Interesting stash of documents, thanks! I'll look through them. The 2005 date looks about right for the developer's guide, and does mention the development kit. I just downloaded it, to go through it later.
The box in which I found the 2414S also had a circuit board I had made at the time, with a connector for the 2414S at one one, and a ...mini phone jack at the other! It brought back fun memories. As some will know, I was heavily involved with ADI and the Ocelot back then, and among my crazy projects, I wanted to be the first person on earth to have the Ocelot turn on an Insteon device using the Insteon protocol. The Ocelot didn't have serial output (though a serial bobcat became available later). But I did have a SECU16IR, a module with 16 IR blinker outputs. IR commands consist of a modulated carrier (usually at around 38 kHz) modulated on and off to produce a bit pattern. Well in the Ocelot, the IR codes are encoded as 1 and 0 bits, each of which means that the carrier should be on or off, and each time cell is 40uS long. So for example a burst of 120 uS of carrier followed by 80 uS of nothing would be encoded as 11100... continuously from one byte to the next. Knowing that, I figured that I could replicate a serial signal envelope by encoding the serial bit durations as IR codes. So the mini phone jack connected to a SECU16IR output. Then the modulated envelope went to my little board that filtered out the carrier (using a simple RC network) and the resulting signal envelope fed into a MAX232 chip to produce RS232 level signals, and I fed that into the 2414S. It worked! (I'm laughing as I'm writing this). That's why I remember that the baud rate was low, because the 40uS time cell resolution wasn't good enough for anything above 4800 baud.