I have to jump in again and doff my cap to PJ2, who is doing a sterling job of laying out the complexities of the situation regarding flight data - not just in terms of the ramifications regarding disclosure, but also in terms of the scientific nature of forensic use of that data. To reiterate, the raw data must be processed heavily from its initial digital format into a format that can be read by humans as well as machines - whether that be the tabular format in the earlier examples shown or the plotted graph format as used in the AF447 report. Running that into a 3D animation requires further processing (e.g. smoothing) that, if not done properly, runs the risk of providing a misleading presentation of what went on. That, due to the relative ease of understanding such an animation can illustrate to non-technical people can cause widespread public misunderstanding and lead to long-running misconceptions about accidents becoming received wisdom. This is enough of a problem anyway, given the press's habit of promulgating unproven information to the public.
The irony is that flight data and CVR recordings/transcripts are so tightly regulated in terms of disclosure precisely because of the pilots' unions refusing to countenance use of these recorders unless such assurances regarding disclosure were given back when they were first introduced. A few of those same pilots unions have called for further disclosure and second opinions when the outcome of an investigation is not to their liking, and I can't help but feel that you can't have it both ways.
What is even more puzzling in the case of AF447 is that the investigation report, while acknowledging dubious actions on the part of the crew, goes to great lengths to highlight that the likely roots of those actions include improper or insufficient training on the part of the airline, insufficient oversight on the part of the regulator and possible design issues on the part of the manufacturer. Cursory use of the "Find" tool on any of the reports reveals that nowhere is the phrase "pilot error" used, nor any other phrases to that effect.
As a result of this, in a criminal or civil case neither Air France, the DGAC nor Airbus can point to the report to limit their liability because their part in the genesis of the accident is part and parcel of that report. I'd be prepared to wager that if the raw data were released, converted and parsed by the legal teams of the union and the families (which to do scientifically and correctly would in and of itself be a massively expensive undertaking), they'd find no germane information that they didn't already have.
Last edited by DozyWannabe; 16th Jul 2013 at 15:22.