Thursday, 15 August 2013

Dog watching or dog's dinner?

Watch_Dogs WeareData alleges to be the first web site to gather publicly available data about Paris, London and Berlin, in one location.  They map the data and create a visualisation of the soup we live in. Now I'm not going to be picky...oh, ok, that's the point of the blog!

So first off...it's not the first. CityDashboard by my friends at CASA has been around a good while and does a great job. Clean, simple, not over-selling what it does, informative. There are others.

So is the new one any good?



Well it looks 'nice'. it has a really strong aesthetic and is definitely very strong visually. It uses colour very well, particularly in receding contextual information while highlighting the detail using highly saturate colours with excellent use of shadows, glows and transparency.  It's one of those web maps you look at and genuinely thing...wow, that's actually 'different'. Beyond that is where the problems arise so I'll just offer some thoughts in no particular order...



Buildings are rendered very well with just a hint of outline and shadow to create the illusion of depth on a sphere from the central focal point. As you zoom in we increasingly move away from planimetric to a 3D panoramic view. The landscape gets a carbon fibre style treatment and we have building walls in transparent gradient fills. Very few names though and at large scale we need labels to determine where we are. Buildings aren't clickable. More labels would help.




Is there too much going on? There's movement, interaction and a lot of data being displayed.  Note I said data...but with a quick click we can reveal information for some data types for instance down to the one-thousandth of a second until a particular tube train arrives at the next station.  Really?  Does publicly available data really give us accuracy down to that degree of precision? Once you click on a moving train symbol though, the train becomes the centre of the map which pans to its destination...nice touch.

As you click other objects (e.g. CCTV camera symbols) and a Bladerunner style set of lines appears, shudders and then persists, glowing.  What? Are the cameras connected? Why can't I see a web stream? What is this connectivity showing?

Mobile cellphone masts are similarly clickable and seemingly connected.  But in reality they're not are they? Neither are ATMs or any other class of object other than purely as a map category.

Boris Bikes are symbolised with a green circle and give details on the number of bikes available but why not use the proper symbols? Yes, I know...copyright, but geometric shapes lose much of the power of mapping when common objects are mapped. Same for the London Underground stations. Some symbols are so engrained; so ubiquitous that seeing new representations is awkward and unfamiliar.

Traffic lights are included but...why? There's no roads mapped (though some road labels are visible) and no data associated with the lights. There's no information on traffic problems, accidents, heavy traffic or anything like that. If we could control them then maybe it would be fun but this is an example of mapping data because they managed to acquire/scrape it rather than having a purpose to mapping it.

What are the green polygons? Parkland? Again...not clickable and only a knowledge of the real world enables you to figure out what you're seeing.

Then we have the standard social media feeds...tweets, instagram, foursquare, Flickr. Really. Isn't this all getting a bit tiresome?  I clicked one instagram post and saw a picture of an empty bowl of sweets.  In the context of the city...what precisely is this telling me? Maybe that Rasheeb has run out of Quality Street? That's all. Nothing more. Pointless noise.

With all the nice geographically accurate buildings why do we have a stylised Underground network? Straight lines connecting stations? I'll tell you why...because that's how the data exists from the source used. But it jars, visually.

We get area stats that include a latitude, longitude to 6 decimal places...for an area? Where in the area? Many of the data fields are blank for some areas. Others appear static such as electricity consumption....oh, they are static...it's the average consumption for an entire year. So we have tube trains position reported to thousandths of a second and electricity consumption for a huge area of a city averaged over a year.  How is any of this comparable temporally when presented on the same map?

White text clashes with white symbols etc etc. And what are the annoying white micro-dots that seemingly hover over the city? Non are clickable. Non have any explanation? Argh.......

Heck it even has audio. Stick your headphones on and you're immediately in the trash compactor scene in Star Wars Episode IV...no, wait...there's some trancy music. It IS Bladerunner!

I could go on but you get the gist.  Apparently we can "discover more about the data that controls modern cities". Oh really? I hate to break this to you but you actually can't. Visual eye candy but beyond that, this is a random collection of data sets thrown on a map loosely and with no sense of purpose.

2 comments: