Google Blogoscoped

Wednesday, February 9, 2005

Behind Google Maps

Joel Webber analyzed the DHTML behind the new Google Maps. It’s nice to see a clean XHTML + CSS approach with absolute positioning and “console game" style, tile-based scrolling. Instead of using XMLHttp requests like in Google Suggest, there’s a hidden Iframe which fires the parent’s “load” function.

There’s two things which may surprise many: Google Maps uses client-side XSLT to convert the XML packages pulled via the Iframe into good old HTML. Also, the shades are generated using PNGs. And this works more or less cross-browser (while Google can’t possibly cover all browsers, they do cover more than you might expect).

It looks like Google is doing the hands-down best DHTML out there today. It’s not even they make the most complicated things, but more the fact they actually use the best and easiest approaches available. And the real magic, as Joel says, must be going on behind the application on the server-side we can’t see:

“The fact that Google’s servers can handle all of these images requests, route finding, line drawing, searches and the like so quickly is the real magic. I also want to point out that their map renderer (or the one they purchased) works much better than all the other ones I’ve seen on Mapquest, Mapblast, and the like. That alone makes it worth using, if only so you can actually read the map!”

And a reader at Slashdot comments:

“One of the things that delights me about Google is a certain kind of freshness I haven’t seen elsewhere as often as I’d like. They have the characteristic you used to see in innovative software that when you describe the latest Google feature, it doesn’t sound all that new, yet when you use it you get that feeling that something unexpected has been revealed.”


Blog  |  Forum     more >> Archive | Feed | Google's blogs | About


This site unofficially covers Google™ and more with some rights reserved. Join our forum!