<-- Advertise here.

Greetings from the InfoVis conference in Columbus, Ohio! Today we heard a compelling keynote presentation by Jake Kolojejchick, Chief Scientist at Viz / General Dynamics. Jake is one of the creators of the "Command Post of the Future" system, currently in active use by the U.S. military.

Jake talked a bit about the system itself, which is geared for realtime collaboration in situations that are, to say the least, stressful and demanding. But his main messages were lessons learned during the system's design and deployment. Three points struck me as particularly interesting.

1. Jake noted that his users were extremely smart and resourceful, which ironically made testing tricky. Given simple tasks, users were too good at figuring out even hard-to-use software. To get around this, users were given extremely difficult test tasks, so that they had no extra brain energy to solve interface problems.

2. Military commanders have always loved maps. Jake speculated that this wasn't just because maps convey information clearly, but they also convey possibilities for action. A bridge almost begs to be crossed, for example. Jake then showed a slide of various visualizations of the current financial meltdown. All of these charts showed how bad things are, but none held any hint of what to do next. (Alas!) Is there a way we can create visualizations with map-like "affordances for action?"

3. The part where I really sat up and took notice was when Jake said that he viewed visualization "as a medium." This is something I've heard (and occasionally said) before, though usually in the context of artistic or social software--so at first it was a surprise to hear the same message from someone talking about a military system. The viewpoints people choose and the annotations they make on visualizations tell you a lot about what they're thinking... and apparently this is as true on the battlefield as it is on a Web 2.0 social network.

Guest blogger Martin Wattenberg is the group manager of the IBM Visual Communication Lab. He is known for several successful visualization projects such as Many Eyes, Baby Name Voyager and Shape of Song.

See also Infovis'08 Conference Coverage.


Very interesting post. More from that :)

Wed 22 Oct 2008 at 4:55 AM

Are you in Columbus? Would love to buy you a cup of coffee.

Wed 22 Oct 2008 at 12:02 PM

Of course Matt McKeon who worked on Many Eyes came from MAYA Viz and helped build the Command Post of the Future so there is an interesting 6 degrees linkage.

Thu 23 Oct 2008 at 1:34 PM

And if you go wayback one of the founders of MAYA (Jim Morris) worked at Xerox PARC on the alto and star programs that started much of this mouse, pointer, desktop world. You can follow a nice line from there to 1990 and this demo (in hypercard!)
http://www.youtube.com/watch?v=H5-T_S50Sr4 (link)
to this in 1993 (document centricity and interaction physics)
http://www.youtube.com/watch?v=H9F17JrG-SE (link)
to this in 1996 from Jake and his team at MAYAViz... (information-centric interface and synchronous and asynchronous collaboration)
http://www.youtube.com/watch?v=7SjEwrZ3cDI (link)
that later became Command Post of the Future...
Of course Jake can't really show the coolest stuff yet, you'll have to wait a few years for it to be shown in the wild.

Thu 23 Oct 2008 at 1:50 PM

It's a medium, because you can push back on it! The thing I wonder is whether it's the same medium as video games?

Point #2 is actually a little scary; are bridges crossed just because they provide an affordance? That sounds a bit like shoddy tactical thinking. The economic analogy is interesting, because the magnitude of the current mess is largely a result of its complexity (derivatives etc.). If there were a visually-apparent way out, we might not be here to start with. =)

Point #1 is interesting, I like the idea of "squeezing out" excess cognitive capacity to find out how the system performs in crunch time.

Thu 23 Oct 2008 at 2:08 PM
Commenting has been temporarily disabled.