Mapping

Mapping — 1 Comment
05
Jul 10

The 2010 World Cup in maps

I’ve spent too much time watching the World Cup in recent weeks to really post anything on here – what better way to get moving again with a look at some of the maps used in web coverage of the tournament in South Africa.

The BBC (Disclaimer:  I work for the BBC, although had nothing to do with this map) turn out a decent venue map with some good photos and brief information on each stadium being used for the tournament in South Africa.

BBC World Cup Map

Mapsoftheworld.com has another example of the classic venue map which offers more in the way of city information but doesn’t really make the best use of the interactivity offered by the map.

The official Fifa website have also gone down the map route for their guide to the host cities of the South Africa World Cup but have only really used the map as a basic piece of navigation, although there is plenty of information about each location when you click through.

The Telegraph newspaper has taken a slightly different approach and simply embedded the Google map for each stadium location within their site – useful for those actually attending the match or visiting the city but less useful for the interested supporter back home.

SA-venues. com have a website dedicated to finding you accommodation in South Africa and have used maps quiet extensively to help you select the best location in a specific World Cup Accommodation section of the site.

SA Venue Map

The main website is quite confusing and it took me a while to work out that selecting a city on the map changed the contents of the right hand column on the page whilst the map often stayed exactly the same, allowing me to select a different city in the same area.

However, I do like the simplified but effective addition of distance markers on the world cup stadium maps to help traveling fans find somewhere to stay within a certain distance of the ground.

Theworldcupmap.com are claiming a “worlds first” by providing a map that combines Microsoft Silverlight Deep Zoom technology with high definition aerial imagery to  create a really pleasing guide to the world cup venues in South Africa.

theworldcupmap.com

According to the press release:

In a pioneering online collaboration, the fields of High Definition (HD) aerial filming and web based mapping have been combined to create the world’s first cloud hosted interactive map that integrates HD aerial footage.

For fans not traveling to South Africa – specifically those living near San Francisco, USA – here’s a handy map for people living in the San Francisco Bay Area produced by the San Francisco Chronicle displaying local places to watch the action live.  Also includes an indication of any national affiliation the venue might have to help you pick the perfect viewing spot for your needs.

And finally… There’s something about a major sporting event, especially one that only takes place every four years, that tempts people to go that extra mile.

The 2010wallsroadtrip blog documents one family’s trip around South Africa, taking in a total of 16 matches, visiting each stadium, and driving over 4,000 km’s in the process.

There’s a Google map of their route on the website with comments and blog posts added as the journey unfolds.

Conferences / Location Services / Mapping / Technology — 3 Comments
29
Apr 10

Location Business Summit – Day 2

No blogging this morning as I was busy presenting a view of location services at the BBC and then speaking on a panel discussion.

Thoroughly enjoyable, and the best panel discussion yet according to @uphamb.

Follow #locbiz f you want a rundown of events so far.

Now back to sitting, watching, listening and blogging.

Joel Grossman from Wavemarket is talking about location aggregation, with particular reference to mobile phones.

Family Locator, a successful 2005 downloadable mobile application although plenty of difficulties and challenges.

Version 2 launched in 2006 and migrated the interface to the web which resulted in exponential growth over the initial 3 months.

Apparently there are 600 billion SMS per year in the US that ask “where are you?”.

Add to this the fact that 100% of phones use SMS but only 15% are smartphones with access to location API and there appears a great opportunity.

Not for the first time at this conference we’re talking about how to obtain metrics on the success of mobile advertising to convert an advert view into a visit to the store.

I’m not the first to mention this, but why not focus on selling the product directly via the mobile rather than trying to force consumers from a mobile advert to a real-world store – surely that would be easier?  And measurable.

Motti Kushner from Telmap is now going to talk about a winning strategy for successful location based services.

  1. Ensure location capability on every device
  2. Make it local, relevant and valuable – understand the consumer/audience needs
  3. Then monetize – and this needs more thought than simply adopting a standard advertising approach

Good emphasis on the second point on how users behaviour differs depending on their social/local profile and location.

Importantly, the services/tools people use for local tasks (e.g. finding a restaurant) are different in countries across the world.

On the monetization front, a strong view that the answer lies with the mobile operator.

Pillars for success

  • An open LBS platform
  • Local content and services relevant to the location
  • Active widgets library integrating products and tools
  • A rich location companion on the client side with location based advertising

Tony Jebara from Sense Networks talks about the old view of LBS data – obtaining the single point reference to the nearest Starbucks for a potential customer.

The new view consists of a building a location data history to enable better understanding of the individual user, and therefore offer more personalisation.

Sense Networks have a couple of applications that are built on this analysis of historic data as well as using the present, real time location information.

CitySense – Real-time density of users on every street corner in San Francisco captured every ten minutes.

CabSense – Tracking 13,000 New York City taxis to help people find the best location to hail a taxi.  Also captures data by day and time to allow views of best places to get cabs at specific times of the day and week.

CitySense will not just show you where lots of people are, it will show you where lots of people like you are.

This is achieved by a user segmentation model derived by analysing mobile phone data, GPS trails, etc and tracking the places each individual user lives, eats, drinks, and spends their time.

Mathematical models then work out the probability of user x being of a certain age, wealth, life status, and so on.

Lots of fantastic network models and diagrams – who is like who, who drinks with who, which people are similar to others, and so on.

The data model is updated approx once a week and the analytics API is due to be released this summer which will allow everyone to tap into this user data.

Next it’s Claire Boonstra, from Layar, with some great examples of augmented reality in action.

In making a good case for AR as a powerful new tool it’s clear that it is interesting, entertaining, clever and enjoyable to look at.  But what about the consumer need or, more to the point, the business model?

Layar is the world’s largest AR community with over 1.6m users, 3000+ developers, 500+ layers published, and 2000+ layers currently in testing.

Layar.com/create if you want to get involved.

Layar positioning is derived from GPS and in-phone compass so only really works outdoors in areas of good GPS visibility.

The final session before lunch is a panel discussion on destination content and travel services that users are prepared to pay for.

A familiar, trusted editorial voice and brand is the added value over a search based information service say those speaking for Rough Guide and WCity.

Interesting comparison between editorially produced and edited guides and services such as Yelp which gives a sense of place based on user opinion.

Discussion in the room suggests that most people would prefer to pay for trusted, valuable content where relevant, rather than rely on possibly less-trustworthy user generated content, even when free.

I’m reminded of recent discussions in the UK over local journalism in print and on the web, will paywalls work around news, and so on.

I wonder if this has something to do with the fact that the purchase of a trusted travel guide is always a relatively small additional cost in comparison to the cost of the overall trip, however cheap.

A free UGC based where user opinion is valuable such as a restaurant review might be a good supplement to a straightforward history of Buckingham Palace for example.

The afternoon kicks off with a panel discussion on the topic of “The threat and opportunity of ‘free’ services” involving Philip Gontier from Multiplied Media, Joel Grossman from Wavemarket and Nick Black from CloudMade.

Initial consensus that ‘free’ has driven awareness and consumer take-up of location services which has allowed non-free models to develop due to the increased user basis.

‘Freemium’ is a word that I’ve never heard before, but seems to refer to a model whereby developers get a bunch of code for free, but then pay for more functionality.  I think.

Another question, is it realistic for free mobile applications to deliver enough in advertising revenue to make it worthwhile from a business point of view.

An example of an app that had 500,000 downloads but only generated about £2,000 over 6 months.

Sounds obvious but the point is made that it’s better to have less users on a more focused, relevant app than large numbers of downloads that doesn’t translate into users.

Plenty more discussion about business models and advantages and disadvantages of free services in the market dominated the rest of the afternoon.

Around mapping it was suggested, and pretty much agreed, that base layer mapping would become free (following the model of OpenStreetMap for example) and the ‘free-ness’ would start drifting up through the chain with more and more location based data becoming available for nothing.

David Gordon, from Intel, summed up the strategic view quite nicely.

“Free is inevitable”, he said, it’s just matter of which services become free first and how businesses and individuals move around to find areas that have revenue potential.

Potentially threatening from a business point of view but genuinely exciting from an industry perspective – a world with free data, free tools, free location awareness, and free connectivity… but when?

Conferences / Location Services / Mapping / Technology — 1 Comment
28
Apr 10

The Location Business Summit, #locbiz – Day 1, 28th April 2010, Amsterdam

Some highlights from day one of The Location Business Summit in Amsterdam.

Annette Zimmermann from Gartner began the conference with some analysis and survey results on the usage of location services.

Unsurprisingly it is expected by 2014 that a third of mobile phones in use will be smartphones – i.e. phones capable of delivering location awareness.

Symbian is the dominant operating platform but the gap with others is narrowing, a predicted 35% share in 2013 in comparison to the android platform which is on the rise and predicted to have a 20% market share by 2013.

An interesting slide on who will influence various aspects of the location business in the coming years.

In summary, Nokia and Aloqa/Foursquare are the biggest influencers on location technology;  Nokia, Facebook and Google when it comes to a significant user base; and Apple on device integration and payment solutions.

Finally, some golden rules on the delivery of location services.

Do not:

  • duplicate anything that is already available
  • rely solely on location
  • require users to sign up on the website to use a mobile application
  • wait for the user to search for information
  • connect the user to others without their permission

David Gordon, from Intel, continued the theme of research and analysis on the current state and predictions for the future.

From a study conducted into users perspective of how much location privacy is worth (with 74 respondents) on average they would allow a mobile device to query their location every few minutes for every minute of every day for about £27 p/month.

Next up, Ed Parsons from Google with “It’s all about data, stupid!”.

First the question, Are we there yet?

Well, Ed thinks that we might be given that:

  • Data plans have improved dramatically and are, almost, simple and cheap
  • Mobile browsers have moved on from WAP and now deliver a web like experience
  • Devices are improving with smartphones now becoming common

Another indication of progress is  that the availability of GPS technology within mobile devices has increased 92% since 2009.

GPS has followed in the footsteps of the addition of cameras to mobile phones – the biggest camera manufacturer in the world is now Nokia, not Canon or Nikon.

Crucial to the success of map based services is the innovation of tools to build better base maps – gone are the days when centralised organisations with high levels of resource required long timelines to build an accurate view of the world around us.

OpenStreetMap led the way but Ed suggests this is “geeky data and not necessarily usable for commercial applications”.

Google Maps enables anyone to add new information to the base map data which is then usable via the standard API.

It also allows users to report a problem and notify the Google maps database of any changes to the underlying base map – i.e. when a bridge collapses or a new road is built.

Place = Point of Interest + Person

Essentially, “for a place to be important, people have to say that the place is important”.  An important link between the location in it’s pure form and the value of that location as derived from social means.

I’m not sure this applies to the attribution of value to every location, for example,  the place where I was born is of interest to me regardless of what others think, but it is good to see the social element recognised alongside the value of location information.

Gary Gale (@vicchi) from Yahoo! is next on stage to present “Hyperlocal or Hype (and Local)?”

Anyone that’s heard my thoughts on this subject before will know how much I dislike the phrase ‘hyperlocal’ so I’m looking forward to this.

Great set of introductory slides on the history of location that starts with smoke signals, through homing pigeons in 1000 BC, and into 1600 and detailed maps of the world.

1960 brought early GPS, 2000 the arrival of the smartphone, and now where next?

A hyperlocal service is something that:

  • is built on entities and events defined by a community
  • is intended for local people and visitors
  • is normally created by a resident or visitor

Different genres of hyperlocal:

  • Classic hyperlocal:  Traditional media delivered through hyperlocal services
  • Corporate hyperlocal: e.g. “15th avenue coffee and tea” is a re-branded Starbucks.  Or the  Tesco iPhone app that tells me where my nearest Tesco store is.
  • Social hyperlocal:  Applications such as Foursquare that create services around people sharing information.

There are different levels of localness (local, hyperlocal, microlocal) but several elephants in the room.

First, a user’s IP address is not hyperlocal and not accurate enough right now.

Second, the issue of ‘hyperlocal fakery’ where users can claim ownership over local services or claim local knowledge without any authentication or guarantee of their claim.

Third, different standards and a variety of services leading the way means different reference values to the same location.

I think this means, how can we share information across the Geoweb when we are defining the same place in different ways?

Hyperlocal or Hype (and Local)?

View more presentations from Gary Gale.

Frank Albert Coates talks through the recent advancements from Google in local mobile advertising and also mentions last week’s launch of Google’s free route navigation service to the UK.

Then an extended panel discussion on how mobile location and advertising could start delivering significant revenue streams for the industry.

Top line seems to be that there is plenty of opportunity in the mobile market, with the growth of smartphones in particular, to deliver focused local advertising that translates into real money – either in the real world or through the increasing number of transactions taking place on the mobile platform.

Scott Seaborn, Ogilvy Group, is keen to stress that too many business are looking to build iPhone apps when only 5% of businesses have a mobile version of their website.

He also talked about a book store that was losing between 12-20% of revenue from people walking into the store and buying a book from Amazon via their mobile phone.

The reason?  Discovery doesn’t happen on-line.  Recommendations and profile-based suggestions are targeted promotions.

Only in the real world can you truly discover something interesting in a ‘random’ fashion.

Apparently, in Japan, 87% of all flights are purchased on a mobile device, says Paul Lyonette from Microsoft Advertising.  Presumably that means on mobile web services rather than simply making a phone call from a mobile.

Attention turns away from talk of money and advertising to community with Henk Hoff, board member of the Open Street Map Foundation.

He defines ‘community as fun, not work. It’s about reward, not profit.  It’s about us, not me.

This illustrates the difference between community mapping in the model of OpenStreetMap and other technologies that gather user content and usage data to improve a service.

Responding to Ed Parson’s earlier “geeky” comments, Henk uses the example of the block of streets where he lives to ‘prove’ that ‘commercial’ maps are not that good with roads missing and incorrectly placed.

The point is,  “geeky data” can also be useful data.

240,000 registered users  have created an account to add and edit the OpenStreetMap map data – this number has doubled from the level of users last year.

Some interesting information on a forthcoming change of licensing model for organisations that use OSM maps.

Currently all use of OpenStreetMap takes place under a Creative Commons attribution and share alike license.

This has put off some companies that want to use OpenStreetMap with copyrighted datasets but don’t want, or aren’t able, to make the final product shareable in this way.

The new model – coming soon – is an open database license which allows any map produced from OSM and other datasets to be kept under full copyright if it has been implemented in the correct way.

The correct way means keeping the databases separate and implementing the different data sets as layers onto the map.

However, if a derivative database is created which augments OpenStreetMap data with an organisation’s own data, before this is implemented on the map, then this new derivative database should be made available as a data set under an equivalent free license.

At least, I hope that is correct!

Finally, ending on a massive positive, Henk showed how OpenStreetMap was used to respond very quickly to help emergency services direct the aid effort in Haiti after the recent earthquake.

And a fascinating map of the Kibera slum in Kenya where detail has been added to show the locations of places of worship and locations of water supplies.

An area that is simply a large, blank space on any other map.

Journalism / Mapping — 1 Comment
02
Sep 09

Using maps on news websites

Paul Bradshaw posted some thoughts yesterday on the use of maps on news websites, a subject often discussed on this blog.

Most recent ‘highlights’ are discussed such as MPs expenses, although Mapumental from MySociety is not listed, nor is my favourite longstanding example of a local news map, the London SE1 News Map.

There is a good list of the advantages in using maps:

  • They provide an easy way to grasp a story at a glance
  • They allow users to drill down to relevant information local to them very quickly
  • Maps can be created very easily, and added to relatively easily by non-journalists
  • Maps draw on structured data, making them a very useful way to present data such as schools tables, crime statistics or petrol prices
  • They can be automated, updating in response to real-time information

However, the post doesn’t really get stuck into the difficulties and disadvantages of this approach. I’ve outlined a couple of points below which I believe are major barriers to a successful map-based news website.

User Experience / Usability
User research suggests that most of the audience still see a map as a route-finding device, a answer to “show me how to get from A to B” or “tell me where this building is”, whereas news has long been consumed in a linear fashion, “Give me the big story of the day, what’s the second most important item, and so on”.

Mapping functionality is also quite complicated for a lot of web users and you cannot rely on the audience easily understanding how to pan, zoom, scroll a map, or cope with the differences in graphical pin-points, hover panels, embedded audio/video content.

Web users generally want to get at information quickly, particularly time-sensitive journalism content, and any design or interface barriers that prevent this can be quite a turn-off.

Geo-tagging or “Where do i put this story?”
It would be simply fantastic to be able to assign a latitude/longitude tag to every piece of content we create so that it can appear at the correct location on a map.

The reality is that this is simply not possible and even if it were possible, further difficulties emerge.

For example, imagine a bunch of stories appearing this week on transfer deadline day relating to one football club making several new player signings, whilst also being in the news for other financial or business reasons. We can easily tag all stories with the lat/long relating to the football club’s home stadium, but how do users easily find and navigate between a cluster of stories located at the same point on the map?

Often there is more than one relevant location that can be associated with a news story.
For example, a person from location A, in partnership with another person from location B, is arrested for an armed robbery in location C, and the forthcoming trial will take place at location D.

Can we geo-tag this story with all relevant locations? Does the technology understand the difference between each location? Does it matter?

Importantly, whatever the technology or editorial strategy delivers, how do we communicate this to our audience within the mapping interface? An interface which is already quite complicated and overcrowded with the standard set of mapping tools, place labels, and options for different graphical layers.

Then we have the regional story.
For example, House prices in Devon have fallen by X.

Devon is generally not a point on a map (it might be if the map displayed all of Europe in a small enough image size on screen and it was not possible to zoom in!) and yet we have a potentially important news story that needs to be geo-tagged.

Postcodes introduce a similar problem as they are also polygonal areas rather than specific points. Tagging content with postcode does not provide a dot on the map for that content to be assigned to, it provides a region boundary similar to county, borough, electoral ward, and the many other areas of this type.

News stories can relate to other shape patterns. A news story about a river or a train journey needs a line on the map.

Some of these examples can be solved by providing a single point approximation to represent the story, although this can also be dangerous.

The BNP membership map from 2008 provided several lessons in this area. At one point, whilst trying not to pinpoint exact houses, postcodes where used to make the information less specific. Unfortunately the mapping tool, in trying to be clever with the data, plotted the information at a specific point in the centre of the postcode, thereby seeming to be an accurate house-by-house set of data – but unfortunately pointing at all the wrong points on the map.

What’s the big story?
As I mentioned under the usability heading above, audiences have become very familiar with consuming news in a certain way that is very different to the map based approach.

A map is perfect for showing me something that is happening near to me, especially if it’s a story that wouldn’t normally make the top headlines, but I still want someone to tell me the big news, even if it is a little further away.

And that, I think, is the key target right now. Not simply solving these challenges and aiming to get as much of our content as possible onto a map, but finding the right way to include mapping as part of the consumption of news content.

Speaking at Where2.0 in 2008 Adrian Holovaty stated that ‘One question i like to ask myself is, would my site succeed without maps?‘.

It’s a very good question to keep asking.

Journalism / Mapping — No Comments
29
Jul 09

Microsoft to partner with local newspapers for MSN Local News Map

Microsoft, through their MSN Local portal, are hoping to provide local news on a map, in partnership with local newspapers across the UK.

Peter Bale, executive producer of MSN, said: “We are hoping to take feeds from local newspapers and tag every piece of information to a map. Hyper-local news online has never been more important and we think this is a really interesting growth area.”

Very interested to see how this compares with Trinity Mirror’s beta news map in Merseyside which has improved since it’s launch last year but still does not feel very user friendly.

Geo-tagging news content is a really complex task and presenting this on a map defies the usual logic of consuming news in order of importance, can’t wait to see how Microsoft tackle this.

(From Telegraph.co.uk)

Mapping — 1 Comment
08
Jul 09

Setting the power of postcodes free

Ernest Marples Postcodes Ltd, whoever they may be, are encouraging you to use their free API to geocode your UK postcode data, with a plea to the Government and Royal Mail to make this ‘official’.

Sites already using the service include:-

← Older Entries