As I have already said in my previous post this year the InfoVis/VAST keynotes offered, in my opinion, really great moments: useful ideas to reflect about information visualization as seen and done out there in the real world. Mr. Ericson's talk has been the subject of my previous post, here I want to talk about Stephen Few's talk: "InfoVis as Seen by the World Out There: 2007 in Review" (PDF).

I had already noted during the conference how the opinions of Stephen Few generated controversial reactions in the audience. It was not a surprise then to discover that other blogs had posted about his talk and that there were some critiques around it. If you are curious about that everything can be followed starting from Stephen's reply from his blog. I deliberately decided to skip the reading of all this in order to not be influenced in the writing of this post.

As is clear from the title of the talk, the objective was to show to infovis experts, the people on the edge of innovation, how the field is perceived (or better misperceived) from the people out there and what we can do to communicate the right message to them and reach their real needs. Part of the talk was obviously dedicated to the numerous examples of bad visualizations and how people don't understand how and why they are bad. But personally I don't think this was the most interesting and exciting part. It can be considered a modern version of Tufte's gallery of junk charts.

There are two main elements I really liked from the talk: (1) the analogy of infovis tools seen as old brownie cameras, (2) the story about real infovis users.

1 - The analogy of infovis tools seen as old brownie cameras

In the presentation Stephen said in relation to the analogy:

" ... these folks try to photograph the world using ancient brownie cameras with dirty lenses and expired film and they were never trained to use. They just point and shoot, trusting brownie to do a heck of a job."

I like the parallel because it highlights two aspects of a single problem: from the one hand we need to provide to people better tools, with less intricacies and able to do a fair job even when used in a "point-and-shoot" fashion, from the other hand we have to be ready to teach people how to read and produce better visualizations even with these simple tools. No matter how good the tool is if people don't know how to use it and at the same time no matter how good you are if you always get stuck when trying to design a new visualization.

I believe the parallel with photography is really pertinent and that can be extended further. The large public in photography is served with point and shoot cameras that do everything for them (e.g., auto focus, auto flash, auto everything) and yet if you don't know at least some basic rules on how to take pictures it's really hard to get a good photograph out it.

Personally I know very well the feeling when I see people misusing cameras just because they are not trained to use them, even with the simpler rules, and even more when they don't even want to learn! I give you a simple example: it is not too hard to understand that if you shoot a person's face against a bright background (e.g., the sky) you have to use some flash to fill in the shades. And yet most of the people won't do it. The same is true in simple visualizations like when people use useless 3D graphs or pie charts. What should we do then? Behave snobbishly and don't care about ignorance? Personally I think we have to recognize that there is a whole spectrum of people with different attitudes and needs. And we have to be ready to support the whole spectrum if we want this field to become a real success.

Like in photography we will have the Henri Cartier-Bresson and Ansel Adams, a very long list of professional photographers of any species, a even longer list of people who are totally passionate about photography but who are not professional photographers and then a huge list of people who want to just take pictures of their babies, travels, events, etc. Are we ready to support them? And even more important, are we ready to teach to them?

Ironically speaking both photography and visualization need visual literacy. In my opinion photography is not thought enough in fact .... let alone visualization! But if we expert are not the first to care about teaching who would it be?

2 - The story about real infovis users

This one i liked even more than the analogy with cameras. Some time ago I posted here some reflections on the question "Who is the user in InfoVis?".

Stephen in his talk went a bit along the same line in my opinion, trying to classify and quantify the kind of people who need visualization in their work. He classifies infovis users according to frequency and complexity of their information needs, and presents the following three categories:

1. Information consumers and presenters (around 80%)
2. Informal data analysts (around 19% and growing)
3. Sophisticated data analysts (less than 1% and in need of growth)

Apart from the accuracy or sophistication of this classification (every attempt would have some wholes anyway) I was totally amazed to see for the first time the face of some real people who use visualization in their everyday working activities! Stephen talked about Christian, Kathy, Marcel, Anne, people with very different attitudes, needs, contexts and yet all eager to have powerful tools on their hand to analyze their data and communicate their results to others in a neat and simple format.

Ours is a great community which is not unaware of the fact that we have to provide powerful tools to the hands of people. Each year at the InfoVis conference, and now at VAST too, there are some design or system studies describing tools designed to meet the needs of some professionals. The problem is in the disproportion between the research of new fancy techniques, often far from the real needs of people, and these kind of studies. And also in the focus on only a very small segment of potential infovis users, usually the ones in the third class suggested by Few: "Sophisticated data analysts".

I am optimistic however, this community is growing and it's really experiencing a turning point. InfoVis gains everyday more and more exposure and it's up to us to convey the right messages. The best message in my opinion remains always the same: show the quality and usefulness of what we do with real facts. Simple and effective tools that meet specific goals with elegance.

I'm strongly convinced that we have a lot to offer and that this lot will become more evident as time passes. That humanity is flooded by data is a matter of fact, and being able to deal with it with effective tools will be everyday more requested. And not only in our professional life but also in our private life. Visualization can help us in getting more organized and informed through the analysis of our own data. A large segment of the information that is produced everyday is personal data, crumbs of information about us that we leave around and that we can use to better manage and understand ourselves. A notable example of this view is the MIT Reality Mining Project.

These are the main thoughts that Stephen Few's presentation generated in me. I think I am ready now to read what the others had to say.