Twitter as a Platform for Location Based Services

29Apr07

During my daytime job, I have stumbled across many definitions of what’s a Location Based Service and what’s not – most of them including the requirement that in order to qualify as ‘location based’, the service must determine the user’s location automatically in some way, e.g. via the mobile network operator, a GPS device, known locations of nearby radio (WiFi, Bluetooth) or Infrared beacons, etc.

Meanwhile, there’s lots of examples of location-specific services out in the wild that don’t use positioning technology at all: instead, users just type in zip codes or street addresses. Just check out Yahoo! Local or Yelp, some of the social LBS I covered in a previous post, or – let’s not forget – my own project, FONFinder! Hell, I can even type ‘pizza near mariahilferstr. austria‘ into Google nowadays and it will return a list of pizza restaurants in the center of Vienna. Dead simple. And definitely able to elegantly (and robustly) answer those questions that are typically associated with ’standard’ LBS scenarios: ‘Where is the next restaurant?‘ or ‘Where is the next pharamacy?‘ (Ok… at least that’s what people considered LBS in the early days, in the late 90’s. Nowadays that has, of course, been replaced by ‘Where is the next Starbucks?‘. But, hey, wouldn’t the answer to that question always be ‘just around the corner‘?)

I think it’s definitely justified to call these services ‘location based’. Ok, there’s the drawback of the user having to type – which can be annoying on a mobile phone – and having to know the street address. But it should always be possible to find a street sign somewhere…

On the positive side, it could be argued that these low-tech LBS don’t suffer from the same inherent privacy problem as their ‘true’ LBS counterparts – because the system has no way of telling whether users aren’t actually lying about their location (which, I think, is an important aspect!)

Anyways – the recently hyped social messaging service twitter is a particularly intriguing example of a service that includes such low-tech LBS functionality. I’m not gonna explain it in lengthy words now – just check out twittervision and you’ll see what I mean. (Watching the world map scroll as the messages come in has something almost hypnotic to me )

What has really opened my eyes about LBS and twitter, however, was a recent blog post by Stefan Geens of Ogle Earth, commenting on mologogo’s twitter integration and (here it is again) my FONFinder (thanks, Stefan!!): Not only does twitter feature an open API that allows Web services to react to twitter messages; but also, first applications are now becoming available that link twitter to GPS – just as mologogo is doing it! (Check out thilo’s post at ‘Das Zentralorgan’, for an additional example.)

My personal bottom line is: with it’s simple interface (SMS) and basic functionality, twitter could turn into the perfect middleware/ecosystem for those basic ‘Where is?’-type of location based services. To practice what I preach, I’ve decided to follow Stefan’s suggestion I’m building a twitter interface for the FONFinder. The goal of this being that you will (eventually) be able to tweet something like ‘d fonf l:mariahilferstr.,austria’ or ‘d fonf l:48.232,16.411′ (with nothing preventing you from taking these coordinates directly from a GPS) and get the street address and distance of the nearest Fonspot SMSed back to you. That’s it.

Wow. That’s my longest post so far. And you’re still reading it! Thanks!

---

Possibly related posts: (automatically generated)

• Twitter valuation off by half: NeXt Up Research
• A Wireless Street Fight
• Twitter Serves Up Ideas From Its Followers

Filed under: Development, FON, FONFinder, Geek, Geography, LBS, Location, Mashups, Mobile, Mobile Social Networks, Phone, Technology, Twitter, Web, Web 2.0   |  7 Comments