Friday, May 22, 2009

Where 2.0 topics

So I've been following where 2.0 via twitter and the videos published and feeling like I missed the party :(

Google Earth for NeoGeographers: Frank gave a good summary of what is possible with Google Earth from reading his PPT presentation.

Community GIS for New Orleans: Denice Ross and James Fee gave a good talk on using GIS and web GIS to provide New Orleans with spatial information. What I particularly liked about this was that they mentioned that they did user testing (details). I tend to advocate a less sophisticated testing regime than they describe, I think you can identify most of the problems with your web site without needing to work on users actual machines but it was a good write up.

It was also interesting to hear that James Fee did a good job of assessing technology needs rather than ploughing in and telling the clients what technology they should use.

4D Maps: Brandon Martin-Anderson talked about representing 4 dimensions on a map. There was some interesting stuff in there but I felt he missed out discussing the many examples of time in GEarth and also Hans Rosling's wonderful visualisation tools.


Denice said...

Rich, thanks for the nice words about our New Orleans presentation! The findings from our field usability testing were pretty humbling and surprising. In our case, going out into the field to test with users on their own computers was essential and eye-opening.

One example I had to cut from our talk: Our web stats showed that less than 5% of our visitors were using 800 x 600 resolution, so we felt confident designing for 1024x768. To my horror, 3 of our 5 usability testers -- neighborhood leaders in the Lower Ninth Ward and New Orleans East and Upper Ninth Ward -- were using machines set to 800 x 600. Not only that, but just about every toolbar and taskbar was set to be visible, which left (I'm not exaggerating) like an inch and a half of vertical screen space to actually view the data in our Google map.

Neighborhood leaders are a core audience of ours, so we had to make sure the map worked on the computers they use. My first inclination was to disable the overview map in the lower-right corner, since that was taking up valuable map space. But, it turned out that these users with the low-res displays heavily relied on the overview map to orient themselves and navigate as they looked at the block-by-block data in the main, zoomed-in part of the map.

In this example, we didn't change the design based on our findings, but are now hyper-aware to not add any design elements that would reduce the map's vertical real estate -- and that overview map will definitely stay as critical piece of our interface.

We would have missed this and other learnings if we'd tested in-house only.

On a side note, field usability testing is a fun way to meet users and get a sense for the real world conditions that our site is used in. And those testers have ended up being some of our biggest evangelists -- a nice side benefit.

Rich Treves said...

Point well made!

My comment mostly came from the point of view that almost every time I ask someone producing a service 'how much user testing are you doing?' the answer is none. Its so easy doing 3 x 1 hour testing in the office just grabbing people that I advocate that to start with. Persuading someone doing no user testing that they need to go out in the field is much more of a difficult sell. I totally take your point that field user testing adds useful information and that getting to know your users is helpful in all sorts of other ways you can't easily predict at the start.

Thanks for sharing the story, I've noted it as one to quote at various people in the future :)