Back on Stamps. I used sockets (as explained in JNI workaround) to get access to GSM data provided by the native Symbian code of Place Lab’s and CRAFT’s GSM trackers. I also extended the coverage of the map of Geneva. Latency is good. The current 145 messages of the database are loaded via GPRS in 2-3 secondes. Runs great on a Nokia 6630.
Buzztracker is software that visualizes frequencies and relationships between locations in the Google world news directory.
Buzztracker tries to show you how interconnected the world is: big events in one area ripple to other areas across the globe. Connections between cities thousands of miles apart become apparent at a glance.
Cartograme versions off this kind of news or buzz maps might be more readable.
The Barcelona Media Park will be my future home in 2006 at the University of Pompeu Fabra.
.jpg)
MGMaps that stands for Mobile Google Maps, is a J2ME MIDP 1.0 interface to display Google Maps. Very much still under development…
The Dutch National Committee for International Cooperation and Sustainable Development setup Mapping Worlds that has a nice set of distorted maps to rais international awareness on various topics like the EU, asylum and development aid.
Place Lab that promotes a mixed approach to positioning (GPS, WiFi, GSM and Bluetooth) are gaining some mainstream attention in Cnet’s article “Intel experiments with Wi-Fi as GPS substitute“. One of Place Lab’s key phrase is that “humans spend most of their time inside or in dense environments where GPS does not work”. To help prove the point, Ian Smith (Intel Senior Researcher) wore a backpack for 3.5 months with a monitor. It found that, on average, he spent only 4.5 percent of each day outside. However, the experiment was most probably ran in rainy Seattle.
The article contains a flagrant inaccuracy, by mentioning that “wireless can also determine height and thus figure out what floor of a particular building a person is on”. If it was the case, this blog would not be named “7.5th Floor”. I guess the write meant “roughly predict in some cases” instead of “determine”.
MapQuest Find Me runs on GPS enabled Nextel Phone. It allows to identify your location on a cell phone, find out what is nearby, and share the location with trusted others.
2 articles explaining how to automatically start a MIDlet by using the Push Registry
- How can a MIDlet be launched automatically?
- The MIDP 2.0 Push Registry
Thanks Arvind Gupta for the tip!
The mobile industry is still so desperately technology driven. Even without real services, operators still push the technology to market. Two recent articles on Location Based Services and mobile TV show clear examples of this push frenzy:
Location-Based Services “Gathering Steam”
There are instances where location information could be useful other than getting driving directions or finding the nearest cafe, even if operators can’t think of what they might be. Sooner or later, carriers will have to realize they can’t provide every service themselves, and that some other people might be able to come up with some things to make them money and provide use for subscribers.
Never Mind The (Lack Of) Interest: Mobile TV Marches On
There is only a little consumer “enthusiasm or demand” for mobile TV, an unwillingness to pay for it and low desire to see mobile TV functionality in forthcoming devices. [...] New services are too often based around the “what can it do?” question — which is concerned with the functionality of the technology. But what consumers want to know is “what can I do?” — that’s about services.
John Krumm of Microsoft Research gives an interview on Wi-Fi positioning, based on the paper he co-wrote Accuracy Characterization for Metropolitan-Scale Wi-Fi Localization with other Place Lab members. He goes exactly in the same direction I took for developing CatchBob!. That is WiFi positioning is of 13-40 meters in an uncontrolled urban and residential environment, measuring the signal strength is almost useless, and complicated algorithms are not more accurate.
Using Wi-Fi to Make Your Device Find Where You Are interview of John Krumm by Rob Knies.
We learned that we can estimate the location of a mobile Wi-Fi device to an accuracy of 13-40 meters, depending on the environment, using just the access points that are already deployed in urban and residential neighborhoods.
We found that measuring the signal strength of Wi-Fi access points really doesn’t help that much in estimating position.
One of the pleasant conclusions from our study is that, for this task, a simple algorithm worked almost as well as a much more complicated one.
