Damien Kee from Domabotics.com has written a new booklet on datalogging with the NXT. I recently got a chance to read through it, and my first impression is that it was well-named. Titled "Datalogging Activities for the Busy Teacher", it is structured as a series of almost stand-alone short projects or lessons, ideal for teachers with little time and little or no background with the NXT.
The book starts with a few pages talking about what datalogging is & how it works, and then just a very brief review of the built-in datalogging environment of Educational version of NXT-G 2.0. At the end of the book are a series of handouts and worksheets for the students, saving the teacher the time of developing these themselves. But the bulk of the book is a series of chapters organized by sensor: datalogging with the touch sensor, datalogging with the light sensor, etc, including the new temperature sensor. Each chapter starts with a description of the sensor, and what sort of readings it yields, before using it in a couple of experiments. Many of the experiments even come with a series of guided questions, including sample answers that students might generate. Chapter 8 steps up to experiments using more than one sensor, and shows some new ways to coordinate them.
On the whole, it's a nice set of simple hands-on experiments. There are some that kids will get a big kick out of (some that depend on human behavior of their fellow students), and some that teachers might value because they move towards even more sophisticated topics like physics (like building a "timing gate" with two light sensors). Some of them like, "The Bouncing Ball", are excellent examples of using datalogging to monitor something that otherwise would be very difficult to measure by other means.
There were some things I found to be a problem; for the most part they weren't due to the book, but due to the tools and the setting. Datalogging under NXT-G 2.0 is very easy, but it has some serious limitations as well - since it just logs the sensors, things like recording a series of touch sensor presses is poorly handled. For experiments like this, the students need to go over a graph of the data to manually record each push and the time they happened at. Here, using the built-in environment ends up making more work for the students, not less. The datalogging environment doesn't log as fast as it could (useful for some things, like the "bounce" experiment), and there's no real discussion of the memory limitations (why you don't want to log every 2 seconds for a week). In short, some of the experiments would be easier and make more sense if the students didn't use the new environment at all, but spent some more time understanding what they really needed to record... and how to do it.
The only other thing I had a problem with, is that the experiments seem too simple... there are a lot of other options, alternatives, or directions to take these experiments in, that I found myself asking "but what about...?" often while reading it. But that's not really a problem with a book; in some ways, it's one of the strengths of the book. Here Damien Kee has laid down the basics in a way the a teacher can adopt very easily, with very little effort - yet it opens up a world of possibilities for both the students, and the teacher, to take it much further. Here's where the real strength of this sort of workbook lays: in opening up the NXT to practical experimentation for people who haven't had the chance yet.
Teachers, you'll like this one, especially for grade-school kids and tight time budgets.