I just found a really outstanding article about how the four basic different types (opto, variable-mu, FET, and VCA) of compressors work in the February 2001 Electronic Musician, by Michael Cooper. This reinforces a lot of what we already said:
To sum up the section on optos:
There is a sidechain consisting of an LED shining on a light-dependent resistor. As the volume gets louder the light gets brighter, creating a corresponding increase in compression of the audio signal. Because the LDR has an inherent "memory effect", it releases slower when the light is brighter or has been shining for a while. That means heavier or continuous compression result in slower release times. Also, the elements have a natural lag time in their attack response, and a soft knee. This gives them a natural sound, but makes them not appropriate for capturing fast transients.
Vintage opto units usually had only two controls: gain reduction and gain. Turning up the gain-reduction results in lowering the threshold giving more compression. The gain control is make-up gain.
Good examples of opto compressers include
the LA2A, Anthony DeMaria Labs,
Joe Meek C2 and
SC2.2, and the Millenia Media TCL2.
Contrary to a previous post on this thread, the 1176LN works on the FET (Field Effect Transistor) principle (not opto), which allows much faster attack and release times, but can be more prone to artifacts (pumping) and distortion if not set correctly.