Visuale

There is an interesting intervew with John Stasko at mentegrafica published recently. I really liked the point made by John about "solutions in search of problems" style of research in InfoVis. Since recently (and today still) we have lived with this way of producing visualizations: inventing a new technique for the sake of it and then trying to find a practical application, i.e., an appropriate problem.

There is a plethora of examples, especially on the web. And many are pointed by this blog too, because they are just fun! After all, why are we so attracted by visualizations? You can respond in many ways but I'm sure that the deepest feeling and the real reason is that they are just cool: beautifull to see, engaging, colorful, etc. However, I agree with John's point and I also believe that in order to make a dent in this world (quoting steve jobs) we must care about the utility of things we make and struggle to find approriate solutions for specific people.

The same and similar points were raised at the BELIV'06 workshop we organized last May 2006 in Venice (colocated with the AVI Conference). It was a really enjoyable event and we discussed a lot about these things in such terms.

Anyway it's worth to note that this is not unique to infovis only but also to all the sciences which actually are engineering (thus almost any areas of computer science). This is just the way new scientific fields develop. At the beginning some small niche of people get interested in the concept and produce new ideas, new prototypes and the rest. Then, when the field becomes more mature, people start asking themselves if these applications are useful or not. I am glad today we are facing these questions because it means we are entering a new phase and thus the field is getting more mature.

One side effect of this is that it is becoming more and more difficult to publish papers in good conferences (like IEEE InfoVis Symposium). Now the reviewers expect to find strong claims about the utility of the proposed technique, real improvements over related solutions and some sort of experiments/tests that demonstrate the quality of your work in practical settings. Sure, I understand this approach can be questioned in that it might discourage the production of really novel ideas, but still this is the classic evolution of research fields. The same has happend and it is still happening at Siggraph for example (probably the biggest conference out there), which has a much longer tradition than infovis. See for example this guy who retired because disgusted about the way papers get reviewed at Siggraph, leaving only this laconic message in his page "in summer 2006 I will be leaving Stony Brook University ... you are interested to read above my reasons, click here". Similar complaints happen at CHI conference too. See these funny arguments made by Henry Lieberman, and Shumin Zhai's response.

So ok, I don't want to push this thing too far ... anyway I really liked John's expression and I think I will use it many times in the future for explaining why nice ideas are not enough, why beautiful images are not enough (apart from artistic purposes?), and thus why our ultimate purpose as researchers/designers is to help people accomplish their tasks better (more efficient, more fun, more effective), quoting F.Brooks's famous Computer Scientist as Toolsmith (pdf):

If we perceive our role aright, we then see more clearly the proper criterion for success: a toolmaker succeeds as, and only as, the users of his tool succeed with his aid. However shining the blade, however jeweled the hilt, however perfect the heft, a sword is tested only by cutting. That swordsmith is successful whose clients die of old age.

I know some people do not agree with me, but this is the way things progress: debating about different ways to view the world. Enjoy! :-)

The program of the next InfoVis 2006 conference (the premier conference in the field) is out. I'm really surprised to note how large the proportion of papers about trees, graphs, and networks is.

There are 13 out of 24 papers, that is, more that 54%, classified either as: Graph Exploration, Network Visualization, Tree and Treemap Applications, or Graph Drawing.

I'm really surprised! I wonder why. Is the case that infovis is becoming the visualization of networks and such things only? Are we loosing ground in other areas? Or maybe is just a random variation happening this year?

In order to see things a little bit closer, I checked the last editions of the conference to see how the proportions have evolved. Here are the percentages:

- InfoVis 2006: 54% (13 out of 24)
- InfoVis 2005: 35% (11 out of 31)
- InfoVis 2004: 18% (5 out of 27)
- InfoVis 2003: 13% (4 out of 29)

What do you think, is it a real trend?

Anyway ... this year I will probably not go there and it's a pity since the conference is always so great. This year there is also the first VAST conference on visual analytics that seems to be a perfect companion of the symposium with some more real worlds studies, which is fine.

I'm also glad to see my friends Geoff and Alan with the following paper accepted to the conference:

Enabling Automatic Clutter Reduction in Parallel Coordinate Plots
Geoffrey Ellis, Alan Dix

It is the result of a research on clutter reduction we started together some time ago, so I'm glad they have developed things so much further. Good job! ;-)

I've just found this web page with links to almost all infovis 2006 papers.

It's nice that so many authors publish their papers online before they become officially available in the common repositories (e.g., ACM digital library).

And it's also nice there's people like Hong Zhou, spending her own time collecting and making them available online for the rest of us. Thank you!

I found really fascinating this eyetracking heatmap form one of the latest Jakob Nielsen's Alertbox articles.

The areas where users looked the most are colored red; the yellow areas indicate fewer views, followed by the least-viewed blue areas. Gray areas didn't attract any fixations.

As argued by Nielsen, the dominant reading behavior follows an F shape: a first horizontal movement, a second horizontal movement a bit below, and a vertical scan on the left, and it is quite consistent over several different types of web pages and tasks.

It's incredibly fascinating to see how visualization can help in tracking and then displaying human behaviors in a way that makes emergent patterns readily visible. There is a growing interest and focus on collecting bits of human information and given them a visual counterpart ready for analysis. Other examples have been posted in this blog before.

The key factor of this kind of visualizations is the fact that visualization permits to see the evolution of things summed up in aggregate visual information. Frequency is often the key parameter that is used to visualize behavior: frequency of road paths, visual glimpses, airplane trajectories, etc. Alternatively, the visualization can be used to replay things at a different rate and thus to expose dynamic behaviors that couldn't be detected at a human pace.

It's a notable fact that replaying things at a different rate permits to see some patterns that would not be apparent otherwise. The image that comes into my mind is the typical video of a flower's entire life replayed in a few minutes (see this wonderful Amaryllis). Or a sequence of pictures taken always from the same position over an extended period of time. A more elaborated idea is Alan's Slow Time: if we could replay our life and use band-pass filters to select things that routinely happen at very low frequencies (e.g., every day, month, year) we could notice interesting recurring patterns of ourselves. Notably, finding your keys!:

Take the home. Band pass in the millihertz range. Start off by removing the things faster than millihertz - to and from movements of people disappear and we are left with the things that change only slowly - rather like a series of long exposure night-time frames, the hurrying people become ghostly mist against the background. Now remove the things that do not change in the millihertz range, things that are the same for hours and days. Watch the resulting images, just the moments of millihertz change - the things put down and picked up, or perhaps put down and never picked up - "that's where I left my keys!"

It would be great to see one day this idea realized in a real prototype.

I find it quite bizarre that the name information visualization does not contain any explicit reference to interaction. And that is true even if its traditional definition (and others) mention it clearly:

"the use of computer-supported, interactive, visual representations of abstract data to amplify cognition" (Card, Mackinlay, & Shneiderman, 1999)

After all, there should be a difference between paper-based static visualization (i.e., Tufte's like) and interactive visualization, isn't it? There should be some added value when interaction is taken into account.

Maybe the name would be too long or complex with additional words? Or maybe who invented the name (my guess is Stuart Card) implicitly intended to include interaction in the name as it is? I don't know. However, I think it is a matter of fact that information visualization is often perceived only as the science of finding the best visual representation for depicting data, completely neglecting the primary role interaction should have.

I don't want to seem a purist, but personally I think we should be aware of this fact when speaking of visualization and, more importantly, when we teach about it. In my experience, students and newcomers, get stick with the idea of representation and completely forget the other half of the story.

These can be some of the causes:

• The name. As I said before, the name information visualization suggests the idea that it is only about visualization that we talk about and not interaction. Notably, the information visualization research page at PARC (where infovis was, partially, invented) is now called "Information Visualization & Interaction".

• Visualization is "visible", interaction is not. When showing visualizations to people, what really stands out at first is the visual part not the interactive. Colors, shapes, regular patterns, remain impressed into one's mind and create a pleasurable experience that strongly influences the perception of it.

• Heritage from visualization gurus. Who wants to learn how to visualize data is often (and for a good reason!) pointed to classics like Tufte's "The Visual Display of Quantitative Data" or Cleveland's "The Elements of Graphing Data". Unfortunately, however, these old gurus teach you very well how to transform data in visual structures but not how to interact with them. Moreover, the good books that teach you also the interactive side (e.g., Readings in Information Visualization) seems to be a lot less structured and organized in its description than for the visual part.

• Interactivity is hard to program and thus to practice. As demonstrated by the plethora of web visualizations with few interactive capabilities, it is relatively easy to program few lines of java code to transform data into a visualization. Much more harder (and time consuming) is to program interaction; dealing with selections, filtering, coordinated views, etc. The result is that there are much more static digital visualizations than interactive ones.

• Interaction is hard to teach. Visual mappings can easily be shown through images in books and slides in class, and be profoundly criticized through inspection. Interaction requires dynamics and real demonstrations. Ideally, students should directly "do" the things to learn and criticize and this is not always feasible/possible. Second, the body of literature around interaction for visualization is much more dispersed and not very well developed. While it is quite easy, for instance, to find taxonomies, empirical studies and classifications, of visual techniques, the same does not hold for interactive techniques.
  

It is true that visualization on paper can also help in reasoning and exploration, to some extent, but its main purpose is to communicate and elucidate something that has already been elaborated and digested by a designer. It is only with interaction that one can really reason about a data domain, explore it under different points of view, and manipulate the medium to accommodate new questions coming to mind.

There are various reasons why interaction is really fundamental and a big leap forward with respect to static representations. As I was saying, data exploration is useful to find new questions more than clear answers (notice that this last feature seems to be at the same time the strength and weakness of information visualization), therefore interaction is really fundamental because as soon as new questions pop up, new views an mappings are necessary. Also, when the information to put on the screen is too much, as often happens, it is necessary to select what must always be visible on the screen and what extra-information can only be accessed by means of mouse clicks. Without interaction the designer is forced to put all or nothing.

It's strange the fate of interaction. The measure of its success is the extent to which it gets unnoticed, that is, how naturally users can manipulate things without too much cognitive effort. However if it is successful, it get unnoticed, and thus nobody remembers it.

I'm eager to come back to these points. I feel there is much more to say about it. What do you think? Is it important or not to deal with this type of questions (after all it is just a name!). Am I missing some important aspects of it?

It's when staring at images like these that I think the best visualizations ever are those offered by the nature.

[from NASA Visible Earth, click on the images for details]

It's amazing to see how informative and beautiful aerial and satellite images are. I try to mentally compare the process of producing visualizations of abstract data (i.e., information visualizations) with that of producing representations of real world object, as these pictures above are.

It's incredible how much more subtle and beautiful these images are compared to those obtained from information visualizations, which have regular grids, well segmented areas, few details, less sense of visual continuity and uniformity.

And yet, at the same time, it's crazy how much more effort it's required to design a visualization and how much more uncertainty there is in it. It looks like the mental and imaginative effort required to a visualization designer is much more because the number of free parameters to set is very large. Design decisions can really differentiate between a good and a bad visualization, and thus much more responsibility is given to the visualization producer.

These few considerations make me recall of a recurring debate on what's the difference between "scientific visualization" and "information visualization". I don't care too much about the terminological debate, nor am I in favor of a neat distinction between the two (quite some people went over it already). However, for the purpose of this post I think it's very interesting to re-propose here what Tamara Munzner once had to say in a panel (pdf link to the panel content) hosted by Vis'03 about the topic:

"Although there are still many definitions of infovis floating around, I think we have begun to converge on the answer that the dividing line is whether the spatialization is given or chosen."

"... The central design problem of an infovis system is the choice of how to assign spatial position, which is by far the strongest of the perceptual cues. Our grappling with this huge space of possibilities has led to a strong emphasis on abstraction, visual metaphors, design principles, and evaluation."

Pushing things to the very extreme, the result is: we visualization designers put much more effort to build things that are thousands of time less beautiful than what? A simple picture.

EagerEyes recently proposed a post titled rethinking the user, treating what I think is one of the most interesting open points of InfoVis: who is the user in infovis?

In recent years, information visualization reached a typical maturation state where people start to retrospectively critique the field, highlighting gaps, misconceptions, unexplored areas, false myths, etc. And the question "who is the user?" is in my opinion one of the most relevant, and with most consequences on how we will perceive and develop the field in the future.

As pointed out by Robert in the post, the problem of infovis is that we often design our tools in a way that they are somewhat easy for us to use but extremely difficult for domain-experts:

"The current view of course is that visualization systems should be designed in a way that is useful to domain experts with little to no knowledge of visualization. In reality, of course these are not the actual users of visualizations: rather, the tools are run by their developers, in communication with the domain experts."

Robert suggests also to classify users in two classes: visualization experts, who develop the tool and use it for others and casual users, who do not develop the tool but use it for simple purposes. I think this classification is too gross and that we need to dig deeper into the question. For instance, does a visualization expert necessarily have to be a visualization developer? I don't think so. Does a visualization user necessarily have to be a casual user? Maybe not. And then, after all what do we mean by visualization expert? And who is a casual user? What do they produce and consume?

I started thinking along these lines on how to classify people and tools and here is the result of my reasonings so far.

User types

• Visualization tool designer: a person who actually implements (and often designs) a visualization tool. It is very rare that a visualization developer is not also a visualization designer and thus a visualization expert (most visualization researchers fall in this class).
  
• Visualization designer: a person who actually designs visualizations but does not develop the tool to design them. To this category belong, for example, all the consultants who produce business intelligence visualizations (e.g., dashboards) for companies. Less often, but more frequent now, there are designers who design custom (interactive) visualizations by means of what I call "meta-visualization tools" (see below).
  
• Domain-expert: a person who uses a visualization as a support tool to analyze data in his specific domain (the person I picture myself when I think about the slogan "using visualization to think") (e.g., in business, biology, security, etc.). He uses it to make sense of data, discover new facts, and take informed decisions to act in his own field.
  
• Casual user: a person who uses visualization for short amounts of time without a clear goal, mainly because it's cool or to investigate some data of general interest and discuss about it. Most visualizations found on the web attract this kind of people. A recent trend is to analyze things like census data collaboratively to extract information of common general interest.
  

Visualization types

• Meta-visualization: A visualization tool that permits to create visualizations. It is generic both in the kind of visual techniques that can be generated and the specific domain or data type that can be analyzed. Example:Tableau.

• Generic visualization tool:The tool is general enough to handle data coming from different domains, but a limited set of techniques and interactions are allowed. The degree of generality can vary however. Examples: TimeSearcher (time series), IN-SPIRE (visualization of text corpuses), HCE (biology experiments).

• Domain-specific visualization tool: The tool is designed to support people working in a specific domain. The visualization is optimized to represent relevant data features from the domain and to support custom typical tasks. There are numerous examples from several domains. Examples: Rumint (Network Security), Visual Sciences (Web Analytics), Visual i|o (Business Intelligence).

• Visualization instance: A single screenshot or printed graph to communicate important findings to interested people. A lot of tools, especially in business intelligence or web analytics, provide functions to create ready to print summaries to resume and communicate relevant information.
  

Current practices and limitations
I've designed a table to show what is produced and what is used in visualization. The table links user type to visualization type indicating if the user type mainly uses or mainly produces the corresponding visualization type.

After having segmented users and tools I think it's time to wrap-up and share with you some ideas I have on it.

• A visualization tool designer produces almost all kinds of visualization. Often the problem here is that the designer is not aware for whom he is designing the tool. And this happens especially in research, where people is not enough motivated to ground their research on real-world cases. A meta-visualization should be designed for visualization designers not domain experts! A generic visualization must be tested on specific domains to see if it works equally well in all of them. A domain specific tool cannot avoid involvement from domain experts.

• A visualization designer is often a consultant. It's an interesting figure, and not enough supported in my opinion. A visualization designer is an expert and often knows very well what works and what doesn't in the real world. I have the feeling that current tools are still not flexible enough to support these people. Either the tool is too specific to let the designer invent new techniques or too generic and complex to find solutions in reasonable time. Excel is the current standard for static visualizations. For interactive visualization we still lack a working tool.

• A domain expert can be a person using the tool to find things for himself or for others. As an example, a biologist uses the tool to explore the results of his own experiments, a web-analytics consultant uses the tool to provide findings and, hopefully, suggested solutions for others. It's important to make the difference between the two because the first is supposed to be less expert (and less passionate maybe) in visualization than the second. But still our goal is to be able to support both of them. Too often, in my humble opinion, in research we forget to partner with these people when building something for them. We pretend we are their domain's expert as well as visualization experts.

• A casual user can be anybody. I included this type because I feel it describes well all the people playing with the thousands of little visualizations we can find in Internet. Some tools like Swivel or Many Eyes are useful to discuss and share ideas with visualization. Some others are just curious but nothing more. I suspect that the large majority of these visualizations are used only few times and then are forgotten. Are they useful? I don't know. A good thing about that is that they contribute to the diffusion of interactive visualization among people.
  

What do you think? Am I missing something? Am I getting it right?

Wow, only now I realize it's around one month and a half from my last post on Visuale! Well ... I have a good reason this time. Here is the main and beautiful "cause" of my dormancy.

My first son Matteo was born on last 14 July and as you can imagine I didn't have much time for Visuale since then. "Ubi maior minor cessat" :-)

Here is him making his own "visualizations". Most probably a million of times more beautiful than the ones I will ever develop in my life.

But this does not mean I didn't have time to think about visualization. To the contrary I have new energies to spend! I'm ready to post new things soon and to be as regular as possible with my postings.

Here is a summary of probable upcoming posts buzzing in my mind for the next future:

• Challenges of evaluation in InfoVis and BELIV'08: assessing the quality of a visualization tool is still today a great challenge. Each of us uses custom methods and relies more on experience than recognized and shared methods. This is the topic of BELIV'08 a workshop I'm organizing with other people at the next CHI 2008.

• Web-based visualizations: I'm sort of "obsessed" these days by the idea of web-based visualization. What are the unique challenges and opportunities of visualization on the web? What technologies are there ready to be used? Does it make sense to develop desktop-based visualization yet?

• Vis/InfoVis/VAST Review: The program from the IEEE Vis/InfoVis/VAST co-located conferences is out and it seems very rich of good papers and events as usual. I plan to speak a little bit about it soon and then again after having attended the conference.

• The SpiralView: We have a paper published at VAST 2007 this year about a system named SpiralView that we have developed for the analysis of intrusion detection alarms in private networks. It is a joint effort between the University of Fribourg and NEXThink, a company in Lausanne specialized on the topic. I would really be glad to share some ideas on that with you all.

• Current trends in visualization: I have the strong feeling that something is happening today in visualization. From the research point of view, I really believe that we are starting to address some problems at a higher level, leading the whole field to a new maturation stage. The same it's true form the business side. Visualization is really gaining momentum. I'll try to expose some of these trends I see.
  

I'll try to publish these and other posts striving, as usual, to be as informative as possible. The number of visitors in Visuale is steadily increasing, I'll try my best to make it better and better. Thank you.

The VIS/InfoVis/VAST conference has been as usual a great event, with lots of good presentations, events and interesting people to meet. The conference venue was really nice (trivia: the Hyatt Hotel is the place where Governor Schwarzenegger lives when he is in the capital) and I definitely enjoyed some Californian sun and was really pleased to find old and new friends.

Despite the high-level technical program and the abundance of good presentations my best moments have been the InfoVis keynote and the VAST keynote respectively by Matthew Ericson, from NY Times, and by Stephen Few, from Perceptual Edge. I was really happy to see these two guys describing, from very different perspectives, things on information visualization as done in the real world: for people who don't want to spend excessive efforts to understand what a visualization means and by people who are not traditional visualization researchers/developers. It was a breath of fresh air for my mind.

I'll start with Ericson's talk here in this post. I intend to write something about Few's talk too in another post.

Matthew Ericson described quite in detail the work they do at NY Times to produce effective visualizations that are informative and easy to understand at the same time. I was impressed by the quality of their work and the heterogeneity of the people in the group. And I was also impressed by Mr. Ericson's argument that they consider themselves first of all "journalists" rather than designers. As such, their primary purpose is to tell a story to the reader. Looking at the graphics produced it's impossible to remain indifferent, the eye and the mind are suddenly engaged, there are stunning visualizations, complex and simple at the same time. Each piece is extremely rich: annotated with concise and well placed text notes, multiple tiny views arranged in a way that the whole set tells a story, pictures and/or diagrams added when/where needed.

I also liked the concept of "honest portrayal". Tufte and others have for a long time warned us about the dangers of visualization; for the very fact that it is so potent in conveying information, it can also be used to send wrong or partial messages. Mr. Ericson goes a little bit further, in my opinion, saying that it is important to keep always an eye open to that fact that visualizations may convey partial truths and, more important, that often in order to convey the whole picture a single visualization is not enough, it is necessary to present the data under different perspectives. The example of the US 2004 elections made the case clear.

The picture is not necessarily "wrong" or purposely "false", but still it contains a partial story that can be misinterpreted: the amount of red in the map is enormously higher than the amount of red because it represents only two values: Bush (red) vs. Kerry. But the picture does not tell anything about margin of votes ... and neither about the population density! Here is how the maps have been reworked and assembled in a full story (click on them to see a big picture).

Another element of interest of the NY Times people is how fast they can produce these graphics. Ericson explained that they work in very very tight schedule because they have to follow the news when they are hot and cannot wait weeks or months to produce a story.

What remains totally obscure to me is what kind of tools these people use to produce such a beautiful and complex graphics in such a short amount of time. Especially because the kind of visualizations, charts, and diagrams they design are not at all trivial and I would bet that most of the time they have to mix the outputs of various tools. Being able to turn data into pictures in such a short amount of time looks to me some kind of magic.

In short what I learned from the talk is that if we want to reach the large public with visualization we have to take care of every detail and present a beautiful, rich, engaging and self-explained piece of work. Sure, this does not take into account how people would interact with interactive visualizations when provided with them, but still I have the feeling that the same principles remain: design complex and composite solutions to provide depth and richness and, at the same time, strive like crazy to make it simpler, simpler, and simpler. This is what most of the time people need to reason about their data.

[Talk's slides from Matthew Ericson's website (Zipped PDF)]

Since the term Web 2.0 was born, a number of tangential fields have also utilized the same terminology to indicate how they have been shaped by it. Notable examples are: Business 2.0, Enterprise 2.0 and even Bubble 2.0! :-).

At the risk of seeming a hype follower, here I ask the question: can we speak of Vis 2.0?

[See the classic Web 2.0 article by Tim O'Reilly to learn more about it]

When I think of Vis 2.0, my intent is to group all the recent visualizations I have seen appearing on the web under a single label. I have the strong feeling that something new is really happening and that the whole domain of visualization is being transformed by the forces acting on the web. Its future shape will depend a lot on it. Think about it: even the sole thing that a visualization, once it is designed and developed, can instantly be made public and potentially reach millions of people is a revolution in itself!

Following the tradition of defining Web 2.0 by the observed new patterns vs. old ones, here I provide my personal list of patterns:

• Web vs. Desktop: The application is distributed over the web and accessible through a simple web browser. No installation, no configuration, no hassle.

• Communication vs. Exploration (and Discovery): The traditional open-ended task explore and "discover the unexpected" is somewhat subverted. The vast majority of Vis 2.0 applications are meant to communicate something that cannot be seen in raw data but that when visualized is quite obvious to understand. Sure, it is clear that discovery and exploration are still attractive and can be largely encouraged, nonetheless this does not seem to be the driving force anymore. In addition, many tools seems to have a clearer goal, a direct connection between task and tool. Traditional InfoVis applications, conversely, have always suffered this limitation of being able to do anything and nothing a the same time.

• Many and Diverse vs. Single and Specialized User Base: Users come from many different sources with a whole spectrum of interests an goals (often curiosity). The visualization is there ready to be observed and used in way that could not be anticipated by its designer. Compare this to the traditional data analyst, using a very technical tool and spending hours alone figuring out what's in the data. Some systems also allow collaboration and discussion, which is another revolution. The user is not alone and the task is never fully ended. People discuss around a topic aided by visualizations. It's the full power of the collective.

• Small and Targeted vs. Large and General Purpose: If we consider the nature of the the tool we have an opposite trend. If the audience is generic and diverse, the tools are small and specific. Forget monolithic desktop applications connected to huge enterprise data warehouses issuing complex queries to data cubes. Here we have thin tools with a single and often very simple and clear purpose, which becomes obvious at first sight. The interaction is often limited, what you see is really the only thing you get, but the purpose is clear. Nothing less, nothing more.

• Shallow vs. Deep Interaction: The tools appearing on the web often employ very limited interaction techniques. In fact, the tool is not mean to be used for complex tasks, few clicks and the job is done.

• Funny and Empathic vs. Cold and Technical: Many of the realizations on the web bring a lot more emotional involvement compared to the traditional tools. This is somewhat paradoxical if we consider the power of visualization to bring not only information but also emotion. The desire to convey emotions is so evident in certain web visualizations that would be an error to consider only as sporadic events (see Iraq War Coalition Fatalities for an excellent example).

• Maps and Charts vs. Fancy Visualizations: A very large segment of web visualizations is realized with maps and simple charts like: bar charts, line graphs, sparklines, etc. It's true that there are also many "esoteric" designs out there but they seem to cover the less useful portion of applications. Traditional InfoVis has instead a clear bias towards creating always new visual designs, often completely useless. This is a very personal consideration, but I strongly believe that we yet have to discover the full potential of simple charts when more clever interaction schemes are attached to them.

• Scripting and XML vs. Java and DBMS: I'm getting a bit technical here, however, even in terms of programming models there are some new trends. Web visualizations are realized with lightweight programming models using technologies like: JavaScript and Flex for the UI and remote and asynchronous data access with XML, JASON and similar technologies. Traditional visualizations, realized for desktop environments are realized with more solid languages like Java and similar and often retrieve data from complex DBMS.
  

One thing I want to clarify is that I'm not necessarily assigning a positive or negative value to these new trends. In fact, I believe there are good and bad things about them.

As an example, I have already said before in this blog that I'm not very excited by the proliferation of badly designed, too simplistic and often useless visualizations on the web. But, it is also true that they have greatly helped spread the word about visualization as an interesting field.

Before closing this post I want to provide a list of examples of Vis 2.0 tools that I consider really great:

we feel fine: for its great emotional value
crimespotting: for its usefulness and advanced interaction
hindsight: for its beauty
hotpads: for their heatmaps
google finance: for its F+C interaction
finviz: for its complexity and simplicity at the same time

That's all folks. I hope you will have something to say about this.

Since the first BELIV'06 workshop on information visualization evaluation, which I organized with some other people at the AVI 2006 conference, I started thinking about why evaluation in our field is so challenging and so poorly supported by the existing techniques.

Among thousands of ideas I have collected and discussed with peers (actually very long discussion are still going on with some of the people who participated to the last BELIV'08) there is one I think has some real value: "purpose-driven evaluation".

If we want to capture the value of a visualization we have to first understand its purpose and then find a way to quantify to what extent the system under analysis (being it a single components or a full system) support this purpose. It seems a too simple idea, I agree, but looking at the current state of evaluation it's clear how this question has been eluded for a long time and how important it might be.

The problem with current methods

Insight generation as purpose

"... it seems an appropriate time to reopen the question about what the ultimate purpose of visualization is and how it should be evaluated.

One potential claim is: The purpose of visualization is insight. The purpose of visualization evaluation is to determine to what degree visualizations achieve this purpose.

If this claim is true, then evaluating visualizations should seek to determine how well visualizations generate insight."

I'm stunned by how these few sentences resonate with my initial idea of purpose-driven evaluation. One fundamental point addressed by Chris is that to date we have not yet seriously addressed the question: what's the purpose of visualization? This papers argues that its purpose is to generate insight and then it goes on defining what insight is.

Other Potential Purposes

Specifying potential purposes of visualization is a useful exercise that can be done at different levels of detail. Here is my list of common high-level purposes of visualization in terms of goals/task:

• Understanding:To learn about a domain or simply understanding what a given data source contains.
  
• Discovery: To find new information and facts that were not known beforehand (and probably unexpected).
  
• Problem-Solving: To model and represent a problem in a way that permits to find and evaluate solutions.
  
• Decision-Making: To make decisions based on data and evaluate their quality and impact.
  
• Communication: To communicate information to an audience.
  
• Monitoring and Situation Awareness: To take under control the state of a dynamic system and react when needed.
  
• Art & Fun: To produce pleasurable artifacts.
  

If we take this list of potential purposes as reference, it's not hard to see for each item how an evaluation could be designed. If my purpose is to create art or fun, I will want to measure how much pleasure people gain with my visualization. If the purpose is discovery, I will want to measure how many discoveries people can make, and what their quality and impact is. If the purpose is understanding, I will want to measure how much information people can absorb about a domain and so on.

That's it. This is all I had to say in the urge to communicate as fast as possible the idea I had in mind. I hope it is clear enough. There are thousands of possible implications around these ideas which I don't have time and space to convey. I hope this few lines can stimulate some interesting and new thoughts in you. Comment on it if you wish! :-)

I have just read this post from Stephen Few's blog "Are visual analysis tools poised to become pervasive?" in which he speaks about infovis flawed principles as reported by Christian Chabot, co-founder and CEO of Tableau Software, during his last InfoVis keynote.

Inspired by these thoughts I have found the courage to speak out about 5 false myths (FMs) I believe we have in infovis.

• FM1 - InfoVis is about data exploration: I have heard it millions of times since I started reading papers and books on visualization, it is a sort of mantra: "infovis is there to support people in data exploration". Me myself I have also described infovis in these terms tens of times in papers and reports. But is data exploration a real activity or goal? Nobody really wants to explore data for the sake of it (apart from us infovis geeks who derive pleasure from it). Data exploration tells nothing about the goal of a user and the reason why he is willing to invest time in learning and using an infovis tool. Biologists don't want to explore data, they want to understand how genes react to certain interventions. Security analysts don't want to wander through millions of alarms, they want to spot intruders and react as fast and accurately as possible.

• FM2 - InfoVis is about discovery: This is another mantra of infovis, repeated millions of times. While it is true that infovis can help discovering new facts, its true value does not come from discovery but rather from understanding. The main reason why an infovis tool is useful is because it helps make sense of data and because it does it in a more efficient way. It permits to efficiently understand what the data has to tell. And its quality can (should) be measured in terms of how effectively and efficiently this process is supported. I remember John Stasko having said in his presentation at BELIV'06 something like this: the main activity supported by infovis is to learn about a domain. This is what we mostly want to do with infovis and this is what should be supported.

• FM3 - InfoVis is about new visualization techniques: InfoVis has already hundreds of techniques available which we can draw from. The real challenges we are confronted with are: 1) understand how to use and customize the techniques we have now to make them useful to specific problems and people; 2) how to combine different techniques in composite tools able to integrate them and get the best out of their composition (as an example why nobody tried, as far as I know, to integrate all those n-dimensional visualizations we have out there?). That said, I am not saying that inventing new techniques does not have its role or that it is a waste of time. I just believe it's time to shift a bit the focus.

• FM4 - InfoVis is about vision: I have already talked about this point in one of my posts some time ago titled: "the neglected role of interaction in information visualization". InfoVis is by no means only about visual things it is also about the way we interact with a dynamic display that is able to react as we interact with it. It is this level of interaction that permits us to efficiently manage screen real estate and allows us to reason about a domain. The big challenge for an infovis designer is not only to map data items on the screen in clever ways but also to support through careful interaction design the very tasks it is designed for. We know quite well the perceptual issues and the design principles needed to design of a visual mapping, but when we come to the point of designing interaction we are lost. The only support we have is to draw from simple ideas developed in other designs (hovering, link&brush, dynamic filtering, etc.)

• FM5 - InfoVis is about the data: We tend to see infovis as a way to support a one directional channel: from data to our brain. But this view underestimates the role of the knowledge we put into the process. When a user interacts with a visualization, he brings his assumptions, background knowledge and skills that play a large role in the interpretation of what is seen on the screen. This is the reason why two different persons can very likely end up seeing different things from the same visualization. There is another hidden channel that goes in the opposite direction, from the human mind to the data, enriching it with the knowledge that is already in our heads. Notably, infovis tools fall short terribly when there is the need to manage this knowledge and let it play a role in the analysis.
  

This is what I had to say in the urge to react to the blog post I read.

Pleeease ... feel free to harshly critic my false myths!!! They are here to be dismantled ... or even better to be enriched by your views :-)

Take care.

-----

UPDATE -- November 3, 2008

I have received quite some interesting references from Christopher Collins related to the FM5: "InfoVis is about the data". From his post:

"...there are some good examples from the VAST community where prior knowledge can be explicitly entered into the analysis process (e.g. i2's 'Analyst's Notebook' or IBM's Research's HARVEST project). My U of C colleague Torre Zuk has also done some analysis of how a physician's prior knowledge affects their decision making when presented with a visualization."

Thanks Chris for your references!

I've recently come across this incredibly good book: "Data Mining for Business Intelligence". I was at first a bit skeptical, my academic background naturally led me to wrongly assume a book on applied business intelligence had nothing more to give than the two other respected books I have on the shelf. Wrong wrong wrong!

As I started reading, chapter after chapter, I felt refreshed by a new stream of ideas, like if all those notions I had accumulated year after year could be seen from a new and fruitful perspective. The book is full of applied examples, compact, with a direct and simple language and, above all, made me finally understand what data mining is and what is it for in the real world. It is the first time I feel I can walk in the same pair of shoes of those guy in the trenches who desperately need strong technology to resolve *their* problems.

So, why do I blog this?

Yesterday in a meeting with our industrial partners I received yet another lesson. Simply put: though fancy and well-crafted visualization is useless if it doesn't help people take actions.

Ok I must admit it, this is maybe only true in business sectors (is it?) but what I come to realize is that we infovis enthusiasts are too much focused on the never ending refrain that visualization is useful to explore data and that we need it to make sense of things.

This is certainly true but this is only part of the story. Take the million managers out there. Not trained to cope with complex stats or charting tools but desperately in need to take decisions based on data. What do they need? To explore and make sense of thing? Sure, to some extent ... but ultimately to take complex decisions in a very constrained setting and tight time limits.