An LDR actually is just a resistor that changes its resistance based on how much light is shined on it. So from that standpoint it is actually passive as far as the signal path goes. The high distortion artifacts that were quite apparent in testing are from the non-linearity of the LDRs used, poor LDR implementation, poor power supply design/layout for the LED light part that shines light on the LDR(s) and whose intensity range is controlled by the volume control potentiometer, crude wiring paths/layout, etc. Nevertheless, it is a far less linear and smoothly controllable approach to a passive preamp than a simple potentiometer or step attenuator - without the whiz bang element of using light.
Also, just because it has a power supply doesn't mean it's not passive preamp. I could build a passive preamp around a remote control motorized Alps potentiometer and the signal path would only be going through the passive variable resistor (potentiometer). The power supply would only be powering the remote controlled motor that turns the potentiometer based on inputs received from the remote control and to power the digital readout to let you know what volume level it's set at, etc. That is a purely passive preamp approach that has a power supply, and adds the convenience of remote control, and volume level readout.
Nevertheless I have built numerous passive preamp prototypes, including an LDR version, and I still generally dislike ALL passive preamps for a couple of reasons. First, they have no gain to preserve the dynamics of analog sources which are not very high in output voltage like digital sources are. Second, the output impedance of a passive preamp varies depending upon the resistance setting of the variable resistor and at some volume control levels may be unsuitable for driving some power amplifiers, affecting the overall sound quality - which then demands the inclusion of an output buffer circuit which in turn introduces some negative effect on transparency, etc.
