Here in
ITS, we've begun some earnest talks on location based services. Location based services are hard to talk about. They are hard to talk about because they inevitably lead to discussions on fruitless services. Services that sound "cool" when you are sitting at a table talking about them, but in actuality, are of little use. Like "find restaurants near you"; such that, if you are in
Tomlinson and you log on to the
Portal, a little window widget says "you're near a Subway restaurant." See, right now, right in your head, you may be thinking to yourself, "
ooooh, that would be neat." No... it... wouldn't. It would be annoying. It's an example of information
pushed at you. Information thrown into your face whether you want it or not. If I want to know what restaurants are nearby, I will
poll for that information. All you need to do (and by "you", I mean a "service" or "set of services") is make sure it is easy to discover where I go to poll that service and make it as seamless an experience as possible.
Push information == bad :: easy to find and easy to use
poll information == good. But, that's a tangent. You can think about that on your own. What information is
pushed at you (or is "interrupting" information) versus what information do you
poll for; which do you like better; and how easy was it to find the place to
poll. Etc. A phone call versus an email. A big blinky popup versus a targetted search for consumer goods. So, back to location based services. Or, the genre thereof. People have been playing with
Google Maps, and so have I. I want to be able to map people posting to
Blog@Case. To do this, all I need is to put location-based data in each blog's
RSS feed and hook into the
Google Maps API. There is a pseudo-standard for including location-based data in
RSS feeds, it's called
geocoded RSS and it, basically, looks like:
<?xml version="1.0"?>
<rss version="2.0" xmlns:geo="http://www.w3.org/2003/01/geo/wgs84_pos#">
<channel>
...
<item>
<title>An example annotation</title>
<link>http://example.com/geo</link>
<description>Just an example</description>
<geo:lat>26.58</geo:lat>
<geo:long>-97.83</geo:long>
</item>
...
</channel> </rss>
So, mapping an RSS feed with that data becomes slightly trivial. Last night, I spent some time doing just that. Here is a mock up RSS feed –
http://blog.case.edu/jms18/code/googlemaps/rss20.xml. And, here is a stab at mapping that –
Playing with Blog@Case and Google Maps. Of course, the real trick is getting that geocoded data into each blog's RSS feed. There are several ways I can do that. I can get the blog system to look up your latitude and longitude coordinates by your IP address, or I can offer a text box you can populate with a street address when you are creating an entry or it can be a check box that just says "I am in my dorm room" or "I am at work" and then lookup that information in the directory server. Much more thinking and playing to be done. And, obviously, combining this with being able to
map
Wiki@Case entries will be equally amusing.
Because I know
Greg is gong to ask for them, here are the SQL queries I used. The query to generate the pages with the most different editors is the following:
Because I know
Greg is gong to ask for them, here are the SQL queries I used. The query to generate the pages with the most different editors is the following:
