Doug Marman crushes compression questions
He's on the blog roll, but because he posts only once a month or so and doesn't have an RSS feed, I don't always get to Doug Marman's posts right away. Thus, I'm only now reading this very good treatise on the ridiculousness of court admissibility problems with compressed video. Make sure you read the comments because he provides more evidence for his arguments there. It's especially heartening because the session we had at TechSec tackling this very same issue traveled along the same lines. The video expert we had there, who works often with the Dallas prosecutor's office, didn't care a whit about compression, etc. All he cared about was whether the image was good enough to see what was happening. Sure, watermarking was important, but way more important was chain of custody, just like any other piece of evidence. If the gun that was used in the murder just happened to take a swing through Burger King, where it was left on a table, forgotten, and retrieved later, then it's going to be a problem when it's submitted as evidence. Same with video. You have to know where it's been and who's handled it. But, big deal. That should go without saying. There's some good stuff from Doug about the differences in compression types in general, too. The whole thing is a must-read for people selling video systems where the video will likely be used in court.