Categories
Data Graphics

LinkedIn Network Maps

I’ve just come across the network map generator from LinkedIn Labs, the “cool fun stuff” page where LinkedIn employees put their “20%-time” projects. I don’t use LinkedIn hugely, but have built up enough contacts on the “professional” social network now, from accepting connection requests, for a reasonably interesting map to be produced. You can see mine here. Four clumps are immediately spatially apparent: I identify them as University, Graduate Job, Current Job (CASA) and Orienteering. The software itself identifies and colour-codes six categories – it separates out my graduate job clump into the interns and the people I met once I came back for real – and splits the current job clump into the current and previous role with a closely aligned group (the quantative geographers at UCL.)

The maps are reminiscent of what can be produced in GePhi, an open-source network visualiser that is becoming increasingly popular here in the CASA lab. I produced a similar kind of map a good 15 months ago of my Facebook connections – this latter map has a richer set of connections but people are connected by the simple application of Hooke’s Law (masses on interconnected springs) with straight lines, rather than the sweeping curves of the LinkedIn Lab map, and without the automatic categorisation. You also don’t get yourself placed at the centre, with all the lines leading to you :-). My connections did however also group roughly into the same categories, showing that once you’ve got your connections, it’s difficult to lose them, no matter what network you are on… 😉

Categories
Bike Share Data Graphics London OpenStreetMap

Flow Animation of Barclays Cycle Hire Bikes

Dr Martin Austwick and I, here at UCL CASA, have been working on an animation of the Barclays Cycle Hire bikes (aka Boris Bikes) in London, based on the historical flow information that was released by Transport for London (TfL) last month.

Taking one of the busiest days of the scheme – the 4th of October last year, a Monday which coincided with a London Underground strike – Martin has created an animation showing pulsing blobs, or motes, representing the bikes, moving through the 18 hours of the day that the data is available for. As each hire is made, the docking station dot flashes red, and and blue trail starts to leave it, heading towards the destination dock which flashes yellow as it receives a bike.

At the rush-hour peaks (08:45 and 17:45) the map becomes a sea of a 1000 blue pulses, many congregating on a number of key routes in London. The few bridges across River Thames can be picked out as intense bars of light, as commuters travel between Waterloo/South Bank and the City/West End. Hyde Park (middle left) and Regents Park (top left) are noticeable from having few docks in their area, and only a few bikes crossing them. The east seems busier than the west, as the City workers typically commute to work earlier and so dominate the scheme on strike day.

Martin’s used Processing, a rich Java graphics library, to create the animation, which has been then output to video. This allows the up-to-1000 bikes to be animated smoothly and effectively.

The bikes are in official Barclays Blue, although if you don’t view the video in HD, they look slightly washed out. Watch the video on the Vimeo website in HD, although you’ll need a fast computer and a broadband connection.

The routing is done based on the OpenStreetMap data for central London. I used Routino to do the routing, producing a routing file for each of the 137,000 possible journeys between docks in London. The routing is directed, meaning the bikes won’t cycle the wrong way down a one-way street. They also generally avoid trunk roads, such as Euston Road, preferring to use the quieter roads and dedicated cycle lanes nearby. Being able to use the new cycling infrastructure in the routing, is one big advantage of using OpenStreetMap.

A disadvantage is where the routing is wrong. For example, access from the Embankment is not shown correctly. Another problem was the reluctance to cross Trafalgar Square in the centre of the city. This meant I had to move a couple of the docking stations slightly. An example of the latter is shown in the picture here. These quirks, and a few others, result in some bikes flying around the animation extremely fast, as the router sends them a mile up in one direction, around a roundabout, and back down in the other direction. The speeds of the bikes are based on the duration information for the journey, which is included in the data, so they start and finish at the right time.

The routing is the “best guess” route, based on the assumption that the majority of cycle users will know the “best” route to take. Casual and multi-stop use will be less accurately shown. Bikes which are returned to the same docking station they started from, are shown “orbiting” the dock for four times, before returning to it.

The work follows on from a recent animation showing the TfL buses in London, by Anil Bawa-Cavia, also here at the UCL Centre for Advanced Spatial Analysis in London.

Categories
Data Graphics London

London Surnames: An Onomap of London

I’ve created a website to showcase a number of bespoke typographic maps that James Cheshire (a Ph.D colleague UCL Geography) has created. The website shows the origin of the most common 15 surnames in each MSOA in London – MSOAs are spatial units roughly encompassing 7000 people.

It’s important to emphasise that these are the origins of the surnames, not of the people themselves, i.e. it is a map of names, not ethnicities. The categories are chosen descriptions of the Onomap groupings that appear when matching surnames and forenames and therefore don’t necessarily line up with associated ethnicities or countries. For example, the high numbers of “Welsh” names appearing is likely not due to lots of Welsh people!

Communities which have more homogenous surnames are more likely to be highlighted on a map like that, at the expense of communities with more mixed surnames – another reason why this map cannot tell you about the proportion of people in a particular area – just their names.

The extract below shows a small Jewish “cluster” of names appearing on the border of Harringay, Hackney and Waltham Forest boroughs – the area known as Stamford Hill, which is an area noted for its large Orthodox Jewish community.

The website itself is nothing particularly special, except that it allows easy panning, zooming and scrolling of James’ eye-catching maps. OpenLayers powers the website, and a custom “pixel-coordinate” projection is used. There is a JQuery slider to scroll through the maps, and the user interface elements adopt the “London street sign” look. The data comes from 2001 so doesn’t account for the likely recent significant population movements around the city in the last few years.

More on James’ blog.

Categories
Bike Share Data Graphics London

Barclays Cycle Hire – Extending East

Alexander Baxevanis, maker of the excellent free Cycle Hire Map app for the iPhone, has obtained a list of 227 proposed sites for the eastwards extension (and expansion of the existing area) of the Barclays Cycle Hires scheme through a Freedom of Information request on MySociety’s What Do They Know. Unfortunately TfL didn’t provide the exact locations of the proposed new docks, rather just the street names, or occasionally junctions.

I have taken the list and geocoded it – using Google Maps and Google Fusion Tables as a first pass, then manually geocoding the 40 or so that failed using OpenStreetMap data.

Red dots show the proposed new locations, with yellow dots showing the existing stands as of January 2011.

You can download the locations from the Google Fusion table here or view a larger version of the map here. See the FOI response for the source data set.

Very important caveats: Because the names are often only street names, the “dot” representing the new dock is placed fairly arbitrarily along the street – in reality, the actual location may be quite far along the street from this place. Consider that these locations are simply my guesses. Also, it is really important to emphasise these are the proposed locations – TfL has not yet started the planning process or consulted with the councils/residents yet. It is likely that quite a few of these will not actually be built, or will be relocated elsewhere, come later this year or early 2012 when the expansion goes live.

Along the way I discovered a number of curiosities, such as:

  • the official name for College Green – the bit of grass outside the Palace of Westminster where MPs are often interviewed – being Abingdon (or Abington?) Street Gardens.
  • a street that has just been born (photo) and doesn’t appear on any public web maps except OSM (now).
  • the various “marketing” names for the new residential skyscrapers appearing around Canary Wharf, such as Streamlight, Ability Place and Pan Peninsula.

Indeed, many of the proposed sites are outside these large new residential blocks, and also outside many of the DLR and train/tube stations in Tower Hamlets – unlike the initial launch of the scheme, there seems to be no shying away from placing stands right next to the stations, where commuters are likely to be piling onto them.

(I was very impressed with Openlayers/Canvas heatmaps the other day, so the first picture above is a heatmap showing dock density, for the fully extended scheme. The background for that picture is OpenStreetMap.)

Categories
Bike Share Data Graphics London Mashups OpenLayers

The First Million London Bike Share Journeys

Thanks to a FOI request from Adrian Short, Transport for London have recently released to their developers area details of 1.4 million bike share journeys. The data is believed to include all the journeys between 30 July 2010 and 3 November 2010, except those starting between midnight and 6am.

I’ve created a map which visualises these journeys – select a docking station and a time, and it will show the journeys that start/end at that dock, depending on the options chosen.

You can see the map here. On launching the site, an initial docking station – one outside Waterloo station – is selected, and an “interesting” timeframe is chosen – the morning of 4 October, which was a day impacted by a tube strike.

Heavy usage along the Broad Walk through Kensington Gardens, particularly at weekends:

The predominant flows from a docking station near King’s Cross station, in weekday mornings, are outwards (red lines), particularly south towards the river. Only a few inbound journeys happen (blue lines):

The reverse is true in weekday evenings, as commuters head back to the stations:

The map bears a resemblance to my live Barclays Cycle Hire scheme status map, as I’m reusing a lot of the same code and graphics.

Categories
Data Graphics Mashups

Boris Bikes – The Flows Are Coming

Adrian Short (@adrianshort) sent an FOI request for flow data for the first million journeys on the Barclays Cycle Hire bike shares in London (the “Boris Bikes”). TfL responded with a test dataset of the first 99 journeys -from roughly 6-7am on 30 July – and a promise that the data for the next 999,901 are coming!

I’m working on an adaptation of my bike share visualisation to show these flows. It’s not possible to show a million lines on the screen at the same time, so consolidation, selection and filtering will be applied, but it certainly is possible to show the first 100 – click on the graphic for a full-size image:

I’m using colour to indicate the direction of travel (see the wheel.) I’ve also shown the flows to a couple of specific docks:


I’ll build out the visualisation with the full data set and release it soon.

As start and end timings are included in the data, I’m sure it is only a matter of time before someone builds a version with little animated bikes moving from the stations into the city during the rush-hour, similar to Matthew Somerville’s excellent real-time tube visualisation, for which the underlying data unfortunately got pulled.

Categories
Data Graphics OpenStreetMap

Nike Grid – Visualising Runners on the Streets of London

My last eight posts have all been on bike share, time for a slight change of topic – running rather than cycling.

In the last few days, I’ve been taking part on the Nike Grid alternative reality game (a futuristic take on street-o). The concept is a great use of social media – with an active Facebook group, key updates pushed to participants phones and Facebook walls, and a Foursquare-esque concept of “checking in” to the phoneboxes which act as the run timers, starting and stopping clocks and noting locations. How do you “check in”? You make a (free) phone call.

There is a strong mapping element to the game – online maps show the locations of the key phoneboxes in each postcode, the maps appear in printed form and as artwork on the technical T-shirts included in player packs sent to key participants.

The maps are based on OpenStreetMap data, heavily stylised in black, grey and white with a “region”-specific pattern for the background and another pattern used for parks. The phoneboxes are “pin” style icons placed on top. The maps have been produced by Stamen Design in San Francisco. It’s not the first time they’ve done cool things with OSM data.

Stamen are also producing daily visualisations of the runs. The run lines have a hexagonal style to them, which goes along with the hexagonal tiling of the 48 postcodes being used in the game, although the start/end points are geographically accurate. A hexagonal cartogram is used on the main website to show the postcodes in pseudo-geographic space, in some of the visualisation the hexagons then “explode” and move to their correct place on the geographic map – a clever linking of cartograms and geographic maps.

Categories
Data Graphics Mashups Technical

Fewer Cities, More Cities

Some bad news and good news about the Bike Share visualisation.

The bad news – the operator behind the schemes in Paris, Seville, Vienna, Dublin, Brussels, Valencia and Toyama asked me to stop getting the current bike share data from their websites. Although I was just loading their webpages, “in practice you are extracting data from [the operator’s] databases and re-utilising it” and “[the] databases are protected under the harmonised sui generis database right, as provided under Directive 96/9/EC: chapter III article 7 (1) and (2).”

For these seven cities, you can still see a historical snapshot from last Monday, when the feeds were switched off, but not the live status, historical animation or trend graphs.

This is despite a quick search on the web revealing a six-month collection of data for one of the schemes (at four minute intervals), the resulting trends being shown at a conference; a better-service campaign website, again for one of the schemes, with regularly updated performance tables; and an iPhone app pulling in the data from numerous schemes run by the operator, amongst others.

Digital Urban also mentioned this in the context of Bike-o-Meter, which uses the aggregated data from my Bike Share maps.

Now for the good news – I’ve added in five more cities – Rennes, Bordeaux, Zaragoza, Mexico City and Rio de Janerio. Yay! The inclusion of Mexico City and Rio should hopefully counter some claims of an European/English-speaking bias! Mexico City’s scheme appears to be concentrated in one very affluent district of the metropolis, while Rio’s is based on the seafront south of the city, rather than in the main urban area.

Rennes is a particularly interesting example, more about that shortly.

[Update – turns out I’m not the first.]

Categories
Conferences Data Graphics

Visualising Bike Share

Here’s the presentation that I gave at the #geomob London Geo-mobile developers meetup at UCL last night.

[slideshare id=5528647&doc=visualisingbikeshare-101022061626-phpapp01 width=”590″ height=”480″]

Please note the data presented is preliminary and unreviewed and should therefore not be considered to be definitive or necessarily correct.

Categories
Data Graphics

Real Life Tweet-o-Meters

I was at the British Library yesterday for the launch of the Growing Knowledge exhibition of innovative research techniques. One installation has been built by Steve and Ben at CASA and is a real-life version of the Tweet-o-Meters (which were also the inspiration and technology for the Bike-o-Meters I mentioned yesterday.)

The installation has dials for nine cities around the world, showing the current level of Twitter activity (i.e. geo-located tweets) in these locations.

I love the “1930s retro” design of the installation. It is notable that all the other installations in the exhibition involve computer screens, in several cases these are used to display old maps (e.g. the New York Public Library rectification service) or historical paintings (using a Microsoft Surface screen.) I love the irony that the exhibition that is showing the data right now, i.e. coming live off Twitter from around the world, is the one which doesn’t involve any computer screens at all – although they are of course computer-controlled behind the scenes.

There’s something wonderfully organic about seeing the needles go ricochetting off the ends of the dials, as sudden bursts of tweets from a particular city come in. I hope the distinctly analogue technology survives. I think we get the work when the exhibition closes next summer. I’m pretty sure, when Steve’s not looking, it will be quite straightforward to “retro-fit” it for a physical monitor of bike share schemes. 😉

Steve has posted some more pictures from the exhibition, including some behind-the-scenes shots.