\ \ \
 
     
 
 

Locative Concepts
Ben Russell

This section attempts to give a brief overview of the potential capabilities of next generation location aware mobile devices; capabilities, which, if combined could potentially lead to significant shifts in the economic and social landscape.

Location and proximity sensing

Beyond GPS there are now a whole range of location sensing technologies:

GPS, WiFi triangulation, WiFi and bluetooth XML feeds that broadcast location information, 3G phone location finding, conventional current cell tower locating. With more coming from ultra wideband through chip level inertial navigation systems.

This means that a device can ascertain its position by multiple means, both indoors and outdoors and to varying degrees of accuracy.


Spatial annotation

A piece of text left at a place can be displayed on a mobile device when the user reaches that place. Conversely because the device knows its location, a user can leave media at a place.

Because of the recent proliferation of location sensing methods, locative media can be placed both indoors and outdoors and can be found and interacted with even when GPS satellites are inaccessible.

Beyond leaving things at specific places, such a system facilitates the (location aware device mediated) spatial demarcation of zones and paths (see headmap.org for more on spatial annotation).


Collaborative mapping

Collaborative mapping is the process by which groups generate their own maps.

People who use GPS systems tend to take records of waypoints, specific points of interest. But GPS devices also track the route you take, they generate GPS traces. If you upload the traces recording the routes you take to a database and others in the same area do the same the combination of traces soon begins to look like a map.

The Waag society put together a project called 'Amsterdam Realtime' (realtime.waag.org) they supplied locals with GPS systems and collected the traces and combined them to create the beginnings of an intricate map of Amsterdam. After 40 days the map was remarkably complete given the small number of participants.

This art project illustrates a serious point. If a lot of people have a GPS and they pool the traces you can make usable maps.

Mapquest produces commercial grade maps for the web. The maps start out from freely available Tiger census maps, they are then finessed using information from commercial drivers with GPS devices in their vehicles, and by an army of real people driving around validating and amending the maps against real places.

In India some maps are produced by university departments using commercially acquired satellite photos and people driving on motorbikes with GPS's.

As maps become increasingly object orientated structures rather than mere pictures, the question arises (once you have added all the utilities and the building outlines), what data are you going to associate with those map objects. The wealth of data about places lies in the hands of ordinary people; they can add a whole range of information that no centralised department could ever hope to accumulate. The opportunity exists to push data collection out to the edges (a very successful contemporary strategy).

Once people are annotating maps and generating the traces that keep those maps up to date, once cartographers delegate a large part of there work to a huge distributed workforce, maps are going to be different, and very dynamic animals.

It remains to develop software to allow large scale collaborative mapping that could result in maps more useful, mischievous, accurate and up to date than the current commercially produced maps.

In the UK copyright protection prevents a large part of the population getting access to maps they could annotate and contribute to, and worldwide, mapping is a big industry that is failing to address this new trend towards massively decentralised mapping. Mobile location aware devices are the key to providing the data that could transform commercial cartography.


Group coordination

High resolution screens, 3d capable, high quality audio, gps, wireless internet connectivity, open many to many voice channels… we're very close to full on location-aware gaming. Manufacturers are releasing extremely powerful handheld devices that have GPS and wireless networking. The next generation of devices are going to be new animals, less easily defined with preconceived ideas descended from static tele-visual media.

Games companies can create an infrastructure for an ad hoc distributed mobile clan. they don't have to mob (but they can), they don't need a base, or a leader, they can cover and survey a large area without line of sight, and they can refer to and contribute to continuously updated collective collaborative maps, territorial definitions and asynchronous (and synchronous) communications channels. Developing what will turn out to be the new tactics (positive and negative) that arise out of these kinds of capabilities reaching street level probably requires using them, being familiar with them, taking them for granted as a starting point. It requires kids. This is not text messaging or email, or word of mouth, its open high bandwidth communications, GPS and powerful processors.

Some PC games already have entire almost accurate 3d models of major cities (ideal for urban gaming when combined with positional information). These new handhelds easily have the processor power and several gigabytes of storage needed to store and render those models.


Soft architecture

We are headed towards a massive shift in the plasticity of the built environment. Location-sensing capabilities allow marking out spaces and paths that can subsequently be found, interpreted and followed electronically.

Consider hybrid objects that can demarcate both in the real world (e.g. using light) and in a virtual representation of the real world (transmitting location). Lights with location awareness, so that a perimeter or a path can be marked out in real space and virtual space simultaneously.

All the elements of architecture, points, lines, and bounded space will be available and perceivable to next generation mobile devices. The cost of transforming a space in software will be considerably less than the equivalent transformation in stone or steel.


Filtering and event triggering

One problem associated with locative media, and with data in general, is how to wade through the mass of information to find what you need.

At the moment people find things mostly through a combination of Google, personal referrals and links.

As data becomes more structured the kinds of filtering that are possible become more interesting.

Social networks, keyword searches, location, and context all become criteria for narrowing down what a user sees.

You can choose to see only what people in your peer group have left, what you are close to, what you are interested in, what matches your keyword search. The intersection of these criteria will be a useful subset of the mass of information available.


Identity

Identity systems are key areas of research with implications for digital currency evolution, reducing spam, increasing trust, exchange and patterns of real world social interaction.

Distributed, decentralised peer to peer identity solutions (see the discussions relating to FOAF and PGP based solutions in the sections below) offer a very different social model to centralised models (e.g. Microsoft's passport and the corporate centric and vaguely federated Liberty Alliance), which view the individual as purely a consumer.

There is a strong commercial incentive in the short term for developing such a distributed system. Whoever defines the most socially acceptable identity model will be in a position to exploit that definition commercially.

The choice of a dominant identity model will have implications for the evolution of money, marketing and new forms of social exchange.

On a broader level. Identity is not just about consumption.

Identity is encoded in clothes, interests, social connections and affiliations.

Mobile devices can encode and project these ideas. A structured, personal, digital identity data layer can be 'digital clothes', mediating between people in the same physical space and triggering social interactions that would not otherwise happen.

Each day you digitally encode your electronic projection of yourself (a bit like an answerphone message).

The opportunity is larger than pure identity. Its about making an identity system that mediates on a higher level than just 'i am who i say i am'. A system that gives information beyond what others see when they look at you. A layer of information visible to others. This layer can be protected using a FOAF based system (see below) so that only those in your extended social network see this identity information.

The goal is to build an identity system with higher, more complex functionality, including features like serendipity augmentation (social network links, interests in common, geographical proximity triggers).

Such a system needs to be peer to peer with the data resident on the device so that two people in the same room can interact even if no connection to a wider network is available.


Social network mapping

There are a number of websites that allow people to collectively map their social networks and take advantage of previously unseen connections. The principle is that you get to see not just your friends, but the friends of your friends (who you might not otherwise be aware of). The list of these sites is long and includes Friendster, Ryze, and Linkedin. These sites all rely on a centralised database.

Since the emergence of XML and the semantic web, it has become a lot easier to maintain decentralised databases. A key early example of this is the FOAF project, founded by the W3C's Dan Brickley. FOAF stands for 'Friend Of A Friend' . The premise is that you maintain a list of friends in a FOAF file on your own website. The FOAF file is a personal, machine readable, list of your friends. It uses a semantic web schema to assert a set of logical relations, mainly of the form 'A knows B' (where B is a functional weblink to B's FOAF file). If B's FOAF file (stored elsewhere) asserts that 'B knows C'. Then a software agent can jump between the two FOAF files in different places on the web and assert that A and C are connected through B, and report back to A that A is connected to C through B.

Essentially FOAF is a digital extension of who you know. Once in place a distributed FOAF social network can form the basis of a more complex system. For example if the geographical locations of A and C are known and they are close to each other, the software agent can deduce that:

person A knows person B

person B knows person C

A is geographically close to C

So that A and C can be notified of their proximity and their connection through mutual friend B

A decentralised social network mapping system incorporating FOAF ideas with encryption schemes like PGP or GPG could be a basis for a powerful and useful decentralised identity and trust system.


Reputation and trust

Aligned to the problem of Identity are the problems of reputation and trust. Once a persons identity can be verified, the next question that arises is whether they can be trusted. This can be made measurable in terms of validation by implication (e.g. they belong to a community that is trusted) or by direct validation (e.g. they are vouched for by other people; either by friends or people that have traded with them without being ripped off).

Trust systems (and theory) already exist and many function well even without being absolutely water tight. The ones in use often refer to some kind of social network mapping system; that is you are validated because other people link favourably to you (e.g. blogrolls and FOAF).


Non-contact communication

Companies like Amazon have recognised the power of implied (rather than directly asserted) social network mapping. When you buy CD's at amazon, they keep track of what you buy or browse through, and suggest alternative products that people who bought the products you are interested in also bought.

If individuals maintained lists of the books and CD's they like on their own sites (or more importantly on their own mobile devices). Then Amazon’s model could become the basis for a kind of decentralized, peer-to-peer, “non-contact communication”.

More importantly this decentralised system could trigger real world events.

Through your mobile device you could meet someone on a long train journey, based on a combination of a mutual friend (FOAF) and an intersection of interests (books in common) and geography (proximity).


Economic exchange

A FOAF based system, aligned to semantically structured lists (of things liked, needed or for sale for example) is the basis for an exchange engine.

If inventories (lists) of all kinds of things are exposed (only to those who fit user defined criteria), and those lists are structured and machine readable (and allow machine inference), then machine mediated exchange systems become possible.

Add location information and host your node in the system on your mobile device.

Allowing, for example,a group of neighbours to expose and pool resources like tools or books.

Up until recently only the money markets met these structured conditions, but now the effects of structuring and automation on markets can be applied peer to peer, and to every kind of exchange. Augmented capitalism.


Ultra wideband

"Ultra Wideband radio is a revolutionary wireless technology that transmits digital data at very high data rates over a wide spectrum of frequency bands using very low power. Within the power limits allowed under current FCC regulations, UWB can not only carry huge amounts of data over a short distance at very low power, but it also has the ability to carry signals through doors and other obstacles that tend to reflect signals at more limited bandwidths and a higher power. In addition to its uses in wireless communications products and applications, UWB can also be used for very high resolution radars and precision (sub-centimeter) location and tracking systems." [From the Time Domain ultra wideband FAQ]


Existing technologies integrated

Integrating existing and emerging technologies into a new way to think about place, social interaction, and exchange is a huge opportunity.

A direct, head on, attempt to make a real device that not only has a feature list that incorporates all the capabilities outlined above, but to make them work together, could result in huge level of social change and offers an unparalleled commercial opportunity

Location aware devices, cam phones, RFID tags, geo-tagged blogs, GPS, WiFi, the semantic web, 3d gaming are a disparate set of buzzwords. Combined together they could alter the very fabric of commercial and civil society.


links

Identity

IDcommons

http://www.idcommons.net

openprivacy.org

The Liberty alliance and Ping ID

http://www.projectliberty.org/

http://www.pingid.com

Trust

advogato trust metrics

http://www.advogato.org/trust-metric.html

newsmonster

ww.newsmonster.org

affero

http://www.affero.org

FOAF

http://rdfweb.org/foaf/

FOAFnaut

http://jibbering.com/foaf/

Semantic web

Dan brickley

http://rdfweb.org/people/danbri/

RDF Primer

http://www.w3.org/TR/rdf-primer/

RDF known ontologies

http://rdfschema.info/2003/known_ontologies

Dave beckett's RDF resource guide

http://www.ilrt.bris.ac.uk/discovery/rdf/resources/

Context computing

Peter J Brown's papers on 'context computing'
www.cs.ukc.ac.uk/people/staff/pjb/pubs.html

Ultrawideband

Time Domain
www.timedomain.com/products/ourtech/faqs.html