should not be published. Misinformation does harm. In the world where I work, the harm is real, not something that we can brush aside as merely academic. The research community should prevent this from happening. If it did, there would be no need for others like me to expose its flaws. I’m encouraged by the efforts of people like Jeff Heer and his colleagues at the University of Washington to address these concerns, but this is not enough. The first step to recovery is to admit that you have a problem. In those rare instances when leaders do this, we appreciate it and respect them all the more.


__________________
Stephen Few
danz

Registered:
Posts: 182
Reply with quote  #60 

Steve,

I rarely take anybody work, including yours just as it is. I use to take my time to dissect the contents, to connect it to my own knowledge, to read adjacent information, to shape my own opinion and only after, in case I consider is needed, I provide a comment on it.

Your latest article was one of the fewest you ever wrote it went straight to my heart. Except your counterexample article, for which I expressed my own concerns in an email to you, the rest of your article is a reflection of my own thoughts. I am not interested in the critics you received regarding your tone. It is a pity that your forum does not allow hiding comments we are not interested in, so the whole thread can be reduced for each participant to just a few comments of interest for the topic.

I will focus now on Ben Jones blog about this matter. As a practitioner, he wants to have his work memorable. Based on his experience and own perception memorability can or should be the goal of data visualization. "I wanted to give them an image that would remain in their minds until the point of decision had arrived. So yes, memorability was a specific goal of mine in the visualization design process." - Ben Jones. This is a clear statement in favor of having memorability studied and therefore used in data visualization.

I am confident that his solution was not studied for 10 seconds only. Instead, his redesign caught the attention of his audience, so it triggered a more complex investigation process which eventually leads to a memorable matter rather than a memorable display. Let's ignore for now that iconic memory does not last. But we cannot ignore the fact that the real reason his solution was remarked was his ability of finding the issue and emphasize it using an appropriate technique. Was not the design itself the one to be remembered, but the matter he found which probably urged an immediate action.

Data visualization is a powerful way to communicate information. Our goal is to make it comprehensible. I am a practitioner myself, but I cannot agree with Ben Jones. It is nothing wrong that we want our work to be remarkable and remembered. We can do that providing effective designs using a highly developed sense of data rather than developing or using memorable techniques. 

Steve, when you wrote “we are overwhelmed by information, not because there is too much, but because we don't know how to tame it” you could not foresee the avalanche of noise in data visualization research area, could you? If you would have known that I am sure you would have write a different intro.

Dan Zvinca

Previous Topic | Next Topic
Print
Reply

Quick Navigation: