Hellschreiber is a mode that was patented back in 1929, by Rudolf Hell. Like SSTV or facsimile, Hellschreiber is a "visual" mode. Whereas RTTY or PSK31 software can interpet the received sounds and convert them to characters or bytes, Hellschreiber must be read by the eye.
Hellschreiber was used originally for passing newspaper headlines. To whom they were passed, I'm not sure. Modern computers have created a surge of interest in this old mode, since it no longer takes elaborate mechanical equipment to generate the required sounds.
The image above (before I changed my call to K4JCW) was created by using two PCs, one acting as a transmitter, and the other as a receiver. The audio was transmitted through the desk-top stereo speakers, and the receiver was a laptop with a built-in microphone. The two machines were about 3 feet apart. The ghosting you see between lines is a result of the audio reflecting off the hard surfaces in the office. Hellschreiber, like Morse, takes advantage of the signal processing capabilities of the human brain. While a PC *might* be able to decode the text using Optical Character Recognition (OCR), it's unlikely that if the signal got any noiser that the OCR would have much, if any, success. The human brain, however, can quite easily decode the text (well, assuming you're literate. Of course, if you weren't, you wouldn't be on the web...)
There are a few good pages on this mode, but the best is probably "The World Of Fuzzy Modes".
There are now several variants on the original Hellschreiber mode (generally known as Feld-Hell). One is PSK-Hell, which uses Phase Shift Keying. Another is MT-Hell, and has two variants, which are C/MT-Hell (Concurrent Multi-Tone Hell), and S/MT-Hell (Sequential Multi-Tone Hell). All told, there are 12 different Hell modes (some of which aren't legal in the US). Check the modes page at 'The World Of Fuzzy Modes' to see what exists.
I could duplicate a lot of good explanations about the various modes, how they work, what the advantages and disadvantages are, etc. Instead, I refer you the above link. ZL1BPU has spent a lot of time putting together detailed information, and it's pointless to reiterate it all.
I will add that supposedly one can find Feld-Hell on 14.063.000 or 14.063.500 (depending on which web page you believe). I can say that in the week of 11/25/99, I spent a fair amount of time tuning around trying to find a live transmission, and have yet to hear one. I'm looking forward to a good copy of a live QSO. According to one our local pundits, you'll find a fair number of German and Russians experimenting with this mode. For the sample above, I wanted to use something I captured off the air. Instead, I had to resort to putting two PCs near each other, one sending, and one receiving.
While I've not worked any live QSOs, the software that I'm most impressed with is the IZ8BLY package. While the Gram 18.104.22.168 spectrum analyzer is really good, it takes more time to get it setup and going (which means grabbing a .WAV file from the web, and tweaking the FFT (Fast Fourier Transform) parameters until an image is decoded), and you have a much narrower window to see any messages in, before they scroll off the screen.
The IZ8BLY package, on the other hand, displays a lot more samples, and doesn't require tuning the FFT. I'd start with it, and once you've seen something, experiment with the Gram package. If you're looking for a general spectrum analyzer, the Gram package is killer.
There are a lot of modes being used on the air that are interesting, but for some (perhaps bizarre) reason, the Hellschreiber mode really piqued my curiosity. Kind of like SSTV, there's just something intrinsically cool about images forming on the screen from a modulated tone...