• Where 2.0: Enabling the GeoWeb by Mass Market Geotagging

    Johan Peeters

    Spin-off of Philips, geotate.

    Think there’s one thing missing, mass-market tagging of metadata. I want metadata coming from my wife, cat, canary, shoes, everywhere.

    Slide showing cyberspace on the left, real world on the right. Cyberspace shows data, real world is organized by “place and time”

    Need to tag the data.

    What is a geotag “Adding an (instant) notation of place and time to a “real world” user generated or observed event / content.”

    “Place and time becomes the new URL”

    If we look at a typical GPS, it has [multiple parts]. Can take 45 seconds - a minute to work out location when turned on. That’s crap for instant geotagging. “Not a good fit for geotagging”.

    We worked out a new way of using GPS. “Cature & Process Later SwGPS”. We have a radio chip, stream the radio data from the GPS into memory. At some time later we put that unprocessed data through a computer and do the number crunching, mix with some additional information and we get a position fix.

    “What’s the big advantage?” - This radio device can go in any device. Uses 500x less power than traditional GPS. Small cell can last for a year or so.

    2.5 million pictures are geotagged by hand each month. Imagine what would happen if this was automatic? “We will make that happen”.

    We have put the technology in a point and click camera. Just take a photo, no waiting. small radio chip, stores <200ms raw data and then turns off radio. When you upload pictures it uploads the data and calculates the locations.

    [Demo]

    Basically the same speed as uploading photos to a computer. Showing that this is very easy, anyone can use it.

    Examples: GeoWatch, CreditCard, Cameras, Bike computers, laptops and UMPCs.

    Enabling the GeoWeb by Mass Market Geotagging

    Technorati tags: geotag, gps, where, where2.0, where2008

  • Where 2.0: Building a Programmable GPS with BUG

    Peter Semmelhack

    Hands in the air, mainly software developers in this room.

    We’ve been talking a lot about mashups. Mashups is generally software, information, mashing things up that already exist.

    What if as a software engineer you could more control the platform that the data comes from?

    Hardware Mashups

    • Can hardware match the progress we see in software, web services and APIs?
    • Sensors - location, image, environmental…
    • Network communications - sharing
    • Real real-time (instantly requesting information and getting results)

    As a software developer I can look around for what hardware might support me but can find nothing. So might get Make and try to build something but it doesn’t scale. I believe that in 10 years the distinction between hardware and software will be much more blurred.

    We’re building a modular platform. Can combine pieces and parts in a way that is completely modular, one module is a GPS. Want to show you how we’ve taken this further.

    [Demo]

    This is an SDK, I’ll use it as a way to show you how it works. There’s a picture of the base unit with the GPS plugged in. This is based on the “virtual bug”, our emulator. Can put two GPSes on, perhaps for comparison. Important thing is it’s up to you.

    Title is “programmable”, that’s a very important point. This GPS is backed by a powerful full linux computer. I can build a tracking device by snapping a GPS and a modem onto the device.

    One of the most important things about bug is that every module in the system is rendered as a web service. If we look at the GPS, there’s a URL that gives the location. Every module gives a URL. Showing GPS output as XML web service. As we add more modules we can mash up more data as it becomes available.

    81 different modules available, all kinds of sensors, IO, etc. All of them follow the same metaphor.

    “Software + Hardware mashups will release new wave of innovation”

    Building a Programmable GPS with BUG

    Technorati tags: hardware, gps, bug, buglabs, where, where2.0, where2008

  • Where 2.0: Monetizing Maps & Mashups

    Greg Sterling

    Tiffany Chester: Specialises in geotargetting advertising Ian White: urban mapping, license data Jay Brice: skyhook wireless: location engine, iphone wifi positioning Steve Coast: Cloudmade: Provides services on top of OSM similar to Redhat Walt Doyle: Where.com: Publisher of Where, number one LBS platform

    This panel is about addressing the practical questions of who’s paying for this?

    Online and mobile, there’s a lot of money out there but it seems difficult online to make use of the local opportunity.

    JB: Agrees, a lot of money can go into making the data that’s needed, but it’s very difficult to collect money use of this.

    GS: Why did you shut down Mappam?

    SC: A lot harder than it looked. Technology was easy, becoming an advertising firm was hard. Cloudmade was being started.

    TC: Seeing a lot of success in the local market. Advertisers have struggled to target audience traditionally but local advertisers find out map-base technology appealing. Often with realtors, market brokers, banks to be able to fine tune their advertising messages to audiences online. The map is the basis for them to be able to do that.

    IW: This is controversial point number one. As an add is more and more granular, the efficiency is, the cost to serve that goes up. The more hyper-local you are the less value there will be.

    GS: What do the people in these areas do to support their apps?

    SC: Wait for google to switch on adsense for maps. It’s going to happen, but when? We tried this with Mappam before google switched it on. It’s an open question as to why they haven’t.

    GS: So you say this will be positive. But don’t you think adsense makes limited amounts of money?

    GS: Lets talk about local response rates.

    JB: When we started trying doing more local targetting and being able to insert a lat/lon. We had to flood adsense with a lot of local information to get any information from them at all. We had a click-through rate of 7/8/12% as if you can position ads around a relevant map you get a higher click-through. But there’s a problem that there isn’t enough ads to serve, requires a huge inventory to serve at such a granular level.

    IW: the keyword conspiracy. If you have these highly targetted ads, you’ll find that a keyword doesn’t have large volume.

    WD: The irony is that we’re getting to the point where there’s a larger audience but not a larger yield.

    GS: Brands and folks tied to national entities. Most of the dollars are tied to local. Won’t some of that money go into online?

    WD: It’s happening more, we’re seeing experimentation. Burger king saw success with driving directions.

    GS: So Burger King pays you for each lookup?

    WD: Sortof

    JB: We got contact by Kraft foods, most visited food website in the world. They’re building a location based menu creator for the ipod “mac&cheese; finder”. Will give you directions to the shopping. That’s a revenue model. Branded/sponsored things.

    TC: We’ve had a lot of success with franchising markets. There’s the head office and the individual business owners who have a very small draw region and it comes down to them. They might not have an online advertising idea until they came to us. They need a method where they can modify the creative for their own region.

    JB: The most visited part of many sites is the store finder.

    GS: Mobile: Can’t really charge consumers for online stuff. In mobile you generally can. For how long? Is that a viable model?

    WD: We’ll see all sorts of model, iPhone SDK interesting as it allows both models, free and pay-for. We’ve learned that the monetisation is quite similar whether it’s subscription or advertising. The increase in overall consumption made up… [something]. Our approach is very much to go free. Seeing dollars flowing from marketing services too.

    GS: Phones are a much smaller screen, don’t you compound the problems of the internet with advertising on such a smaller screen.

    WD: It’s certainly harder but there absence of limitations causes lack of innovation. Companies like greystripe put up interstitials and you have no choice but to engage with the adverts. Mobile ads might be more interruptive but be more effective and be things that you can’t do on the web.

    SC: People hated ads on maps more than they hated normal maps. Don’t really want MacDonalds logos on maps, especially if they obscure features.

    TC: Our ads come under a lot higher criticism and that’s because they’re incorporated as part of the content. We’re really focusing on making sure ads are highly relevant. We’ve had user studies done but the response is very good so long as the ad is targetted they feel it’s part of the experience.

    GS: We’ve had cost-per-action forever is it viable for mobile?

    JB: Tying two sides together means you get paid from multiple sources. Advertising and then routing people to the location for instance.

    IW: As a data provider, value is clearer. On mobile it can be harder to make clear what we’re charging for. If we can give the consumer something that’s more enriching…

    IW: Relative to location awareness: more phones will be shipping with GPS, yay, sucks for some things but great for others. On a repeating basis, what type of people subscribe to location aware services? We’ve had location aware for years: vindigo: when are we going to get to critical threshold, will we, will iPhone be game changer?

    GS: iPhone is absolute game changer. Windows Mobile, Nokia, everyone is trying to develop. Mobile usage is increasing. Volume is going to be there just how fragmented is market? iPhone appears to have built market as response to apps built is better.

    Q: I work for the history channel. We’re trying to find the model for putting content in good LBS devices?

    WD: I’m a fan. We’ve found that vertically segmented content you need to build 1-1 relationship with your consumers. It has to do a lot with your existing advertisers and with the applications that will be serving the content. Reference points to location are enormous and you have a lot of content so there’s a lot you can do.

    GS: CPM and something else

    JB: We’re based in Boston and have done great historical data with another company. They were looking for a business model to use that data but the cost of acquisition was high. Monetising might be a problem.

    Q: Subsiding services with traditional ad networks/sales channels?

    TC: We’ve had a lot of newspaper groups approach us.

    GS: You need scale though yeah?

    TC: Sure.

    IW: Direct marketing business has a lot of this money. I’m convinced that LBS will become a kind of direct marketing. It’s simply another channel.

    JB: We’ve seen the reverse of this. The company that’s the largest printer as they generate advertising and coupons. Field sales force of 800 people selling paper-base local ads. We may see a huge inventory of advertising come online through this organisation. They’re already doing location targeting and it may feed to online.

    Q: I’m calling Bullshit on saying it’s a good thing when Google puts ads on maps. Ad subsidies will make it so that google android will be the only phone.

    Monetizing Maps & Mashups

    Technorati tags: monetizing, map, where, where2.0, where2008

  • Where 2.0: The Business Case for Simulation, Gaming & Virtual Worlds

    Denis Browne

    Intro to Matrix…

    SAP has been around for 35 years. Taking you through what we’ve done for a customer Implenia. 3D rendering of a building so that customers can immerse themselves in an environment. Better than styrofoam models they’ve been using before.

    Can look at how buildings will function when they’re brought online, safety issues, etc.

    Showing a model of a house.

    Have inserted sensors into the model. Can control it from the virtual environment. Turning lights on and off on the virtual model and it’s reflected in the real model house.

    Blocks the light with his hand to simulate the bulb blowing. Virtual model spots this and creates a ticket to request that the bulb is request. Client can then go in and see how long it’s going to take to be replaced and change the priority.

    Denis sometimes forgets his keys… with the Unity system and this environment, Sebastien (his helper) can open the door for him).

    Also have asset tracking.

    More and more companies are interested in managing their workforce.

    Opening and closing the door on the model makes the door open on the virtual model, it’s very accurate, good for security monitoring.

    Can optimize the building, find the carbon footprint, what’s turned on and off, what systems are using energy. Can do all this and remotely turn things off and on. Implenia get paid for reducing costs for their clients and use these systems to do this.

    Implenia manages entire metropolitan areas in some cases. Every building they build today has sensor networks embedded.

    Everything is going to be tracked and traced.

    Showing a picture of the Internet. In 5 years, the sensor network will be hugely bigger. “A mere flea next to this picture”. The sensor network will run the world in the future. Should make the world a better place to live.

    Might be odd that old stodgy dinosaur SAP is at Where2.0. Hoping that this presentation will make it clear what we can do in the location based business.

    The Business Case for Simulation, Gaming & Virtual Worlds

    Technorati tags: sap, sensor, where, where2.0, where2008

  • Where 2.0: Ride the Fire Eagle: Open Location for All

    Tom Coates

    “Hello Everyone”

    Yahoo! Brickhouse, environment for startup type projects in Yahoo! Small teams working on projects that are “bets for the future”. Ones that might be a big deal in 18 months time.

    Fire Eagle, very much a bet on the future. Location tracking services still haven’t gone as mainstream. Fire Eagle is trying to solve some of the small sticking points.

    Launch FE 2 months ago, response has been amazing.

    “Thank You!”

    FE helps people:

    • Share their location online
    • Control their data & privacy
    • Easily build location services

    Most location-aware applications have very tight coupling between finding the location and using the location. Both the problems are pretty hard. Because it’s so hard finding the location, a lot of uses for the location don’t get made. One good model would be one person finds the location and many people use that, but that’s a bad model too as it only gives you one source. FE is better as anyone can make an application to find your location, then anyone can make an application to use your location.

    Yahoo! Internet Location Platform

    Language that we can use to understand location. E.g. when we all talk about San Francisco, we’re all talking about the same San Francisco.

    Can update location by entering text on their website, or by providing a string, some cell tower IDs, lat/lon and other methods to the API.

    Dopplr - “A social network for frequent travellers”

    Dopplr can integrate with Fire Eagle. Authorising requires going to FE site, confirm, then go back to Dopplr. Very simple. FE just manages the location stuff.

    OAuth makes this work. It does a lot of security stuff. Dopplr doesn’t necessarily know who I am as far as FE is concerned. Stops people on different services from being able to work out who you are by comparing tokens.

    Privacy, we have an idea of how much you want to share. Information can come in but not go out. You can turn update off completely. You can also purge your location information out completely. There’s a new contract that’s being created between websites and users to allow you control of your data. By default we will send you a monthly alert that if you don’t respond to, it will disable your account, so you can’t forget you’re being tracked.

    Navizon - positioning system using GPS, WiFi and Phone positioning

    Loki - from Skyhook, MyLoki can broadcast your location to FE.

    ZoneTag - runs on Nokia S60 phones to geotag photos but now also updates FE

    Firebot - allows you to update location using a twitter direct message

    Brightkite - can check into a place then submit images, etc. about the palce, updates FE.

    Plazes - goes by a wifi hotspot and records location, asks you to enter information or uses information if there is some, updates FE.

    spot - combination GPS unit and satellite phone. Doesn’t update FE, we would like it to.

    Isaac Daniel GPS & Satellite phone shoes - awesome, nothing to do with FE, Tom wants one.

    Using location…

    Wikinear - designed for mobile phones, shows 5 nerest wikipedia articles.

    Lightpole - helps find restaurants, bars, many POIs Lets you set location but also gets your location from FE.

    Outside.in - news stories

    Fireball - Conferences

    Fire Widgets - local weather forecast

    Friends and Family widgets - show where you are

    Proximizer - show how close a friend is

    Movable type - plugin

    Twitter

    All user generated content

    last.fm - doesn’t integrate with FE, but would be great to find out where you listen to certain songs.

    Friends on Fire - facebook app

    Ride the Fire Eagle: Open Location for All

    Technorati tags: fireeagle, where, where2.0, where2008

  • Where 2.0: Convergence of Architectural and Engineering Design and Location Technology

    Geoff Zeiss

    Talking about the construction industry, the gaming industry, geo-spatial. There’s things happening in these sectors that are important. Construction industry is 40% GDP, worldwide 2-4 trillion dollars. One challenge is global warming.

    71% of electricity consumption is involved in buildings. LEED certification, green building councils. 14,000 projects that are LEED certified. Carbon footprint, natural lighting, heat sinks, etc. This has really become a worldwide effort. That’s a big area.

    Water company in Australia had been mandated to be carbon neutral. First step, replace headquarters.

    Requires massive renovation of reconstruction of facilities.

    State of infrastructure in North America. “Report Care for American Infrastructure”, every 2 years or so. 2003 roads and highways got d+, 2005 got a D.

    Another example of challenges in the construction industry. Impacts a lot of people’s lives.

    Bridge collapses, gas explosions, etc. People don’t like this, governments do something about it.

    “Who’s going to do all this stuff” “In the US alone., an additional one million workers will be needed to address these challenges in the construction industry.”

    Not enough workers to do this.

    “And productivity is lagging”

    Graph showing industrial productivity (non-farm) is going up, but construction productivity is going down.

    “Geospatial enabling”

    80% of IT can benefit from location intelligence.

    “Building Information Modeling”

    Good for accounting, also different subsystems can be developed by different people so makes sure they don’t encounter problems.

    “3D Visualization”

    Use data from when building was designed and combine with underground utility data and realistic visualisation, can really simulate entire cities.

    Have many benefits

    Urban reconstruction

    • Right to light
    • Noise abatement
    • View protecion
    • others…

    3D visualization of underground network of a city

    Allows right to light to be found, based on geo data and gaming technology. Can identify issues with shading. Good for LEED too.

    Noise abatement. “Who’s going to hear a soccer game?” for example.

    This is what you can do with a full simulation of an urban environment. Showing very good simulation of Seattle. Elevated freeway. If there’s an earthquake, this thing will fall down. Choosing to go underground or elevated. Contracted company to work out how much it would cost. Wanted citizen involvement but instead created 3D model videos. Showing what it would look like to remove the elevated road and put it underground.

    Summary Real challenges IT developing techs convergence Enables intelligent simulation of urban environments

    Convergence of Architectural and Engineering Design and Location Technology

    Technorati tags: construction, where, where2.0, where2008

  • Where 2.0: Katamari Damacy

    Greg Sadetsky

    CEO of Poly9, startup based in quebec city. Been doing custom mapping applications for 3 years. Flagship product is FreeEarth. 3D web globe. Round ball, you can see earth, can zoom, tilt, see terrain and high resolution imagery.

    FreeEarth lives on the web, integral part of the web. Makes a huge difference. Being part of the browser means you don’t have to download, install anything. The version of the .net framework doesn’t matter. Uses Flex/Flash. If YouTube works, FreeEarth works. We should not be asking users to install/download stuff.

    Entertaining uses of FreeEarth:

    • Skype Nomad using Freeearth.
    • Sanyo Japan - users can leave messages of well-being to the globe.
    • Wild Sanctuary - one year ago, Where 2.0 2007, web design company in san francisco were introducing interactive mapping application of sound-scapes for wild sanctuary. You could click on a region and hear wht it sounded like. Although they had no problem with the 3D version with google maps, they had difficulty with the google earth version because people had to download and install it or couldn’t install due to corporate security products. 30earth realised that using google earth loses your users to google earth, no longer on the site. They cam to us and we helped them in 2 days for a BBC interview.

    We’ve seen why web deployment means so much. We put out a press release yesterday about a joint venture with another canadian company to launch “GeoAlert”. System to save lives. GeoAlert is all about saving lives.

    Every time an oil company wants to pump oil from a well, they have to go to the government and ask permission. Energy commission looks them in the eye and says “are you sure you can call everyone around the well if something happens?” Energy board looks them back in the eye and says “no problem”, which is, in effect, false. The oil companies do 3 things, go out to all residents with paper forms, take contact information, children, medical information. Put the info into excel file, database. Then they really hope nothing happens.

    There’s no organized way of reaching people, tracking who didn’t get the information, have to lookup in the excel file. No audit trail. Very scarey. GeoAlert does two things very well. 1st lets the commission click on a 3D map, FreeEarth map, to locate the incident and 2nd thing is call everyone around that well. GeoAlert will take into account the wind and call people directly in the gas plume first. Isn’t that amazing? … (one guy says “Yes”)

    mapmkr - Launching Summer ‘08

    “How do I see my data, on a map?” Do we have to go to see specialists just to write text documents? No, so why do we have to for maps? Making a service to allow this. Hoping that people at Where 2.0 will use this.

    Katamari Damacy

    Technorati tags: geoalert, mapmakr, map, poly9, where, where2.0, where2008

  • Where 2.0: Modeling Crowd Behavior

    Paul M. Torrens

    Why would you model crowds?

    Important factor of modern life. Some of the most important tipping points. Evacuation, emergencies. Understanding how crowds work is important for public health. New forms of mobs.

    “We [really] do not know as much about crowds as we would like to know.”

    “Simultation can serve as an artificial laboratory for experimentation in silico

    “I build complex systems of behaviourally-founded agents, endowed….

    “What does this have to do with geospatial technology?”

    Business opportunities for Geographic Information Technologies”

    Way to do it is convoluted, build models, AI, rendering, outputting analysis, statistics, GIS, etc.

    Have technology to model individual people, often don’t have the data. Generate synthetic populations, downscaling larger data sources to individual level. Gives characteristics to our agents. Customer loyalty cards, GPS, cell phones.

    Physical modelling and rendering

    “Small-scale geography from motion capture and motion editing”

    Record spatial and temporal information in studios, 100 frame per second, spatially in order of a few cm. Graph of movement of skeleton through time.

    Physical simulation, doing bad things to the modelled skeletons.

    “Behaviour is simulated (!= scripted) using computable brains”

    Using a turing machine, socio-communicative emotional agent based model, wrapped in geographic information. Given GIS functionality. Geosimulation.

    Taking basic model and wrapping in geography. Agents can “see”, can deploy mental map, plan past, parse to waypoints, navigate to goals. Identify what they’re interested in, they can steer, locomote. Use motion capture date.

    Should be able to drop them in a city and they’ll get going.

    Data relies on GIS, video of a 3D model of a city. Space-time signatures, space-time patterns. Given someone’s usual geo behaviour can figure out all their possible directions.

    Social network monitoring. Monitored children on a campus every day for three years, watched how they formed groups and how they play. High performance computing. Binary space partitions.

    “Applying this to real world issues”

    “Quotidian crowd dynamics”

    Screen scenes. Pick one person and follow them.

    Showing video, model of people. Old people, young people, drunk people. Each behaving autonomously. Positive feedback, negative feedback, all sorts.

    “Extraordinary scenarios”

    Building evacuations, bottlenecks.

    Run through space-time GIS, look at egress behaviour, if they run more people get hurt.

    Urban panic, out of buildings into urban environment, look at how they evacuate. Showing video people jogging.

    More diagrams and graphs.

    Dynamic density map.

    Riotous crowds. Standard riot model and wrapping with geo-spatial exo-skeletons. Generates a riot. Can test for clustering to see if people with similar motives and emotions are grouping. Small scale riot behaviour turns to large scale very easily. Devious behaviour, rioter see police and pretend they’re not rioting, run away when chased. Inserting police who are told not to arrest people will calm the crowd but not completely.

    Crowd response to invasion of non-native stuff. Showing Cloverfield.

    Small scale epidemiology of influenza.

    Zombies.

    Putting crowds into digital environment, what happens to location based services.

    geosimulation.org

    Modeling Crowd Behavior

    Technorati tags: geosimulation, simulatation, modelling, crowds, where, where2.0, where2008

  • Where 2.0: Best Practices for Location-based Services

    Sam Altman

    Last year concentrated on our social network mobile applications. Can share your location, get alerts when you’re near people, geotag flickr photos, etc.

    Idea is to use location.

    As we developed service we discovered why live location apps suck so far.

    Talking today about what we’ve observed is the problem and how we’re solving it.

    When we hear about people doing well with location tracking we cheer each other. Loads of devices and apps coming out for location.

    Location Challenges Today: What we want

    • Inexpensive
    • Low friction access
    • Relevant accuracy
    • Strong privacy
    • Availability and Reliability

    Structural Challenges

    • Cost
    • Availability
    • Privacy

    Doing 300 pings a day of user’s location, costs a few cents per go, we need to reduce this 100 times.

    Loopt Core Location Platform

    • Mobile Client
    • Location Privacy Manager
    • Location Server
    • Location Access Manager

    Mobile Client

    Need a client device to let you do location in any number of ways. Can use SMS, GPRS various ways to request your location. Clients only around for certain classes of devices. Runs at lowest level.

    Privacy Manager

    Consider privacy a lot. Do a lot to ensure your location doesn’t go to everyone. Ways to manage this: Global on/off, app by app, restrict by time, day, location. Set/obscure location. Can set on mobile or by web, for accounts, sub-accounts, many options.

    Mobile Location Platform

    Trying to get to per-user per-month models, getting as close to free as possible. Ads.

    Access Manager

    Current Location APIs are very bad. Trying to replicate the quality of Apple’s API. Only specify when the phone leaves an area. Only access location using a web app, or a wap app, choices for building apps, can access in number of ways.

    Believe Loopt have the best location platform the world has ever seen.

    Solving the problems with:

    • Structural Challenges
      • per user per month model
    • Availability
      • open APIs, easy to access
    • Privacy
      • end user control and management, device and 3rd party app integration

    Best Practices for Location-based Services

    Technorati tags: loopt, tracking, privacy, where, where2.0, where2008

  • Where 2.0 - Live Blogging

    As you’ve probably noticed I’ve decided to have a go at live blogging the talks at Where 2.0. These notes are taken live so there’ll be lots of spelling mistakes and possible errors. I’ll try to go back and clean them up later but hopefully they’ll be useful for people who can’t be there.

    Break now, I really need one!

    Technorati tags: poly9, where, where2.0, where2008

subscribe via RSS or via JSON Feed