Comment faire, donc, pour réunir le monde "en ligne" et le monde "en dur" ? Quelle interface, quelle passerelle, permettra de passer de l'un à l'autre ? La réponse semble évidente, il faut introduire dans le monde en ligne la composante la plus importante du monde matériel, l'espace géographique bien sûr ! On pourrait donc avoir une interface ressemblant à une carte, telle Google Maps, ou à un monde 3D, comme Second Life mais avec le réel comme matière première plutôt que l'imaginaire. Ajoutons la géolocalisation par GPS (ou autre) et imaginons des solutions pour faire des choses aussi évidentes que d'afficher le site web correspondant à l'endroit où nous sommes, ou encore, moins évident mais tout aussi pratique, consulter le plat du jour des restaurants alentours. Les communicateurs apporteront donc, en plus du mode de navigation classique, un moyen d'accéder à l'information en partant d'une représentation physique du monde, réconciliant peut-être ainsi le réel et le virtuel.

Mais il existe une autre composante essentielle qu'il ne faudrait surtout pas négliger, c'est le temps. Comme dans le monde réel, rien n'est permanent sur le web, les choses changent, tout peut se transformer à chaque instant. Les communicateurs proposeront donc des mécanismes de navigation dans le temps afin de répondre à des questions du type de "quel était donc ce restaurant à la place de cette boulangerie ?" ou "quelle pièce ce théâtre présentera-t-il cet été ?" ou bien encore "cette photo a été retouchée, quelle était donc la version précédente ?".

Je dois maintenant aborder des aspects un peu plus techniques afin de montrer brièvement comment tout cela pourra fonctionner. Où les données seront-elles stockées ? Que deviendront les fournisseurs d'accès et les acteurs du web en général ? Pour répondre à la première question, je dirais presque une évidence, les données des gens devraient pouvoir être stockées par les gens eux-mêmes. Les communicateurs seront donc en mesure de stocker des données et de les diffuser à d'autres communicateurs, sans aucun intermédiaire, à la manière d'une architecture P2P. Cela dit, pour des raisons pratiques, on confiera souvent le stockage des données à un fournisseur tiers, mais ce dernier n'aura rien à voir avec un YouTube ou un Flickr, ce sera plutôt une sorte de disque dur distant, un "datastore" personnel, stockant indifféremment tous les types de données. Ce service pourra être proposé par les fournisseurs d'accès ou par tout autre acteur. D'autre part, on aura la possibilité de déplacer ses données à tout moment d'un fournisseur à l'autre, et cela sans "casser" les éventuels liens qui pointeraient vers ces données. À cet effet, une distinction sera faite entre l'adresse physique d'une donnée et son adresse logique qui servira de référence. Les adresses logiques pouvant quant à elles être données par les fournisseurs d'identités numériques, les seuls acteurs auxquels on sera véritablement attaché puisqu'en changer reviendrait à changer d'identité. Mais là encore, on pourra être totalement autonome si on le souhaite en hébergeant son propre serveur d'identité.

Voilà pour le stockage des données mais il faut aussi se demander qui prendra en charge les éventuels traitements réalisées sur lesdites données ? Car il faut bien des traitements pour créer, modifier et faire toutes sortes de choses intéressantes avec les données. La solution envisagée, une fois n'est pas coutume, offre un maximum de liberté : les traitements pourront se faire partout. À la fois localement, dans nos communicateurs faisant tourner toutes sortes d'applications, mais aussi à distance, les traitements pouvant alors être pris en charge par des services tiers auxquels on aura donné des droits d'accès à nos données.

Tout cela commence à prendre forme mais il manque une dernière pierre à notre édifice. On peut voir sur le web un certain nombre de sites qui font des choses intéressantes en agrégeant des quantités plus ou moins importantes de données. C'est le cas des moteurs de recherche bien sûr, mais il y a aussi tous les sites, tels que YouTube, qui offrent des fonctionnalités communautaires en s'appuyant sur les données qu'ils nous proposent d'héberger. On peut alors se demander ce que deviendront tous ces sites si le stockage des données se fait directement dans les communicateurs ou dans une myriade de "datastores" indépendants ? C'est ici que les datahubs entrent en scène ! Pour réaliser un moteur de recherche ou un YouTube, il suffira de mettre en place un datahub afin d'agréger toutes les informations qui nous intéressent. Peu importe où les données sont stockées, si elles sont publiques elles se propageront à travers les datahubs, et tout le monde pourra jouer avec !

C'est ainsi que se termine cette première série d'articles consacrés aux communicateurs. J'ai essayé d'en tracer les grandes lignes, et cela va sans dire, il y a encore beaucoup à explorer, à découvrir et à inventer ! Merci de m'avoir lu jusque là, n'hésitez pas à me contacter pour toute remarque, suggestion ou proposition. La porte est grande ouverte et il y a encore tout à faire !

In her talk "Just-in-Time Info" Pattie Maes, Associate Professor, MIT, Program in Media Arts and Sciences gave a fascinating plenary talk at the SSP meeting yesterday. She has challenged her students to develop prototypes of how humans can be fed information about the world around them as they go through through the everday lives. Many of the techniques involved wearable sensors tied to cell phones on the look out for infrared beacons or RFID chips. The sensors can then be tied via the cell phone to the millions of databases on the Internet to then feed back audio or visual cues to the user. Pattie's students have already prototyped concepts for locating books in a library, gathering product information in a supermarket, photography automatically tied to geolocation, even possibilities about finding people with common interests as you walk around the street.

Words cannot really best describe the mind boggling impact such combinations of technology might have on the world. Check out the MIT class web site here (especially the Vision section). You won't be disappointed.

I've always been impressed by the ability of Google Analytics to accurately pin-point my visitors right down to particular areas of a city. But how accurate is this information? Certainly in my own case it's very accurate, usually attributing my visit to my district of town or the neighbouring one.

So how accurate is whatismyipaddress.com for you?

Check it out and leave a comment with your city/country and how accurate you feel they pin-pointed you.

I'll start the ball rolling. Ixelles, Brussels, Belgium. About 1 mile out.

By the way, here's the definition on Geoblocking

We have detailed geolocation data for 83 countries live and now almost all town level data is returned in under one second. The state of California is the laggard at just over 1.5 seconds. This removes a huge adoption obstacle for us as geolocating videos, member profiles, or real estate shows had often previously involved a 15-20 second wait - a virtual eternity in internet time but we had bigger programming fish to fry at the time.

We'll continue to focus on making performance improvements in a number of different areas over the coming days and weeks.  Go ahead and give it a try either while uploading a video (http://upload.vidlisting.com) or geolocating your member profile, videos, and real estate shows on our social networking site (http://catalog.vidlisting.com)

There are now geodata for 79 83 total countries live at http://catalog.vidlisting.com.  We welcome any new members from the following countries (or those that have media from these countries):

- Burkina Faso
- Gabon
- Gambia
- Georgia
- Germany
- Ghana
- Gibralter
- Grenada
- Greece
- Greenland
- Marshall Islands
- Palau
- Papua New Guinea
- Philippines
- Qatar
- Romania
- Sao Tome and Principe
- Russia
- Rwanda
- Saint Vincent and the Grenadines
- Seychelles
- South Africa
- Senegal
- St Kitts ad Nevis
- Tanzania
- Togo
- Tonga
- Tunisa
- Turkey
- Turkmenistan
- Tuvalu
- Uganda
- Ukraine
- Uzbekistan
- Venezuela

More coming soon....

20 May UPDATE: Added 4 more countries last night (British Virgin Islands, Slovenia, Trinidad and Tobago, and Vietnam)

We'll be adding China and India later this week.

We currently have the following countries in our database and activated so that members, videos, and properties can be properly geolocated:

- Antigua and Barbuda
- Afghanistan
- Algeria
- Azerbaijan
- Albania
- Armenia
- Andorra
- Angola
- Argentina
- Australia
- Austria
- Bahrain
- Barbados
- Botswana
- Bermuda
- Belgium
- Belize
- Benin
- Belarus
- Soloman Islands
- Brazil
- Bahamas
- Bhutan
- Bulgaria
- Canada
- Chile
- Colombia
- Costa Rica
- Dominican Republic
- Ecuador
- El Salvador
- France
- Guatemala
- Japan
- Mexico
- Nicaragua
- New Zealand
- Paraguay
- Peru
- Panama
- Portugal
- UK
- USA (all 50 states + Puerto Rico)
- Uruguay

We'll be adding an additional 20-30 countries over the next week or so. If you live in or have video content from any of the above countries, feel free to join any of our social networking platform partners such as http://catalog.vidlisting.com.

You'll start to see some of the things that we are doing in the geolocation space over the week or so as well.

This Euston Station luggage trolley is a bit lost....

this is my first zonetag pic. It uses the geolocation info from the n95 GPS. main problem is that the GPS takes over a minute to get a lock on the satellite and is very power hungry. Also doesn't work that well unless you are outside.

It may turn out that GPS is overkill for easy geo location. cell or even wifi trig are ample for most situations. (check out google mobile maps )

I'm looking forward to seeing what fireeagle can do, it's a development of zonetag in private alpha right now, just waiting for my invite.....

We've still have a long way to go with the developer API - it is neither complete nor "customer ready". However, we decided to front load some parts of the programming interface and put together a live commercial example.  The live working example demonstrates some of the types of things that developers will be able to do with our upcoming developer API for video that ties directly into the ForSaleByLocals multimedia platform as well as give us a chance to identify and fix issues. 

Here is a sneak peek at some of the capabilities of the programming interfaces:

Context Sensitive Content Aggregation: We've been wrapping certain contextual data around multimedia content for some time now.  This has been limited to aggregating content by language and geolocation (including latitude and longitude in order to span political boundaries). Now, we are able to define additional contexts and use our fledgling developer API to access only content that matches the desired context.

Organization of Content:  Once the right content is retrieved, we are working on being able to use the API to organize related bits of content into "interfacelets" which then can be included like a widget into pages.  We still have a lot of work to do here.

We now have a live commercial example of the ability of the ForSaleByLocals multimedia and video engine to serve up contextual content. The real estate site vidlisting.com is using the developer API to highlight various videos for a huge real estate show in Spain, SIMA 2008. They are using the language and context capabilities to deliver and organize SIMA related video content on a special page (http://vidlisting.com/sima2008/indexsima.asp) which is also going to be their home page during the days of the event - 8 through 12 April 2008. 

We'll continue to discuss how context can help make content networks more social and effective withi a business context.

Let us know what you think...

Communities As Abstractions of Geolocation

Hyperlocal by definition has a dependency on geolocation. Every location on earth is part of a myriad hierarchy of geolocation abstraction that are based on tangible or intangible boundaries - the most often thought about abstractions are neighborhoods, counties, states, etc. These abstractions often form the sense of formal and informal community for many people - what one person considers "community" may different greatly from what others consider.

Abstractions as Currency For Communicating Useful Information

In many ways, the level of geolocation abstraction is a representative "currency" for informational requirements and often provides a personalized way to think about about community at different levels.  It's important to correctly align the leve of abstraction with the requirements. Choose the wrong currency and the transaction might not make sense. For instance, using congressional districts (which are abstracts of geolocation that can span demographics as well as towns, counties, and zip codes) makes a lot of sense in the context of election results because the currency is the same as the success criteria. Election results might totally be different in a representative system when votes are aggregated by a different geolocation abstraction such as town or state or even nationally. This explains how a candidate can overall popular votes and yet lose an election. At the macro level, just meeting another American while traveling or living overseas can make an instant bond based on the geolocation abstraction of country alone.

Zip Codes and Town Names Might Not Be The Best Abstractions For Useful Data

So is zip code the right currency to use for hyperlocal blogging? Personally, I'm not so sure. I think about my zip code only a bit more than I think about my congressional district and feel about the same level of allegience to both. My chances of knowing a distant location's zip code is slim to none...well, actually just none except for 90210 and you know why that is...to be honest, I'd need to likely look it up each time. I don't even have an idea what state the 4 for zip codes in question are in. This could just be me but I'd need to have more of an emotional connection to a level of geolocation abstraction to think in terms of neighborhood or community than zip code. 

Town names have many of the same issues as zip codes albeit a stronger connection. There is still the issue of being located across the street from the dividing line that zip codes have. Just because I may live in {insert name of your town here} doesnt mean that I don't have hyperlocal interest in one town over.  The lines are simply too arbitrary.

Much like the misfit of the number of hot dogs in a package with the number of hot dog bus, we tend to use zip codes and town because that's how the information most often comes packaged. It's an easy abstraction and the path of least resistance.

Using The Raw Data of Latitude and Longitude

The FSBL geolocation engine uses latitude and longitude as our platform's internal geolocation currency. It's like using raw data rather than summarized data. Most people know what town or city they are interested in and we convert that information to a lat/long.  The math to determine a reasonable radius around that base lat/long is fairly trivial and works across the world. Just about everything in our web engine database has geolocation information wrapped around it making geolocation matching an easy process.  The link is from 2006 and is still very valid - sound functionality never goes out of style :P 

The best part of using lat/long is that it spans other geolocation abstraction boundaries. This is beneficial to both information providers and consumers of that information.  Users dont need to know what towns or neighborhoods are close to each other or what zip code they are in. In a sense, we've abstracted away the abstractions.

Suggestions For Next Steps

- wordpress plugin which allows town level lats and longs to be wrapped around blog posts

- integrating photos and videos posted by others into hyperlocal blogs by using lat and log information from photo and other media upload sites that support them.

- syndicating hyperlocal information and media so that it can be aggregated across hyperlocal sites

See Also:

Real Estate Websites of the Future
Linking Real Estate Content With User Context: Putting Better Faces On Contextual Content
Aggregating The Long Tail: Making Micro Useful
The User As The Center Of The Universe: Integration Of Real Estate Video And Geolocation

GyPSii, a geo-location and mobile social networking provider, has partnered with Shanghai Rannuo and China Unicom to launch 

its GyPSii service during the 2008 Olympic Games in Beijing. 

The GyPSii service combines social networking, mobile, location and web 2.0 technologies to offer China Unicom users a seamless mobile lifestyle experience. It will allow users to create pictures, video and blogs centered on all things Olympic while using their existing camera phone.  GyPSii users can capture their special moments during the Olympic games, upload the content to their personal space on www.gypsii.com.cn and share it with friends, family and communities

One of the programs I am using on my Nokia N95 is Locr to locate pictures. Here I hope is a link to the Locr site

NearBy

Summary:

While online social networks have brought people from distant locations closer together symbolically, people are still dependent on their location to form many of their ties. People often make connections with others who congregate in the same general area (e.g. school, work, etc). These ties are often made regardless of where people actually live.

NearBy is an experiment to see how people react when given information about their friends' home locations. Privacy will be a concern, so the program may choose to show distance between peoples' homes within a range, by number of miles, or show locations on a map. Often people who have met by non-local reasons will be surprised to learn that they live very close to one another. This can lead to closer relationships outside of the congregational realm.

It might also be fun to show other information based on the locations of the users, e.g. traveling salesman shortest path between users, popular dining areas near users and reviews, etc.

Technology Used:

• Bluetooth sensors
• The Umich student staff directory
• Google maps API

Research Goal:

• Interpret the different reactions to granular privacy changes
• Interpret the reactions to locality information provided

Educational Goal:

• Creating interesting "mashups" using different sources of data

Deliverables:

• Predictions on user response
• Working module and documentation
• Results from observed user response

Plazes now has a REST API (the old, pre-Activities version used to have an XML-RPC API). However, the information it provides is sometimes a little verbose for my purposes... I'd quite like to be able to simply ask the Plazes server "where am i?" and be told in a really simple format. Right now it looks like I need to know my user ID, and then go get http://plazes.com/users/<id>/activity.xml, and then get the name of the <plaze> element from the verbose XML that is returned. Something that just returned my current Plaze would be neater (for things like the ecto Applescript that I based off of Peter Rukavina's Adium Applescript of days gone by, which used the old WhereAmI API).

So I turned to Yahoo! Pipes to see how I could abbreviate the output. I've created a simple Pipe which can return a small piece of JSON containing a user's location when a user id (short name or number) is provided.

The base Pipe is here. It is not very useful in raw form as Pipes returns RSS by default, so running it will result in an apparently empty result on the web page. Try the JSON option instead.

You can add a user parameter (it defaults to my ID number, 6032), e.g. http://pipes.yahoo.com/pipes/pipe.run?_id=ZCgigzK93BGPeu60ouNLYQ&user=6032&_render=json

Running this query results in a JSON collection containing something like:

"items":[
   {
      "city":"Farnborough",
      "last_updated":"2008-01-07T14:28:00Z",
      "user":"Andy Piper",
      "plaze":"Andy's Office",
      "description":"",
      "title":""
   }
]

Annoyingly, Pipes only allows me to publish JSON or RSS (I'd prefer just simple text, but hey-ho).

Something more I'd like to do with this is to use Project Zero instead of Pipes, and make it more configurable to enable me to get more information on the user and plaze if needed.

At this stage, we take a look at http://www.ilovewillies.com (Only awaylable in the Google-cache)

Edit: I fixed the links

1. EasterEgg1.jpg
2. Barber.z5
3. Eggdump
4. Sophie's Choice
5. EasterEgg3.part.2
6. Dr. Mike Thorne's flashing dot at the side of your screen

These are all the informations we need.

Episode 1 gives us an image of Moss, the interesting part is the exif-geolocation. See: BBCentral's Blog

If you don't mind scrolling down a little bit, you will find a map in the right column.

My readers are now able to tell me where they are reading me.

The map starts in Bangkok, Thailand by default because that's where I am right now.
I will change it to Paris next month when I will move  there.

Feel free to click on this map, add yourself.

I will be happy to know who is reading me and from where.

Most importantly don't confuse "wardriving" in any way with the illegal act of "piggybacking", the theft of a Wi-Fi signal.

Wardriving is defined as searching, locating, and mapping Wi-Fi networks while driving in a vehicle and using a Wi-Fi equipped laptop. PDA, and now we can probably add smart phones, UMPCs, and Wi-Fi signal detectors to the list. Wardriving software can easily be downloaded for free on the Internet. One such program is NetStumbler.

Wardrivers plot the location of a network and record it on a website for the purpose of mapping an online network. The most popular website is WiGLE.net (Wireless Geographic Logging Engine).

Online mapping of network IDs can be used as a an alternative to GPS called geolocation. This is accomplished by triangulating the current position from the signal strengths of known network IDs. The Rugged Notebooks Blog has posted on this before in the post description about Navizon.

True wardrivers only identify and collect information about wireless access points (WAPs). They do not use the network resources. Using network services without authorization is unethical and illegal.

It may be hard to determine the attraction to wardriving, but it may be beneficial for the future of open source mesh wireless networks and the purpose of advanced network mapping programs.

Photo credit: Joe Mabel

Rugged Notebooks: Extreme Computers for Extreme Conditions™

Esperimento partito da un gruppo di universitari bolognesi che inserendo foto, segnalazioni di posti, locali, negozi, oggetti da scambiare, sbarcano ora in altre città come Venezia, Bari, Cagliari, Milano con molte altre in preparazione.

Se qualcuno fosse interessato a collaborare condividendo foto, segnalazioni di città non presenti tra le sopra nominate datemi voce nei commenti che vi metto in contatto con i webmasters! Innanzitutto non c'è ancora niente su Roma e altre città sparse per l'Italia! Per chi decidesse di aderire o fare anche solo pubblicità vi rimando a questa pagina dove potete trovare milleunmodo per linkarli!

Qui sotto vi faccio un esempio di due contenuti che ho inserito nella mappa di Bologna.

Accorrete gente! Accorrete! 

La Paresse [contenuto]

Piazza maggiore [foto]

Nokia a beau régner sur le marché du mobile en écoulant plus d'un million de téléphones par jour à travers le monde, cela ne lui suffit plus. Le finlandais l'a prouvé hier en signant la plus grosse acquisition de son histoire : le rachat de l'éditeur américain de logiciels de navigation Navteq pour 8,1 milliards de dollars (5,7 milliards d'euros). Car, non content de tirer ses revenus de la vente du téléphone, Nokia veut être au coeur des services Internet mobiles.

« Les services basés sur la localisation sont une des pierres angulaires de cette stratégie », a affirmé le PDG, Olli-Pekka Kallasvuo. Avec Navteq, Nokia met la main sur l'un des deux leaders mondiaux de la cartographie numérique. Fondé en 1985 dans la Silicon Valley et comptant environ 3.000 salariés, Navteq est le principal concurrent du néerlandais Tele Atlas, qui vient pour sa part d'être racheté par le premier fabricant de navigateurs GPS TomTom pour 1,8 milliard d'euros. Aujourd'hui, Navteq fournit des données cartographiques couvrant 69 pays, qui sont notamment utilisées dans les systèmes de navigation embarqués à bord des véhicules. Egalement propriétaire du site américain d'information routière Traffic.com, Navteq a généré, l'an dernier, un bénéfice net de 110 millions de dollars pour un chiffre d'affaires de 582 millions.

Au prix fort

Autant dire que Nokia a accepté de payer le prix fort pour se renforcer sur le marché encore balbutiant de la géolocalisation. Navteq a d'ailleurs reconnu avoir étudié d'autres offres avant d'accepter celle du finlandais, ce qui limite fortement le risque de contre-OPA. Si les analystes ont salué la pertinence stratégique de l'opération, beaucoup l'ont jugée bien trop coûteuse. L'action Nokia s'est d'ailleurs repliée de 1,84 % hier à Helsinki.

Cette acquisition s'intègre en tout cas dans la volonté du finlandais de se développer dans les services, quitte à venir concurrencer ses propres clients opérateurs mobiles. « Nokia se définit désormais par rapport au monde de l'Internet et non plus des télécoms », estime Vincent Poulbère, analyste chez Ovum. Convaincu que les 900 millions de mobiles Nokia en service dans le monde constituent la porte d'entrée idéale pour accéder à l'Internet, le finlandais s'est déjà lancé dans la musique (via le Nokia Music Store) ou les jeux (N-Gage). Il vient de les regrouper sous la marque Ovi (« porte » en finlandais) à partir de laquelle l'utilisateur trouve également le service de localisation Nokia Maps. Fonctionnant déjà sur le terminal multimédia N95 lancé l'an dernier, ce service est gratuit. Les cartes sont actuellement fournies par Tele Atlas. Avec Navteq, Nokia accentue la pression sur TomTom alors que les ventes de systèmes de navigation personnels devraient tripler d'ici à 2010, pour représenter un marché de 12,8 milliards de dollars, selon iSuppli.

En France, 2,5 millions de GPS devraient être vendus cette année, selon GfK, le double de 2006. Le mobile doté d'un navigateur GPS ouvre la porte à une multitude de services localisés et personnalisés.

C'est pourquoi Nokia vient aussi de racheter la société américaine de marketing mobile Enpocket, spécialisée dans les SMS ou MMS publicitaires. Actuellement dominé par Google et Yahoo!, ce marché pourrait peser 11,4 milliards de dollars en 2011, contre 2 milliards aujourd'hui, pronostique Informa Telecoms & Media Group.

Nokia mise sur la géolocalisation avec le rachat de Navteq

technorati tags: mobile, mobile alley, mobile news, mobilemarketing, mobile marketing, marketing mobile, interactive marketing, marketing interactif, marketing, Nokia, Olli-Pekka Kallasvuo, Navteq, GPS, Geolocation, Géolocalisation, Enpocket

2) che c'è gente senza braccia che ammazza a testate gente con le braccia

3) che, non vorrei essere ripetitivo, c'è la festa del Cactus questo fine settimana e c'è anche il relativo link.

4) che il senso della famiglia, checchè ne dicano i comunisti, c'è ancora e in Francia ne abbiamo candida manifestazione

5) che luminari informatici e comunicatori della amata penisola emiliano romagnola lanciano in prova un interessante servizio chiamato CentoMETRI. Presto vi parlerò estesamente. In breve si tratta di un citizen network che dà una geolocation ai contenuti più vari come foto, stanze in affito, pub, ristoranti e così via. I contenuti li inserirete voi stessi una volta creato un account!

6) leggete Mattoni, romanzo rosa per donne in posa o me ne avrò a mal e se me ne ho a male dovrete vedervelà con lui.

8 ) Camelot!

is just a model

il sette è un numero che non mi è mai stato particolarmente simpatico.

I discovered today that it's really easy. Geolocation is the term used for the conversion of an IP address to a real-world geographic location. For example:

There are a whole bunch of companies selling geolocation services - usually either as web services, or as downloadable databases containing IP addresses and geographic data.

Free geolocation data

If you're looking for something free, check out MaxMind GeoLite City. You can download a huge pair of CSV files (one for IP blocks, one for corresponding location data) from their website and import them into SQL Server, and you'll end up with a schema like this:

Now, there are 4 billion possible IP addresses, so rather than explicitly listing every one as a separate row (which really would hurt SQL Server) the MaxMind people have split up the IP address space into blocks which correspond to a single geographic location. To help SQL Server find which block an IP address is in, they have defined a one-to-one mapping from IP addresses to BIGINT values, as such:

So now you can (almost) instantly obtain geographic data for any IP address by defining a SQL user defined function (UDF) such as:

[sourcecode language='sql']CREATE FUNCTION [GetGeocodingData]
    (@a tinyint, @b tinyint, @c tinyint, @d tinyint)
RETURNS @ReturnTable TABLE (
	City VARCHAR(255),
	Region VARCHAR(255),
	Country VARCHAR(255),
	Latitude FLOAT,
	Longitude FLOAT
)
AS
BEGIN
	INSERT @ReturnTable
	SELECT 	lo.City,
		lo.Region,
		lo.Country,
		lo.Latitude,
		lo.Longitude
	FROM Blocks bl
	JOIN Locations lo on lo.LocationID = bl.LocationID
	WHERE (CAST(@a AS BIGINT)*256*256*256 + @b*256*256 + @c*256 + @d)
	  BETWEEN bl.[BlockStart] AND bl.[BlockEnd]

	RETURN
END
GO [/sourcecode] 

In this example, the parameters @a, @b, @c, @d correspond to the four parts of the IP address (and should therefore be integers in the range 0-255).