Electronic – Best development board for ARM9?

The Launchpad is much, much cheaper than pretty much any other development kit you'll find. Many dev kits are several hundred dollars - How much are you willing to spend?

In my opinion, a dev kit should:

• Not be much more expensive than the components which were used to manufacture it. Some manufacturers put significant markup on their kits, imagining that they have their buyers locked in. They don't. If a kit is ridiculously priced, I buy a different one.
• Not lock you into a toolchain, and really should have an open-source and free toolchain available. The Launchpad does miss this point, but if he's interested in PICs and MSP430s, he'll have MPLAB installed already. For some tools, you won't find this option anywhere.
• Have some on-board peripherals and connectors, but should also make all of the microcontroller pins available with jumper or solder blob options, preferably through breadboard-friendly .1" pitch connectors.

Luminary Micro (aka TI/Stellaris) makes some evaluation kits which fit that bill today. I have the LM3S6965, which satisfies each of my above criteria, and has the components you specified: LCD, Ethernet, USB, and Cortex-M3. If you're reading this next Christmas, you'll have to re-evaluate the options.

In your search for an assortment of components, you should consider three things:

1. What he already has. I can't help you on this one.
2. What is in assortments that are available from places like Sparkfun and Adafruit (and elexp). Sometimes, one of these will fit the bill and have a quantity discount. Other times, these will have a selection markup, and be more expensive than buying individually from a major distributor.
3. What he needs for his specific discipline. There are related questions about Equipment, Robotics, Components in general, and ICs to Interface with Microcontrollers.

That should get you started on a component search.

As for hobbyist magazines, I'd strongly recommend Circuit Cellar. It's a good deal more advanced than Nuts and Volts, but shouldn't be over his head if he's already looking into ARM processors and high-voltage stuff. In fact, Nuts and Volts might be too simple at times, but Circuit Cellar will never be too basic. The goal is always to improve, right? Jerrry [sic] from Slashdot posted a good overview of the magazine options a while back that I just found when trying to remember how to spell Elektor, another option to consider.

But before you buy anything, consider what you want it to mean to him. If you buy him components, they'll go in the parts cabinet and be put into projects over time, and your impact will probably be forgotten. For an example outside of the electronics realm, would you rather he buy you socks or a pretty scarf? Socks are more basic to your hobby of keeping warm, but when you wear the scarf you'll think of him, and when you wear the socks you won't. With that motivation, ignoring the technical evaluations of dev kits, components, and magazines, I recommend that you look for something that could cause him to think of you. A subscription to a magazine would be good (for 30 seconds a month, then, if he's like me, he'd be distracted by the articles...), but a tool might be better. You didn't mention that he had a bench power supply or multimeter - These would be tools that would last a long time, wouldn't become obsolete, and would be used every time he sat down at his bench.

Sorry for the long answer - I've been thinking about what to put on a wish list for Christmas for my girlfriend.

So from the previously linked answer, I think the most interesting/beneficial things for me would be the bullet about learning the tools (compiler and linker), and learning different styles of software architecture (going from interrupt based control loops to schedulers and RTOSes)

Porting a small operating system to a new device could help you to understand schedulers and RTOSs. FreeRTOS is popular and well documented. eCos is another.

Writing a bootloader is a good way to get to grips with a linker as you'll want to divide up memory and flash into regions.

Another tip is to pick a completely new architecture or chip and build yourself a development board. Forcing yourself to start right from the beginning and look everything up in the datasheet is a good way to learn.

Explore Protothreads. Try writing the same programs in both a threaded and state machine style. Once you're done with Protothreads, write a real thread scheduler.