Pat McDevitt, TeleAtlas. Navigating the Future: Mapping in the Long Tail

All navigation is local.  Where is RELATIVE.  It seems the big companies don't map smaller, localized "where".  Why not?  They're hit based, it's too expensive to map the niches.  Pat explained it as hit based vs. niche based mapping.  Who's who in the mapping playground:

• Hit based: TeleAtlas, NavTech, UK Ordnance Survey
• Niche Based: Platial, Open StreetMap, Loopt

The niche based mapping effort is being done by...you guessed it...YOU.  How is this happening?  It's all part of the map creation long tail.

The Long Tail of Map Creation

1. Create your own content. 
2. Compile verify and normalize 3rd party data.
3. Develop enabling tools and do more of #2. 
4. Filter user and community content.

The enabling and compiling steps allow micro-interest mapping.  That means my 70 year old Dad could add content that is relevant to him thanks to steps 2 and 3.  So, where is long tail going?

In the long tail, filtering technologies will be very important in the future.  This brought up the whole Paleogeography v. Neogeography debate (man...I hate those two words...but that's another topic for another post). Really, the debate is about collaboration or competition in creating map data/content.  I think it should be collaborative and so did Pat.  He showed a great quadrant figure to describe how can all work together (read as: Can't we all just get along?).  In Pat's words, Paleo should focus on protocol based visible and verifiable data.  Neo focuses on low protocol low visible/verifiable.  Low protocol/high verifiable/visible data will be created through customer feedback to paleo companies.  If all of these areas come together, paleo and neo can come together and produce really great, unique content and geography for a very long tail in the map creation playground.  I'm not sure I entirely agree with this assessment, but it's a good start. 

POST SCRIPT: I really wanted to do a post on Jeremy Bartley's talk on ESRI's ArcGIS 9.3 but he totally got robbed.  He got 5 minutes and barely got a demo in. He talked quickly (very quickly) about ESRI's REST and JavaScript API's, showed a quick demo of Google Maps and VE integrations....and he was off the stage.  I felt really bad for him.  If you go the ESRI User Conference this year or the Dev Summit next year, connect with Jeremy.  He's a great guy and has some awesome ideas.

Vincent Tao, Director of Microsoft's Live and Local Search and Virtual Earth gave an absolutely awesome presentation this afternoon.  First he spoke about Where is the "where?".  Essentially, he see's location as nothing but an index to information.  As such, he said that we need to move from organizing spatial information to organizing information spatially.  Next, he discussed how people look for information indexed by location?  Here's his lowdown:

First look: Entry points to location data?  (Or where is the "where?")  Specifically, he was talking about location based services vs. location powered services.  So, what does that mean and what's the difference: People come to social network sites for socializing, not for location information.  The location component is value-added.  In general, most location services are powering today's hot social sites.  This is not location based services...this is powered by location services.  Big differences...and most people are using location powered services, not location based services.  So, the where is not so much at "mapping" sites as it is at sites we use every day.

Devices (what do people use to get their location data): PC queries at 71% represent the largest slice of pie.  Cell phones are 5% of the pie, phone data (i.e. mobile browsers) is at 2% but with a growth rate of a projected 71% in the next year, in vehicle navigation is at 1% with a 20% projected growth rate next year, and the remainder is still from print (you remember...those paper map thingies?).

Next, Vincent moved to show all of the way cool stuff coming (or just released) in Virtual Earth.  Microsoft views VE as an enabling platform.  As Vincent said, "We don't want one earth...we want you to create an ecosystem of millions of earths."  So here's the coolness of VE on it's way or here already: 

• VE for Mobile Search: Image mapping + real-time traffic + new voice search (recently released and very cool...I've been using it and it works great).
• VE for Messenger: Not enabled in U.S. yet, but coming
• VE add-in for Outlook: Can tie in to appointments...provide auto-reminders for meetings based on distance from meeting location and real-time traffic updates.  "Better leave for your meeting now...major traffic on I-5!"
• VE and SQL Spatial coming in June.  We're all anxiously awaiting this one!
• 500 world wide 3D cities...adding 200 more in the next year!
• Automated image processing to remove moving objects from VE tiles.
• Automated 3D city model generation: Major Secret Sauce here, but they're cranking out cities very quickly.
• V2 Cities: Upgraded textures and quality.  Much closer to reality with trees and everything...automatically!  Some of the demos he showed were almost indistinguishable from real photos.
• And, the coolest of the cool...VE and crowd sourcing: Your own photo experiences....we're talking 3D vertical photo rectification here folks!!!  Yeah...slightly impressive.  Can't wait for this to come out.
• Beyond your photos:  Dynamic real-time shadows in 3D cities.  It's done and it's coming.

Years ago, I worked on a project that attempted to integrate GIS data into the heads up displays of U.S. Army helicopter pilots.  The technology has come a long way.  Tom Chruchill from EarthData demoed some killer apps for augmented reality for the Denver Police Department helicopter pilots. We're talked fully annotated, vector-feature enhanced heads up displays.  He's also putting together some killer geobrowsers.  Check out his stuff at http://www.earthscape.com/.

Johann showed some very cool hardware from GeoTate that will automate the geotagging of photos via small GPS devices in cameras...they fit in phones, watches, credit cards and other small stuff too.  They have some simple hardware components for hardware manufacturers to embed in their  solutions.  Definitely keep an eye on these guys...this is going places.  Johann predicts that with GPS enabled cameras, geotagging in Flickr can grow from the current 6.5 million manually geotagged photos to over 50 million automatically geotagged photos...that's huge!!!

What FireEagle does:

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

FireEagle works like a brokerage. It bridges the people getting location and the people using location.  It allows users to control how much information is stored and shared.  Tom demoed how Dopplr integrates with FireEagle and controls privacy and access to user entered data.  Essentially, they are using OAuth to manage security for FireEagle.  Here are some cool things people have been using FireEagle for:

• Navizon combines GPS, WiFi and phone positioning
• Loki combines GPS-like location, local search and one-button access to location-based content
• ZoneTag take a picture with you phone and instantly geotag it and upload to Flickr
• BrightKite allows you check into a place an upload text or photos about that place.  Location based social networking.
• Plazes recognizes where you are on a laptop or phone and provides contextual info about where you are.  Create and share activities tagged with location.