Next generation visualisations
With Stamen Design. Approach we have doesn’t come from anything other than a desire to engage with the technology and make a new experience for people.
Map of 2004 election, red states and blue states (MIT work). When you take these things by county instead of state, get a more variegated map. Break it down by degree, get a much more varied map, then add size of population you get a totally different map. More real, more expressive, gives you more to think about.
Working a lot on live data from places like Digg, or vast, or deep. Speed isn’t as important as consistently. More questions than answers.
Cabspotting.org – GPS signals from cabs to show you where people are going, and where the live cabs are, and which are full and which are empty. gives you an idea of where people are moving. Can give a speed map, red fast, white slow. Can see where the taxi depot is, the freeway. Then can animate it. Looks like the heart, it’s a circulatory system, or like a river.
When system breaks you know it immediately with live data. Breakages are not necessarily bad, e.g. on the Bay bridge, you can see directly the paths of the cabs on the upper deck, but the lower deck is a straight line because you can’t get the intermediary points from GPS.
Oakland Crime. Shows all the crimes in Oakland at once, so you can explore it rather than search is, as search implies you know what you are looking for. Can sort the crimes by type, e.g. quality of life crimes vs. violent crimes. Latter show more arrests than former. Can start to ask questions about why these trends occur.
Again, live data can show you breakages very quickly, so site has a blank in the timestream where system was shut down by crime reports end. Can access data any way – one visualisation doesn’t solve everyone’s needs all the time. Visualisations will prompt new questions that require new visualisations.
Modest Maps frameworks allows you to take tiles from any service. Starting to give you more control over how your map mash-ups look.
Digg story visualisation, each dot is a story Digg, so you can see which stories are getting more digs, and clusters of sstories, can see th links between stories and the popularities. See some interesting things when there is breaking news. when really big things happen, the visualisation nearly breaks, but that’s not such a bad thing. But the visualisation isn’t mission critical, it just allows people to keep an eye on what’s going on, more ambient than the crime one.
Similar visualisation where there are stacks with blocks falling from the sky. Can also give a tag cloud-like visualisation. Another that shows users Digging different stories.
Interesting things is that Intel sponsored a Digg visualisation.
Very interesting time, started off in universities, now progressing, starting to be useful, companies are taking notice. Important to provide multiple views into the same information.
Twitter blocks, takes your Tweets and strings them out in a 3D structure. Main stream is you and your tweets, and virtical access is time, so more time between tweets the more time between tweets. Crossing your line is the Tweet timeline for others.
It’s a gardening metaphor, you plant things and sometimes it grows and sometimes it doesn’t. It’s fluid, so you have to modify visualisations regulary to keep up with the data, amount of data, etc.
Trulia, real estate aggregator that connect realtors with buyers, provide a lot of visualisations and maps – heat maps, neighbourhood info, connections between realtors. Took the dates that houses were build and did a visualisation for each year’s builds. Took data to show when neighbourhoods were constructed. Madison you can see it flow down the peninsula, Vegas is like a big stain that feeds on itself. Shows you histogram of housing booms. See weird patterns, e.g. big spike in 1900 which is probably a record keeping spike. Visualising the data reveals unexpected things.
Ryan Alexander’s map of search results on Trulia, people who search for Mass City, MI, you can see where else they searched for afterwards, etc. Get a sense for people searching local, or outside of the state. Shows that people intrested in New Mexico are also interested in other places.
Start to see patterns, see surprising things. Brattleboro, VT, small town, people look for a few other places, but they are very, very interested in places in the rest of the US. college town, and all the other places you might want to be if you are a professor, aren’t local. Cheyenne, WY, is the same, which is a vacation town so all the other towns people are interested in are vacation towns. Anchorage, AK, people want to get out of. Was trying to get out of NY, Queens, and the people who live in Queens who are interested in other places in Queens, *or* places in Florida.