Look at the linked to board specs, it has a built-in debug interface called OpenSDA. And while you can use an external debugger, the Cortex-M0 and M0+ support only SWD, not JTAG, and OpenOCD support for SWD is still under development.
The Cortex-M0+ is an interesting core, but AFAIK none of the open-source tools support on-board debugger interface (but the interface itself should be open, so it's "only" a matter of implementation), and SWD support is still experimental at best, so it looks pretty much Windows-only at the moment.
Normal I2C addresses are 7 bits, so 127 addresses total, of which 16 are reserved for special purposes. There's a protocol extension that allows 10-bit addresses, but it's relatively rare. Now keep in mind these are device addresses, not IDs. The address tells you how to talk to a device, not what kind of device it is, so the limitation is that you can have a bit over a hundred slave devices on the same bus. To avoid address collisions, most I2C devices allow you to configure their address eg. using a few package pins.
There is an optional device ID extension to the I2C protocol with USB-like registered vendor and device IDs, see page 20 of the spec.
I2C doesn't use chip selects, instead the bus master generates a START condition to signal the start of a transaction, and a STOP condition to signal the end of the transaction. After generating the START condition, the master sends a byte containing the 7-bit slave address, and a single bit indicating the data direction (1 = read, 0 = write). So your example becomes: START, indicate you want to send data to device 0x77, send the two data bytes 0xF4, 0x2E, STOP, START, indicate you want to read data from device 0x77, read four bytes, STOP.
There's more to how I2C works, but that is the basic idea. The official standard explains everything in detail, but is maybe not the best tutorial.
If they're an option, Farnell have a warehouse in the UK (RS also have one, but their prices always seem more expensive than the competition). Mouser ships from the US, but every time I've used them has been very painless, no extra fees or anything.
Do you have any good advice for making ribbon cables? When I've tried either the latches on the side of the top part of the connector tend to break, or the whole thing comes apart.
They're also afraid of security issues caused by exploiting bugs in the emulators/VMs, which is why for instance third-party web browsers can't use their own JavaScript engines.
Apple at least used to be pretty weird about all sorts of emulators and virtual machines on iOS (first a blanket ban, then allowed for a fixed set of emulated applications, then I guess they even allowed a C64 emulator to run the Basic interpreter). I dunno what the current rules are, but hopefully this guy has done his homework and won't run into problems on that front.
