Learning from Plazes – location based services


Recently, I started a presentation about location based services by the phrase ‘Location location location ?’

The question mark was not typo!

Where are the mass market location based services?

Inspite of its promise, location based services have yet to fully reveal their true potential.

Location is the chief differentiator for mobility based applications. In user terms, a location based application would deliver any content ‘filtered by location’ – for example, a dynamic map that mirrors a user’s location or a ‘find my nearest’ application where every fast food outlet in a one kilometer radius can be listed.

By definition, a Mobile service accessed from a device such as a mobile phone, is ‘location based’ because the location of a mobile device is always known to the subscriber’s mobile operator. However, merely the fact that your mobile operator knows the location of your device is not enough.

To truly enable mass market location based applications we need

a) The location to be available to trusted third parties

b) The cost of the location ‘ping’(i.e. request) to be feasible

c) Points of interest to be ‘mapped’

d) Privacy issues to be resolved

The conventional ‘top down’ approach does not seem to be working. In other words – the current situation, where the mobile operator knows the location and is the gate keeper to that information has not worked(and in my view will never work). That’s because – there are too many legal, privacy, cost issues involved.

It also needs the major parties (with conflicting interests) to work closely together else the market fragments. By this, I mean to be truly viable, we need an ‘umbrella entity’ which takes feeds from multiple location sources(operators) and then has the capacity to manage that feed.

An alternative is an organic approach

In my previous blog, Mobile web 2.0 – A service blueprint – I mentioned that locations could be ‘tagged’. Exploring this further, I found a site called Plazes – which seems to be doing this. Plazes also has addressed some key issues which would be used in future mobile web 2.0 services(and for location in general)

Firstly, what is it?

Plazes is ‘Location + context’. From their site

Plazes is a grassroots approach to location-aware interaction, using the local network you are connected to as location reference. Plazes allows you to share you location with the people you know and to discover people and plazes around you. It’s the navigation system for your social life and it’s absolutely free.

How it works

It seems to map a unique key created from the user’s network (based on the user’s router information) to a ‘Plaze’ that the user annotates. (I managed to annotate the futuretext office . I also saw Joi Ito online twice – which is an endorsement for the site in my book! )

The user names the Plaze, they can add pictures and comments. Others who are physically present at the Plaze can add information subsequently. On every Plaze, there is a box called ‘Discoverer’. The person who ‘discovers’ the Plaze manages this box and it can be linked to their web site or blog. Other information on that page can be edited by others

By the way, Plazes mixed with Google maps makes a neat mashup – check it out!

All this is fine .. But I was more interested in the social, web 2.0, mobility aspects. This is a pioneering service and the lessons learnt here could well be the norm (Italics are from the Plazes web site)


a) Privacy – You can use a pseudonym

b) Tracking – You can optionally allow plazes to ‘track’ you using ‘Trazes’. This is set off by default. By default we don’t record your traces. The moment you leave a plaze that information is gone. You past whereabouts cannot be compromised, since we don’t have them. Nevertheless we do provide a function called Trazes, where you can record your past wherebouts and publishing them to other users, if you explicitely wish to do so.

c) Visibility: You can choose to remain invisible from your friends(so no one knows your current location). You can use Plazes without the launcher, meaning you are visible, but your location is unknown. All features, obviously except the location aware searches, will still be available to you.

d) Centralised editorial staff: There is no centralised editorial staff

e) Mandatory information about the Plaze is kept to a minimum

f) Not spyware – Plazes makes great efforts to tell us that.

g) Data quality and control: Anyone physically present at a location can incrementally complement or alter the information for this plaze. Therefore the quality of data will increase with the number of users and frequency of usage. The most frequented Plazes will therefore have the best quality of information, because it is being reviewed most often.

h) Censorship: We do not censor pictures and comments. The description page for a plaze is basically a wiki. We believe in free speech and we strongly believe in you, the user. If someone is altering the information in a stupid or malicious way, we are sure someone else will correct it.

i) Compliance with geographic legal issues: There is one exception to that rule: The laws of some countries require us to delete unlawful pictures or writing, if explicitely pointed out to us. We will obey these laws, because we don’t like being in jail.

j) Obsolete and historic Plazes: The following passage is well worth a read because it encounters a problem not yet endemic .. but critical in future as we all map virtual elements to a physical world – the whole question of ‘virtual archeology’

Another issue in matters of data quality is obsolete Plazes. The unique identifier for a Plaze is the network, or to be more precise the router’s mac address. In other words, if you buy a new router for your Plaze or move the same router to a different Plaze, the information is either lost or incorrect. Let’s say you move into a new apartment and take your router with you. Easy you say, next time you log onto Plazes you just change you address information. Right, but what about those pictures of your old house, the comments for that party you threw last year, etc? Even worse, stay in your apartment and buy a new router. All your precious information will be gone, the history of your Plaze will be unwritten, forever lost.

This whole subject is a tough one. At first we thought the solution was going to be that you can just assign a plaze a new router. But this would either open all doors for a new sport called “Plaze-jacking” or an actual human would have to decide for every Plaze if the transition is legit. Not an option. So we came up with something called a “Historic Plaze”. A historic Plaze is an inactive Plaze at the same location. There is two ways for a Plaze to become historic: The discoverer can mark it historic, for instance if he moves out of his flat. This way, if the router pops up the next time it is being assigned a new Plaze.

If a Plaze is inactive for more than half a year and there is another Plaze with a similar description at the same location.

Historic Plazes are linked to the active Plaze’s description page so you can dig down into the Plaze’s history. It’s virtual archeology.

k) Taking social networking to the next level – in their words The Plazes you are frequenting are actually a much better filtering system and common denominator than explicit connections like “he is my friend”.

Plazes takes it to the next level regarding location-awareness and implicitness. The Plazes you are frequenting are actually a much better filtering system and common denominator than explicit connections like “he is my friend”. By being virtually present at certain Plazes like a record label or a certain restaurant and having conversations via comments at that Plaze, the system is much closer to how we actually interact in the real world. By being able to annotate real world locations virtually, Plazes augments, enhances and encourages real-world communication rather than simulating it.

In a nutshell, well worth visiting and trying – for no other reason than – it’s the way to go as I have been saying with mobile web 2.0. Plazes seems to have thought about many of the issues that I have been talking about. Besides, its fun!(By the way, I have no commercial affiliation with Plazes or the company creating it). I just think that an organic approach may be feasible – and the alternative (top down/operator led approach) will not see light of day!

Technorati tags