Tuesday, 20 August 2013

Pretty Maps

There's no doubt that maps that look good are better than maps that look bad and whilst that is, in part, down to subjectivity as technology progresses so we're seeing more better looking maps.

But better looking doesn't mean better mapping. Take for example these maps of people moving around Seattle (actually, a couple of maps). Click on them and go take a look:






They are beautiful (in my opinion).  They have a strong aesthetic that pulls you in to explore. They use colour sparingly and they use a palette that works well. The symbology contrasts well and visually I'd describe them as quite stunning...certainly creative and well composed. As we move through time the individual symbols add a trace to the map so we get a cumulative look at general movement corridors and the blue proportional symbols increase in size as people move close by.  They're grocery stores. The design isn't wholly original but originality shouldn't always bee seen as the cornerstone of something being 'good' anyway (itoWorld's maps of OpenStreetMap edits set the standard for this type of map style and animation some years ago of course). So they look good and follow on from maps designed in a similar fashion. Job done? Unfortunately, beyond the aesthetics are just dumb maps and there's nothing particularly clever about them....they even put the animation to a soundtrack as with many such maps.

The University of Washington's Center for Public Health Nutrition is responsible for them and The Atlantic gives them a short write-ip.  Researchers stuck GPS receivers on 493 people in Seattle to watch where they went across a week in order to see what their food buying patterns were. The maps are a first stab at looking at the data and guess what...people move through space. they move in different ways. Some move faster (cars), some slower (feet) and what do the squiggly lines and symbols tell us...err, nothing... people move from here to there as they go about their daily business.

The GPS wearers did keep a diary of their movements as well as shopping habits etc so it'll be interesting to see if the researchers can turn that data into something more meaningful and actually reveal something and map it.  That's the hard part...turning a research question into a set of digestible findings that can be mapped to reveal them visually.  We wait in anticipation but at the moment the maps are just eye candy. The GPS tracks are in no way broken down into trips to grocery shops...it's just raw data. The blue dots get bigger as people move close to a grocery store located there but that would always happen because grocery stores are located in areas that people move close by. People move. Nothing more or less.


In their second video they tracked speed of movement. The claim is this highlights the walkable parts of the city...but if you look closely it actually doesn't...it just reveals where people who walk...walked. Is downtown walkable...well it will contain more people at certain times of the day who most likely walk to get lunch.  that doesn't mean the urban morphology is any more or less walkable. Correlation does not imply causation.

And here's a second map that's doing the rounds as a collaboration between HERE (Navteq and Nokia Maps) and CartoDB. It's called Living Cities. Go take a look by clicking on the image.




Again...it's visually very appealing and using CartoDB means they've made the UI really slick and modern.  It's certainly pushing web-cartography more towards a higher-quality product and if tools to make web maps begin to provide the ability to design maps like this then that's a really positive move.  The basemap is really nicely integrated into the application for instance and uses some clever symbology. But we're still stuck for purpose....dig a little. What does the map actually show? Again, not very much.  We've got some animated pulses (nothing like visualising the pulse of a city for sure) and factoids popup. There's some clicky things and we have time represented as well.  None of this is particularly useful and hundreds of other city maps also take feeds of data and purport to offer insight into daily patterns.  They don't...they just show partial, generalised and random picto-bites and sound-bites.

This brings me to the general point...should maps always be 'useful'? Well yes...I believe they should but here's the sting: A map's 'use' can be defined purely as an object that is designed simply to look good and bring visual pleasure to a set of data.  There is nothing wrong with stating categorically that your map is designed to just look good.  Maps can just be artistic with pleasure being derived from their appearance or designed to inspire people to design their maps along similar lines.  But...the maps shown here and many, many others go way beyond this simple objective and claim all sorts of nonsense to try and justify their being; to try and purport to show something they have no hope of showing; to raise their standard beyond the purely aesthetic as if they become more valuable by so doing.  But in over-selling their purpose they actually do themselves a disservice. The Seattle maps claim to be about exploring movement, behaviour, grocery shopping and health. The Living Cities maps are designed to tell 'stories'...except they don't, really. They look good...that's about it. They are valuable on that basis alone so let's leave it at that.

Looking good for the sake of looking good is fine by me...and I like what these maps show.  I can look at them, smile, take some ideas and move on....trouble is at the moment it seems people are desperately trying to find a reason to make their own particular pretty map. It's almost as if they are reluctant to simply make a pretty map without having a higher purpose on which to hang their visual ideas. Pretty maps are fine. Pretty maps with a real purpose and which genuinely serve that purpose...now that's the ticket!

2 comments: