Friday, November 12, 2010

Blog Break and Infographics in the Media

After slogging my way through the summer while all my colleagues took leave I'm about to take 3 weeks off myself to sort some things at home and holiday abroad. I haven't been posting much because of trying to kick a few projects into shape before leaving them for a while. I'll be back 2nd week of December.

I'll leave you with a fascinating set of videos I've found by Geoff McGee for a fellowship he completed at Stanford University. Its about data visualisation as a story telling medium and focusses on graphics in the media. What's interesting about it is the same problems I've noticed in using neo-geo tools in education and outreach come up in their topic area. Points that particularly resonate with me:
  • Martini Glass Presentation: The importance of an introduction, context setting and explanation of what you can 'do' with an interactive web graphic or complex print graphic (section III: Telling Data Stories). Without this, your creation is just a set of pretty colours to the user. The Martini glass stem represents the video clip slide presentation introduction and the triangular glass represents the freedom of the user to explore the graphic on their own.
  • Attractive Does Not = Effective: A beautiful looking stream graphic showing box office results for movies with time is discussed in part IV, various commentators point out that it grabs attention wonderfully but then is difficult to interpret what it shows.

Monday, November 8, 2010

Timelines and Tours outside Google Earth

I've come across a couple of examples of GEarth features implemented elsewhere which were worth a mention:

Timeline Example: I thought this timeline from a New York Times graphic is much better than the timeline in Google Earth:
  • Easier to grab and move the jaws, in GEarth the jaws are too small
  • The play button only allows the jaws to move together, in GEarth you can press play and the far side of the jaws will move which is too complex for users to understand and utilise IMHO
  • The time labels are simple and clear whereas in GEarth the labels are more fussy
  • The blue shading communicates 'this is the time range' in a clear way and its semi transparent so you can see the graph below it.
The GEarth timeline remains high on my list of things Google should really fix in GEarth.

Tour Example: I think the tour feature of GEarth is one of its strongest features allowing user in presentations or promotional film clips. I came across a film sequence in a TED talk which has a form remarkably like a tour:


(BTW the clip is fascinating and well worth watching in full)

The clip 'zooms' down from large to small scale and at the destination scale the camera moves around a 3D object which is then manipulated in various ways to illustrate the relationships of neurons. Compare it with this GEarth tour:



The building clip has a poor frame rate and the building isn't manipulated in some way (like showing the inside rooms) but otherwise, the format is exactly the same. I'm in the middle of researching to best design tours at the moment and the Seung clip is a lovely illustration of how the results of my studies will not just apply to GEarth and other Virtual Globes but to any 3D visualisation system where zooming across scales in a film clip is important.

With the Seung clip I defined where it should start, see how to get a YouTube video to start where you want it.

Wednesday, October 27, 2010

Visualizing Motion

Static and Dynamic Motion Viz: Timelines offer an excellent way to visualise motion of people or vehicles. I've discussed timelines in GEarth with a video illustration previously. Recently I've been hanging out at the GIS Stack forum which IMHO is an excellent resource. A recent question was on visualizing motion statically: what were the best techniques? What software could you use? I've pitched in with an answer along with some other contributors.

Static vs Dynamic: What doesn't come out in the discussion is would you choose a static option over the dynamic one if you had a choice? I think for the expert map reader (I'm thinking of a biologist tracking animal movement as an example) a static map has a number of advantages, it can be scanned quickly to find the point/time you are interested in and values of velocity are easy to read off. By comparison, an animation must be played to get to the point you want to view. As it can be printed it's also very portable.

For the bloke on the street, I think the dynamic version has most advantages: understanding the static version requires interpretation, the animation is instantly recognisable. Also, animations instantly attract the attention which is important for the public but not necessary for the expert.

Wednesday, October 20, 2010

Recording Tours via Screen Recorder

I've lots of deadlines coming in at the moment so excuse the lack of posts last week.

Prior to the appearance of the tour feature of Google Earth in v5 I discussed recording a tour by using screen recorder Fraps. Recently Google Earth Blog has discussed a smart variation on this idea: set the fly to rate very slow and then deliberately speed up the movie in a movie editing package (see Be Travellers post). This gets over a major problem of recording a movie with screen capture straight from GEarth: the resulting movie loses the smoothness and looks 'choppy' (the technical term is to say the frame rate is low). Smart idea!

Here's the video example:


The HowTo write up Mike Griffin does is good but I have some suggestions (numbers in brackets refer to their step numbers):

Dialog box in the way (1): The way to deal with this is to simply drag the dialog box so most of it is off screen. Its not particularly elegant but it solves the issue.

Setting Viewpoints with Placemarks (2): as I suggest in this tutorial, if you choose your viewpoint before you create a viewpoint placemark its better because GEarth automatically sets the current view as the Placemark's view.

Manual Flights Clicking (4): The main problem with the technique Mike suggests to my mind is that you have to sit there manually clicking placemarks and then watching as GEarth flies from placemark to placemark incredibly slowly. This would drive me nuts for anything but the shortest tour.

If you can handle XML an alternative is to dive into the KML you've produced in setting up the placemark viewpoints to grab the Camera or LookAt data (in italics below) for each placemark.

<Placemark id="ElkRidge">
<name>Elk Ridge</name>
<description>shortly after the eruption</description>
<Camera>
<longitude>-122.2989801778228</longitude>
<latitude>46.29626193853221</latitude>
<altitude>13.39603579634948</altitude>
<heading>-31.06642935444568</heading>
<tilt>90.62346382665406</tilt>
<roll>-6.361109362927034e-014</roll>
<altitudeMode>relativeToGround</altitudeMode>
<gx:altitudeMode>relativeToSeaFloor</gx:altitudeMode>
</Camera>

<styleUrl>#msn_grn-blank1</styleUrl>
<Point>
<coordinates>-122.2993707244499,46.2969626924056,0</coordinates>
</Point>
</Placemark>

You can insert this data from each placemark into a FlyTo construction in a tour:

<gx:FlyTo>
<gx:duration>3</gx:duration>
<gx:flyToMode>smooth</gx:flyToMode>
<Camera>
...same data as above in here...
</Camera>

</gx:FlyTo>

using a series of these will give you a tour and you can edit the duration tags to alter the speed of flight from placemark to placemark. When you're happy with how the tour looks you go back through the tour KML resetting each duration to be x10 what it was previously. Now you can set the screen recording going as Mike suggests, run your incredibly slow tour and let the computer do the work recording it while you go and have some lunch.

Other Advantages: Recording tours as videos rather than as tours in GEarth has other advantages: you can access all the features of the movie software you use (such as fade in transitions) which aren't available in Google Earth. As time goes on I expect Google will add more features to tours.

Thursday, October 7, 2010

Maps and Art

There is an intriguing outbreak of using maps in art in the blogosphere;


The above image is of the English channel and is from the intriguing Fata Morgana which, using the latest feature in the Google Maps API has taken away everything but the words (and in the UK, what words! 'Hougham Without' in the middle of the image sounds lovely).

Axis maps release a set of posters working on a similar idea but putting a ton more cartographic love into their work.

Via Ogle Earth I found a wonderful reflective essay on the juxtaposition effects of imagery from different times: HiLoBrow.

And finally, Evert Schut finds inspiration in the many and varied patterns Google Earth produces at Google Earth Art.

Wednesday, September 29, 2010

W3G Conference - like Geography only cooler

(Title of this post ripped and mixed from @DJSoup tweet)

A quick post on yesterday's W3G conference, I had a great time, there was a fantastic mix of NeoGeos, GISers and all kinds of other geo-philes kicking around ideas in a conference/
unconference format.

Highlights for me included:

Steve Feldman's fantastic rant (blogged here and here) about data that shouldn't be mapped. His line was that mapping some data doesn't do it any favours because a] it doesn't tell you anything worth knowing b] its misleading. He identified some twitter maps as culprit, like this map of tweets about the Gulf Oil Spill, I agree with him that this one doesn't pass the 'does mapping this add anything to understanding?' test.

Peter Batty's comments in his unconference slot about usability. As I said in my slot, I have been advocating Krug's 'Don't make me think' book and usability testing to all sorts of groups for years so it was excellent to see someone else making much the same points.

I also met Stuart and colleague from Geo.me who have a very impressive engine for fast production of bespoke maps. Well worth a look.

I gave a talk on techniques of clustering points mostly and expansion on this post. I think I misjudged the crowd with my approach but geo.me and Nick Bicanic made some interesting counter points to my argument so it was worth doing.

This was my first unconference and NeoGeo event, its noticeably a lot less stuffy than more mainstream conferences I go to which is great. Thanks to everyone who put it together.

Thursday, September 23, 2010

Cartography of Base Maps

updated 1 Oct to attribute the quote correctly

There's been an interesting cartographic discussion of base maps recently in the geo-blogosphere; Stamen Design blogged some notes about their work on Bing maps, which was picked up by 41Latitude which was in turn highlighted by James Fee. The point that comes out from this, as Justin O'Beirne says, is that

"in trying to make a base map that’s optimized for everything, we’re actually creating one that’s optimized for nothing."

Which I heartily agree with. An example is street names, if you are producing a map for city navigation street names are vital but if your map is a thematic one showing a heat map of crimes per square km, street names aren't important and should be removed. 41Latitude suggests having a range of base maps for different uses, this is sensible but of course you can go one step further: The new Google maps API allows you to customise the base map and of course, if you start with a really basic base map (say land and sea) of any service you can always produce a custom base map by adding specific layers (e.g. roads) as you see fit.

How Many Users: What no one so far has mentioned is that how much you customise a base map is really controlled by how much effort you want to put in. In turn, that is usually a factor of how many users you have and how complex your map is.

sketch map courtesy Rosanne Elkins-Bushnell

At one end of the scale, the proverbial 'back of an envelope' map has as its base map the envelope and works fine for a sketch of how to find a coffee shop for one person on a street. Showing a group of people the location of the hotel you're all meeting at works pretty well as a marker on a street map; my maps by google would be fine for this as the data can be displayed as a simple map. However, once you start having multiple layers with markers or polygons you really want to start thinking about improving your base map as screen clutter and usability becomes a real issue, especially if your map service is going to be used by many users and a client is paying you to put it together. You want to choose from a series of base maps to get the optimal base map or even customise it for your particular use.