Google IO 2011

I've never been to Google IO, I'm told the tickets are v difficult to get but looking at the recordings of the sessions, it could be one to think about for the future.

Mano Marks and Sean Askay on High Perfomance KML is excellent so far, I've only got part way through but Sean has already shown some combinations of Tours and the timeslider, update to change the elevation of polygons and a GPS track and heart rate visualisation that I want to play with. Sean's also launched a new site unchartable.com with similar content that I'm going to be watching.

I'm also watching a video on fusion tables which has some good new features about the fusion table/google maps tie up.

Credit to Google for getting this all recorded and released so quickly.

Does 3D suck?

I was on a road trip holiday the week before last and before I left I read this James Fee blog post he has an image of Godzilla walking into electricity pylons with the tag line 'Google Earth keeps bumping into things trying to work out why it is here'. This led me to ruminate and I quickly wandered from James' post about GEarth enterprise to the more general 'what is Google Earth good for?'. I considered features such as

• layer control
• zooming
• time animation

but, powerful those these are for visualisation, they are all also available in Google Maps and are just as much 2 as 3D*. What truly sets a virtual globe apart from web mapping is 3D and this is what I ended up ruminating on most of all:

Current Uses of 3D. I don't have much time for fancy graphics, so beyond cool looking 3D flights, what does 3D in Google Earth actually add to understanding at the moment? There are several examples:

• 3D buildings: Populating a city with 3D models adds a lot of value to urban geography, you can see what a skyline looks like or enhance what you've seen on a walk through a foreign city by revisiting your walk in GEarth later. 3D trees are available in GEarth too but I don't think they're nearly as useful as the buildings.
• Mountains and Valleys: When considering topics where topography is important a 3D view can add understanding. The Appalachian Mountain removal project is a perfect example as the location of the mines on the tops of the mountains is key to understanding the problem.
• Earth Science Models: 3D models of large scale processes on our planet such as subducting plate margins and hurricanes can add understanding, see the project I'm working on with Old Dominion University and others.
• Streetview and Panoramas: Streetview and gigapan panoramas come close to giving a realistic 3D view of a scene. Streetview enables you to move from location to location whilst a gigapan allows you to zoom in on interesting features of the panorama.

IMHO the Streetview/Panorama example is the most important of these. However, the 3D visualisations discussed above are far less common on the web than the uses of 2D like data mashups, real time maps and map related infographics.

Future Uses of 3D: But what is available now may not be the whole story. Can we imagine a use for 3D in the future that is the 'killer app?'. Here's some possible future uses:

• Real 3D Earth Science Models: Currently the way to present a 3D geology model is to slide it out of the ground to give the viewer the idea that its come from underground. It works pretty well but it would be good to be able to fly beneath the earth's surface and show models in their real situation.
• Thematic 3D data: I've frequently discussed 3D thematic maps on this blog, mostly unfavourably. However, the 'Obama: One People' visualisation I discuss here looks to me like it really adds something, I think 3D thematic maps could be useful visualisations but they remain unproven - no one has done user tests on them.

I can foresee that Streetview and similar panoramas will improve in time with real time 360 views and developments like Microsoft's streetslide helping make them more usable. Beyond that, neither of the topic areas listed above is convincing as a killer app.

2D Dominance: So I don't think 3D is ever going to topple 2D as the best way to visualise spatial data. This may seem a little surprising, we live, work and move about in 3D all the time, why won't 3D will become more common? What this point misses is that we also symbolise all the time too, a road map is far more useful than satellite imagery for navigating with because it strips out everything unnecessary (trees, people, greenhouses) and leaves exactly the information we need: roads and junctions as symbols. Usually the best visualisation is the simplest one possible so if a spatial problem can be visualised in 2D its best to do it that way. Especially if you are rendering the map on a small screen smart phone. We live on a planet that were it reduced to the size of a billiard ball would be smoother than any in the world so it really isn't surprising that most of our spatial data can be effectively reduced to 2D.

*of course you can now pull 3D GEarth into a Google Maps but I'm not really focussing on the technology here.

Gapminder and GA Conference 2011

So at the GA Conference last weekend the most interesting thing for me was Hans Rosling and Gapminder. Hans's talk was excellent, as he was addressing a set of Geography teachers he went into more depth about some map design characteristics of the Gapminder World. If you haven't come across it I recommend you give it 4.5 minutes of your time by watching this video

Thematic Maps: Gapminder started out visualising the data with a standard thematic map of the world which they colored to show changing variables like rise in GDP. Hans said they soon rejected that as countries with big areas such as Australia dominate the view visually. That's a clever insight IMHO.

Map with Symbols: So they tried circle symbols and varied their sizes. This was better but a change in size isn't as visually arresting as movement. Hans's smart phrase was : 'movement has a direct highway to the brain'. So they ended up with their famous graph visualisation and a map tab you can click to see a map presentation. I think the key point about the data presentation is that country location isn't that important, all you really need to know is which continent a country is in. This means the map can be ditched in favour of a graph plot which brings with it the advantages of movement and the ability to draw tracks.

I couldn't help thinking that Hans the statistician seems to know more about good graphical and map design than a lot of GeoWeb developers and GIS specialists I know.

Happy Easter Break!

GA Conference 2011 - Links and Slides

I had a great time at the GA conference last week, I gave two workshops on using Google Earth in education. Lots of people thought the sessions should be longer and asked for links to the materials. These are the slides from the Introductory session (you'll need a Google account if you haven't already got one). They include:

• Links to the first 3 lessons as presented in the session
• A Link to video versions of the other 5 lessons
• Links to the files I used to illustrate the Case Studies

These are the Advanced Session Slides. They include:

• More Case Studies of use

I'll write up some thoughts on the conference later.

7 Foot White Elephant: When Globes Fail

Negative Reviews: Observant readers of this blog may have noticed I've stopped doing reviews of Google Earth projects recently. They took a lot of time and I'd always get a defensive response from the authors which isn't any use in starting dialogue and discussion about map/Virtual Globe design. However, someone actively promoting bad techniques of geo-visualization in a keynote presentation deserves a rebuttle so here's some thoughts on Julia Grace's keynote from Web 2.0 Expo NY 2010:

At 2.29 she shows a graphic with static, old style pie, line and bar charts:

"whenever I see these sorts of charts, something inside of me dies. The bottom line is that we can do better than this"

Simple is Best: Well yes, sometimes a highly interactive graphic is a great answer to a communication problem but a lot of the time simple non-interactive graphics are better. Perfect example, David MaCandlass's graphic of scare stories with time. Its well produced but at its heart it's a simple line graph:

At 5.34 into her keynote I end up agreeing with Julia for a while. She name checks the interactive NYT Oil spill map as being an excellent visualisation. I agree with her on that, its a great use of an interactive map tools to help the user understand the spill.

Map on a Globe: At 6.06 she then discusses a 2D map of IBM offices she had to use, she thought it sucked because of map distortion, her solution? Go out and buy a 7 foot globe and project the map onto that (an image comes up at 6.39). Apparently the globe was so large it wouldn't go through the door of their offices.

The Problem with the Globe: The problem she thinks she's solved is an ancient one: you can't represent the earth's surface on a flat surface without distorting the map in some way. However, although the globe solution she's promoting does solve the distortion it has an even more important flaw: now you can't see all the offices at the same time because they're on different sides of the globe. Even if the globe was the best solution, what's so wrong with a virtual globe on an iPad? A whole lot cheaper and much, much easier to get through doors.

When High Tech is not Cool: But what I really object to in this talk is the tone. Simple visualisations are likened to 'Tron', old, flat and outmoded whereas 'Minority Report' visualisations are cool, 3D and the shape of things to come. Of course this isn't true, the best data visualisation solution is defined primarily by the audience you're addressing and sometimes simple is best. You'll notice she never mentions users and usability.

Investigation into Placemark Clustering

First of all I'll start by making a point by considering a couple of examples of the development of everyday technologies:

College Dropout changes Computer text: The text you're reading on screen would be very different were it not for the actions of a student dropout. Steve Jobs dropped out of Reed college but, because it interested him, still attended a class in calligraphy. The techniques he learnt, about elegant letter spacing and formats, were later applied to the Mac and from there were copied by windows. We've all benefited from Steve's nerdy love of text.

Spam Email didn't have to exist: Email was invented early in the history of the internet but because everyone then knew each other by name, no one bothered to produce code in email programs that checked the ID of senders. We all suffer because this didn't happen, the internet is swamped by spam email traffic that could have been avoided. Google Wave was an attempt to get us all off email to avoid this sort of problem, it didn't take off despite having the promotional weight of Google behind it.

Get Technology right early on: My point is that its incredibly important to try and get early technology right otherwise you may never be able to correct it. One of the issues I see like this in web mapping is the clustering of points, if we get it wrong now we may never be able to undo it and we will end up using sub optimal visualisation techniques just because we've always done it.

Clustering Placemarks: Previously I've written about the problem, placemarks need clustering because at a certain density of points it is becomes impossible to pick individual points out. IMHO some of the ways of visualising these are poorly designed e.g. clustering placemarks into 'blobs' with numbers.

It may be that numbered blobs work as a way of clustering placemarks - maybe users immediatly get the concept of a large blob being a cluster and that outweighs other problems I've identified. But what worries me is that this technique is all over the place in web maps and no one has actually done any user testing to show that its effective.

Enter my new MSc student Craig who's doing his project on this map visualisation. By doing a series of user tests we hope to answer the questions:

* Does blob clustering work compared with other techniques?
* If not, can we adapt it so it does?

Wider Point: This isn't the only example of a technique that is being widely used in web maps without being user tested (anyone for 3D thematic maps?) so I think Craig's work will also be important in promoting the more general idea that we need usability testing in web maps. At the moment I think web programmers are applying these untested techniques because they think the look flash and/or they are easy to apply because the software needed is readily available.

Street View in Tours: Ancient Rome

AFAIK I was the first blogger to spot that the new GEarth tours in GE6 supported streetview. I still think this a great bit of functionality and I'm surprised I haven't seen it popping up more. Today I came across yesterday's blog post by Google about how their streetview trike has been capturing imagery of historic European sites (what a cool job that trike operator must have), really clever use of the technology but I was interested in the way they presented the streeview examples: They linked out to GMaps instances of streetview from within the blog, three examples are shown here:

View Larger Map

Colosseum

View Larger Map

Imperial Forum

View Larger Map

Palantine Hill

It works and is impressive but this form of presentation is clunky if you wanted to see the relationship between sites in which case a GEarth tour becomes the media to choose. As an experiment I've produced these three sites in a GEarth tour, it's a bit rough but I think you can see my point about how this is a better way if the geographic relationship of the three sites is important.

to play in GEarth itself Rome Street View tour

Note I used the wrong forum site.

Good Design Points:

• You'll note that by use of annotations for Rome, and a 'reference' square I give a sense of location and scale to the viewer.
• Use of the square allows people to memorise the locations of the three sites as it acts as a strong landmark.
• I return to high views between sites as this allows people to follow where they're being flown to.
• I also customised the Colosseum views and tour path (by choosing a different high view) as in its original format the tour flew through the Colosseum walls - distracting for the viewer.
• I make explicit remarks about how the user can pause the movie to look around and also about how fast the imagery is likely to be downloading. The former is necessary to encourage active use of a tour, most people will hit play and sit back if you don't actively suggest they could do something different. The comment about imagery download will help those on slow connections who may be thinking 'its not working' as all they can see is a pixelated sludge on screen.

Technical Points:

In KML you can control turning on and off streetview, the sunlight slider and historical imagery. It would be useful to be able to do the same for 3D buildings and roads. However, by using the Tour gadget to produce GEarth in a browser (as I did) you can control these layers without needing to code anything.

I also noted that streetview imagery is low resolution until you click pause at which point it sharpens up.