Yes, you've got it right. They shot a laser through a Cesium Gas(I think) and the detector at the other end, detected an event shortly before the beam enters the chamber. If I recall they figured something was Hopping across the wave peaks to arrive ahead of when it should have. What hit however was just the leading edge and you couldn't encode anything into it. In other words, you couldn't send information to the past as nothing can be determined until the entire wave cycle arrives as the other end of the chamber. It's a bit over my head, so I've probably message up the details. It was uhm, a Japanese team, like Mitsubishi or NEC or something like that.
Mazda?
You did a pretty good job recounting that experiment. Like you say, when something seems to be going faster than the speed of light (or) backwards in time, it turns you can't do anything useful with it (unless you have a wormhole at your disposal).
That was a really bogus experiment. They shot a gaussian laser pulse into a chamber of excited gas. The excited gas acted like an extremely high gain amplifier to the rising edge of the input pulse. Once the gas emitted its photons and became de-excited, the gain fell. This all happens in a gaussian profile.
The exact same "faster than light" effect could be created with an op-amp and an arrangement of caps, diodes, etc.