The National Library of Scotland’s high-quality scans of historical Scottish mapping have been made available under an Attribution licence, which means they can now be used to trace features for OpenStreetMap in Scotland. While the maps themselves were already out of copyright, the high quality scanned imagery itself was still subject to copyright.

You might think that, with the recent releases of up-to-date(ish) Ordnance Survey mapping in various scales and formats for the whole of the UK, historical mapping is less useful for OpenStreetMap. After all, why use 60-year-old mapping, available in raster form and available at (relatively) low resolutions, when resolution-free vector data, and 1:10000 rasters of the same area are similarly available. But these beautiful old maps contain a lot of detail not on the newer ones, and for large parts of rural Scotland, where roads, rivers and mountain features very rarely change, they will still be enormously useful for completing the more remote parts of the country.

The historical maps can be viewed here. As you zoom in a few levels, the projection changes into the regular EPSG900913 (the tilted north lines are a tell-tale sign) that can be used directly in OpenStreetMap editors such as Potlatch.

Here’s Applecross, a place near Skye in North-West Scotland that I have long wanted to visit. In this particular area, historical imagery (OS 7th Edition) is already available on OpenStreetMap, allowing for tracing, but this is not the case for most of Scotland, as many of the sheets are still in copyright.

National Library of Scotland (Historical):

Ordnance Survey Street View:

Ordnance Survey 7th Edition (Historical):

The current OpenStreetMap Mapnik render:

So, I ran in the Nike Grid ARG (alternative reality game) on Saturday, concentrating mainly on the E9 postcode in Hackney, but also going jogging around the City of London (EC1, EC2, EC3 postcodes) doing an informal City of London Race. The aim of the game was to log runs between four specially designated phoneboxes in each postcode, dialing in at the start and end of each leg. The more legs done, the more points you got – bonus points were available for running early/late, doing a fast run, completing every possible leg, and the most number of legs.

My strategy was hampered by having a severe hangover from the night before, so I didn’t make it out of the house until 3pm (the game ran from 8pm-8pm) and was pretty dehydrated. It was also a very warm day – and, to make things worse, the phoneboxes themselves acted as heat reservoirs. One City leg went via a supermarket and its chiller cabinet…

In my first session I essentially ran all of the six possible legs between the four phoneboxes, and several extra legs between the two closest ones. In the later session (after my jog around the City) I again aimed to run all six possible legs, getting the fastest split bonus for each, but realised near the end I wasn’t going to make it to/from the far one, so repeated some of the smaller legs. The many people enjoying a cool drink in the garden outside the Royal Inn on the Park, immediately opposite the most southerly phonebox, must have wondered what was going on.

The map below shows the routes I took between the four phoneboxes, marked with green rectangles:

In total I ran around 16.5km (10 miles) in the E9 postcode. The phonebox dialing process meant I essentially had a two minute rest after every leg – the longest of which I did in just under 10 minutes. My shortest leg was 1m 26 – I tried this one again and again but my times kept getting worse with each attempt!

I ran into the last box about 10 seconds before the game closed – I had to push it for this final leg and got bonus points for running this leg in the fastest time. (In fact I think I picked up all six of the fastest leg bonuses during the day.) The Nike team were filming this last phonebox and interviewed me afterwards.

I was extremely unlucky not to win – notice how close I finished to the eventual winner in the leaderboard below. However I did get 110 of my points in the dying seconds of the race. The guy who finished third appeared at the same phonebox a minute later (i.e. too late) and, had our arrivals been reversed, he would have finished in front of me.

Although I didn’t win, a friend won not once but twice in a different postcode, so I’ll at least get to see what prizes I missed out on!

There were some “bugs” in the game – certain phoneboxes in the City had quite unresponsive keypads which made it difficult to clock in at the end of the leg. Quite often, the automated service appeared overloaded and stopped talking half-way through, leaving you wondering whether the run had been correctly logged or not. The game leaderboard was updated in real time, which was impressive, but it was written in Flash so I was unable to see how I was doing on my iPhone. (A dedicated iPhone app would have been cool.) There luckily weren’t many players in my postcode, but many more would have clogged up the system – it took 1-2 minutes in the phonebox to stop and start each leg. Some clarity on how many points were on offer would have helped me refine the strategy, although I suppose part of the challenge is figuring it out for yourself. A couple of “test” 3am short legs I tried on my way back from the pub didn’t count for “early” bonus points, although game messages suggested they would at that time. Finally the maps weren’t too great – some phoneboxes were in the wrong place. I had however done a bit of online research first though and used a marked orienteering map instead, so this didn’t affect me. A friend of mine greatly benefited from one phonebox not being themed – he was the only person in that postcode who realised it was still a game phonebox and so completely destroyed the opposition.

It must have been a nightmare to organise, with nearly 150 postboxes scattered across many miles that needed theming, maps distributed to them, checking and fixing them – not to mention answering the many and varied questions and complaints on the Facebook event page, and writing the software to handle the automatic logging, updating and cheat detection.

Overall I really enjoyed the style of the event. There was definitely something of “The Matrix” about sprinting through the grimy streets to a phonebox (themed in green and black, too!) and breathlessly grabbing the receiver in front of surprised bystanders. All things considering, it was a nice “Real Life 2.0″ take on the street orienteering theme. Not sure we’ll see this repeated – Nike generally organise a “concept” event in London yearly but each year’s idea changes dramatically to keep things fresh – however I would certainly love to try it again.

The OpenStreetMap project, as of today, has over 250,000 registered users. It is fair to say that most of these will never edit the map, or have just edited it to put their house in and then don’t return, but there are also a large number of active contributors to the project, such as the London community. Over 40,000 “ways” (generally, roads) are being added to the project every day. The project is continuing to grow, and the release of usable Ordnance Survey (OS) data covering the country, at the beginning of the month, looks like it will advance the project, rather than reduce its relevancy in the U.K. For the first time, a quick way to “complete” the roads is available, but there are many other features still to add which will keep contributors busy for years to come. For one thing, none of the OS data includes paths and tracks, and it’s not completely up to date, unlike OSM which sometimes gets roads added on the day they open – or before!

The London community at the moment is concentrating on filling in the building areas in central London, so the map here looks less like a “patchwork quilt” of filled and unfilled blocks, and more like the contiguous mass you see in other cities like Stockholm, Frankfurt or Milan. (Having said that, we are well ahead of Paris, Barcelona, and, suprisingly, Berlin.)

Our next pub/map meetup is in Holborn on Wednesday evening – come along!

Nike are running an event next Friday/Saturday in inner London called Nike Grid. It’s a great idea – basically players run between any two specially marked phoneboxes in a postcode area (e.g. E9). Typically there are 3 or 4 such phoneboxes in each area, each temporarily branded with the event logo. At the beginning and end of the leg, the player phones a special number from the phonebox, entering their player code. As the call comes from the phonebox, it’s proof that the player is there then. Players then earn badges by doing the most number of runs in a postcode, doing all the possible combinations, the fastest run, the hilliest run, etc.

Like I say, a great idea. It’s a technologically advanced version of street orienteering, similar to what my club has been running in similar locations in central London over the winter and it’s a shame that Nike doesn’t mention the “o” world anywhere in their publicity for the event – but maybe orienteering is a bit anoraky for their brand experts? (Nike don’t make orienteering shoes anyway, but their big rivals, Adidas, do – my current o-shoes are Swoop 2s.) It’s a missed opportunity to promote the (sub)-sport to a market that likes running, is happy to be holding a map as a different challenge, but has never heard of orienteering.

On the left is part of the map my club used for a street-o in Bow, below it is Nike’s version.

To pick your way between phoneboxes, you get a map – downloadable from the website, or collectable in paper form from the phoneboxes themselves or the Nike stores in London. There’s four maps, representing south, west, north and east London – the coverage generally extends out to the edge of zone 2. I visited a few of the phoneboxes this evening and picked up the north and the west maps (the south and east ones haven’t been put out yet, or have all been swiped already). On the maps, the phoneboxes are shown as green hexagons and the rest of the map is a rather pleasingly mimimalistic white-on-black design, rather like some of the other great cartography you can create out of the OpenStreetMap data for inner London that I and other project volunteers have collected.

In fact, wait a minute. Some of the detail on the maps around my home area looks rather familiar. Yes, they have actually used OpenStreetMap data for the map. I can see the characteristic kinks in the paths in my local park that I surveyed and that don’t appear on OS/Google/Teleatlas/Navteq et al map data. Nothing wrong with that – using OpenStreetMap data commercially such as promoting a brand of shoes is just fine. Except they haven’t attributed the project or stated the licence the maps fall under – both requirements of using OpenStreetMap data to create a derived work, especially in printed form. Oops.

Why am I bothered? Contributors of open data don’t do it for the money (mostly) but for the “kudos”. In the case of OSM, the project itself typically gets attributed rather than specific contributors, for practical and logistical reasons. The contributors are still acknowledged in the data itself. The project benefits from acknowledgement because publicity will help increase the number of contributors to the project and so increase the quality and completeness of the map data, making it in turn more viable for future uses. Everyone wins.

All they need to do is (a) add a line to attribute the project, such as “Map data (c) OpenStreetMap and contributors, CC-BY-SA”, to the maps concerned on their website and future printed copies, and (b) not be surprised if people make derivative works from the maps, which is allowed by the licence the data has been used under. I’m tempted to create an interactive map for the whole of London or indeed the world, in the same style – the cartography is very nice.

Incidentally the map is created using quite an old copy of the data, from before last September – some of the more recent roads I and others have added to the project don’t appear. The designers have also enhanced the widths of some of the major roads, and added in road names and numbers. Roundabouts have also been added in as proper circles. There are some mistakes in the process they’ve used – the main track (highway=bridleway if I recall correctly) around Victoria Park doesn’t appear, but the paths (highway=path) that lead to it do, resulting in a rather odd “gappy” looking bit of cartography around there, ironically a similar quirk of the Google maps of the same area.

Also, I’m not sure where the postcode boundary lines come from, but they mis-align somewhat with the OpenStreetMap data – in some places the lines wander near, but not exactly along, the centreline of a boundary road. You can see a particularly bad mismatch between the green line (postal boundary) and the white line (here a canal) on the left of the first map above. Just a cosmetic quirk.

It is a really great idea, and a really nice bit of marketing. I will, hopefully, have a go at getting a few of the badges during the 24 hours the game runs. Let’s hope they get the attribution sorted out.

(I notice it’s happening the same weekend as the London Marathon, who have Nike’s rivals Adidas as a key sponsor. The timing is not a coincidence, I’m sure!)

By the way, Nike have made it very hard to be contacted about this – there are no contact details on the game’s website and it is not possible to send private messages to the owner of the game’s page on Facebook, thanks to the way the social network sets up fan pages. Sigh. Of course, people in glass houses and all that, I should attribute the screenshots in this blog post – all the screenshots are of maps created using map data (c) OpenStreetMap and contributors, CC-BY-SA.

[Update - I have made minor edits to improve the clarity of the article and add the note about Google.]

Last weekend I cycled around the perimeter of the building site that will be the Olympic Park in 2012. It’s a big site – over 9km all the way around.

Here’s some photos I took – click through to the Flickr pages for some slightly pithy commentary to go along with them:

Direct link to the photo set.

Here’s a brilliantly produced video showing what parkrun is all about.

It’s shot on location in (I think) Bushy Park in south-west London. Plenty of shots of the verdant park and the wildlife, as well as the runners.

I gave a talk on OpenOrienteeringMap at an (un)conference today – the slides are on my research blog.

One problem which my club has come across when assembling Street-O maps using OpenOrienteeringMap is where control labels clash with other features on the map – they may obscure important underlying topology, or intrude on the circles denoting other nearby controls, as shown in the first image below.

To fix this, you can manually specify the angle that the label appears at with respect to the control circle. I haven’t yet built this in to the online map builder, partly because I didn’t want to clutter the UI with too many options, partly because I haven’t got around to it, and partly because the variables that allow labels to be positioned as such is not yet in the released version of OpenLayers (it is in the trunk) that the map builder is based on, so I’ve had to patch my version specially – I’m reluctant to do too much coding in this area before it’s formally released.

…anyway, to fix it, first produce your map as normal. Then, notice the faint horizontal black line on the bottom left of the PDF? This is actually text which represents a partial URL that allows the map to be recreated. Zoom in to the bottom left in your PDF reader, and copy the line of text into a text editor.

The next step is to edit the label angles. Find the section starting “controls=“. Everything beyond this is a control, each control is defined as four numbers, each seperated by a comma (and no spaces). A comma is also used to separate each control. e.g. controls=1a,1b,1c,1d,2a,2b,2c,2d – where 1a, 1b, 1c and 1d are the four attributes for control 1, etc. The attributes are control number, label angle (clockwise from north), latitude and longitude. The lat/lon coordinates look funny – they use the “EPSG900913″ Google-style coordinates. The label angle is the one to change – they all default to 45 (top right), changing this to, for example, 180, will set the label to be directly below the circle.

Once this is done, copy the long line of text and append it to the URL http://tiler1.censusprofiler.org/pdf? in your web browser. Press enter and you should get back a new PDF with the corrected angles, as seen in the second image below. (N.B. the URL may change in the future, I’m piggy-backing off one of my work project websites at the moment for this.)

Before

After

OpenOrienteeringMap now updates in near-real-time from the OpenOrienteeringMap database. It does what’s called “minutely” updates, although the actual time-lag from updating something in OpenStreetMap (say, using Potlatch), to it appearing in OpenOrienteeringMap, is typically around five minutes – occasionally the delay may be up to 15 minutes, or the update doesn’t appear at all if something wrong happens. This is, however, a vast improvement on the per-month updating I was doing before – that process took a couple of days to run, this only takes a few seconds each minute.

Note if you are viewing OOM in your browser, your browser will probably cache the tiles, including ones which have since been updated, so you will want to do a “super-refresh” of the page (Control+Shift+R) or reopen your web browser, if you’ve made changes and want to see them reflected. Creating PDF maps always uses the database so doesn’t suffer from this caching issue.

A huge amount of adding and updating is going on around the world in OpenStreetMap – the project is growing by around 5% a week – so it’s good to be able to take advantage of people’s efforts on a timely basis.

OpenOrienteeringMap now has a map builder, that will create a PDF map of any area, optionally with score-course orienteering controls on it, for you to print out and run on. You choose the scale, orientation and style – street-O, pseud-O or a special street-O without railway lines. Adornments such as a title, north arrow, logo and attribution are added.

To use it, click on the “Create a Map” link. The builder (which is written in Javascript) doesn’t exactly match the final map 100% as it is just building up a URL that gets sent to another service, powered by Python, Mapnik and Cairo which actually creates the PDF map for printing – the latter is a high-quality vector map so should look great when printed out – no jaggies, except for point-features as Mapnik trunk doesn’t (yet) support SVG icons.

Tip: The URL used to create the map is reproduced in tiny-text on the bottom left of the map – it will look like a small black line until you zoom in to 1600% or so. Copy this text and add it to the end of “http://tiler1.censusprofiler.org/pdf?” to reproduce the map. You can tweak the size (in metres) of the PDF sheet, to accommodate unusual sheet sizes, and also change the title and control positions here rather than having to recreate the whole map in the builder.

I’m going to be documenting how the PDF map is produced, on my research blog, in the next few weeks.