Top banner ad available.

jeffrey_veen_talk.jpg
A 20-minute talk by Jeff Veen from Small Batch, Inc., also known from WikiRank, which was originally given at the Web2.0 Expo in San Francisco a couple of weeks ago. During the talk, he focuses on some of the classic examples of information visualization (John Snow pump, Minard's map, the tube map, and so on), the issue of "decorating" data versus making it accessible, and the emerging challenge to empower lay people to participate in visualizing and analyzing their own data.

Considering the examples and his focus in the democratization of data visualization, I sincerely hope he reads infosthetics.

Watch the video below.

Thnkx Christina. Image taken from Anita Hart at Flickr.

MORE

15 COMMENTS

Very nice. Not only the lecture itself, but I haven’t seen that layout of speaker/slide show done anywhere before. It’s really nice. Perhaps it wasn’t worth it before online video started offering HD, but I hope more sites adopt some variant on this...

Wed 22 Apr 2009 at 4:41 PM

interesting talk. i wonder about the "audience control" aspect -- i think he is (we are?) assuming that the user is somewhat sophisticated and will be able to pull a story out of the data given the right tools.

how likely is that? also, if you have only sophisticated users, might they not want to take the data and do their own thing with it, rather than use your nifty tools?

Thu 23 Apr 2009 at 12:45 AM
d^2

"Image taken from Anita Hart at Flickr."

With permission? The photo is traditional C, not CC licensed.

Thu 23 Apr 2009 at 2:25 AM
Jaakko H.

I think really the whole point to make as a takeaway from this how these data dimension controls will really improve the users lives. This is obviously where data visuals will end up, but what innate behaviors do people have to gain quality of life from reading the data. If its positive enough, maybe their be a bit more patient to take a 3 min instead of 1 interacting with these interfaces. I wonder if data interfaces will go through the same mass production limiting quality that products started going through in the industrial age...at some point.

Thu 23 Apr 2009 at 2:39 AM

I've been a fan and feed subscriber for quite a while, Andrew. Thanks for the post!

Good questions, d^2. I don't think we need to make any judgements on the sophistication of users and how they'll approach what has traditionally been obscure and difficult. That's why I've been so focused on personal data. I think the tools designers traditionally have used to illustrate Napoleon's bad decisions are even more powerful when used to answer questions like, "Did I run faster this week?" and "When will I ovulate next?"

Thu 23 Apr 2009 at 5:48 AM

I love that you are such a data geek that you have a photo of the John Snow water pump! :)

Overall, I agree with your statements. However, I often feel that analysis is still a very important part of infoviz. Your Botswana/GapMinder example makes my point. Rather than turn all users into analysts, I think there are some stories that should be brought to light. Unless one already knows of the Botswana story or is going to dig around all of those countries for hours, they probably won't stumble across that heartbreaking story.

In fact, I think we often need a combination of the discovery - storytelling dichotomy you brought up. Let the users discover their own story, but also showing the important stories that may not be self-evident.

Obviously this point is moot for personal data or realtime information.

Fri 24 Apr 2009 at 8:19 AM

Great presentation. Even though he wasn't mentioned by name, you're obviously a big fan of Edward Tufte (love his work on spark lines). Jeff, I appreciated your emphasis on storytelling. I think what makes data (and its interpretation) so compelling is that there is a story involving real people behind every number.

Sat 25 Apr 2009 at 1:06 AM

Very user-friendly presentation!

I did really appreciated the story-telling approach used to speak about data visualization.

Jeff, thanks for spreading the voice,

d.

Sat 25 Apr 2009 at 5:30 AM

nice speech indeed, widely targeted: barnum effect

Sun 03 May 2009 at 9:48 AM

at 11:50 there's an imprecision: removing "unuseful information" is not always the solution: example both official london undeground map and the geographically accurate version (http://kottke.org/03/01/geographically-accurate-tube-map) miss the point that timing + geographic position is the real people's need.

Sun 03 May 2009 at 10:05 AM

Great talk! Love the water pump photo. :)

The only quibble: you suggest there's a tension between story telling and discovery. It has always been the case that info viz has had two jobs: discovery and story-telling, and they're used in exactly that order.

Your use of Gapminder for the Botswana story is a perfect example of what I'm talking about: the sharp dip in Botswana was only evident after you knew to look for it and thus, knew that you, the speaker, had to dim back the rest of the data and to turn on trails to let us see what you were talking about. You came to the data knowing what question you were trying to ask ("are there any outliers here?" "are there any countries experiencing sharp rises/drops in life expectancy?" etc.) Don't underestimate how hard that step is: asking the right question of the data is the key to insight and understanding, not pretty pictures.

Once you know what question you're trying to ask, then you can design a chart/graph or tool that helps highlight the right subset of the data in a way that lets that answer come to the fore. Only then can you, the investigator/data-designer use that answer (which might be a picture, an animation, or just a table of numbers) to paint a very clear picture for others to convince people of the story you want to tell about the thing you discovered in the data.

These aren't in tension, and they're not in any sort of transition from one to the other.

They'll always both be there, and the critical step is knowing what question to ask.

Tue 13 Oct 2009 at 2:35 AM
Matt Conway

at 11:50 there's an imprecision: removing "unuseful information" is not always the solution: example both official london undeground map and the geographically accurate version (http://kottke.org/03/01/geographically-accurate-tube-map) miss the point that timing + geographic position is the real people's need.

Wed 17 Mar 2010 at 8:14 PM

Great presentation. Even though he wasn't mentioned by name, you're obviously a big fan of Edward Tufte (love his work on spark lines). Jeff, I appreciated your emphasis on storytelling. I think what makes data (and its interpretation) so compelling is that there is a story involving real people behind every number.

i wonder about the "audience control" aspect -- i think he is (we are?) assuming that the user is somewhat sophisticated and will be able to pull a story out of the data given the right tools.

Fri 15 Jul 2011 at 12:51 AM

at 11:50 there's an imprecision: removing "unuseful information" is not always the solution: example both official london undeground map and the geographically accurate version (http://kottke.org/03/01/geographically-accurate-tube-map) miss the point that timing + geographic position is the real people's need.

Wed 28 Sep 2011 at 12:28 PM
ADD A COMMENT
Commenting has been temporarily disabled.