What I'm getting out of Google IO

So, at breakfast, I sat down and mapped out what was of interest to me, an academic developer, at this conference. Here's what I came up with:

• The Google Charts API: Programmatically charting winds up being surprisingly tough. You have to install libraries which often have dependencies which can be hard to get on every platform. If you want to post your charts on the web, you suddenly have either a processing issue on your server or extra image files to upload and maintain. The Google Chart API (and related) fixes these problems.
• The Buzz API: This will be tremendously useful in the classroom. It will help us figure out what's going on within our group.
• I've decided against app engine. It provides certain benefits like obvious web hosting but does not really work great for the kind of batch processing, sometimes long latency stuff I do. Servers still serve some role (pun intended).
• I'm considering elements of html5. As today's Android keynote at Google IO is demonstrating, javascript performance is a big issue in html5. I see this as still a couple of years out even at Google's rapid pace of development.

Update: I forgot to mention the Google Analytics API which will play a big role in teaching.