I have a dear friend who is a film editor taking a look at things (once he manages to stop laughing).
One thing is that apparently there does not seem to be info on the time that the datafile was originally created. The data files sent to me via DropBox now have “creation date” & “modified” date " as the same: i.e the date that they were uploaded to DropBox. Might be a game ender, right?
Anyway I’ll keep you posted if we do find a way forward.
I think Created is the original, legacy method of identifying files. I would have a hard time believing they would leave it out.
We should remember that Windows likes hiding information. Do you have filename extensions revealed? Let’s count the number of forum poster complaining that Audacity will not open an MP3 file which turns out to really be MyMusic.m4a or something else entirely and Windows is hiding the “.m4a” part.
You could try a test:
Create a new file and note the file creation time.
Upload it to dropbox, then download it on another device and check the file creation time - my guess is that it is later.
Is Windows really helpless here? How about the “attrib” command in Windows 10 Terminal? I know Terminal and the DOS commands aren’t native any more, they’re an app, but still.
If everything else falls apart, it’s still possible, since you have the transcription, to match the six second chunks to the whole script rather than just the relationship to nine other files. You can even get the computer to help a little. Type out a six second snippet’s words and let text matching compare it to the transcript.
Find:
“My dear Aunt Sally sells slimy sea shells and market’s them to the visitors in the village square.”
That’s about six seconds at normal reading speed and the text is not likely to be mistaken for anything else in the show. That should give you one hit.
The process gets faster as you go since you have fewer and fewer files to match.
The Hollywood rescue, since you have the script, is get somebody to read it in a studio and simulate the original presentation.
I’ve been doing a lot of alpha soak testing of Unitary Project format and in real-life use it holds up amazingly well (I’ve been bold enough to use it for my production work for a while now).
What’s holding it up is that we have a few edge-case bugs (mostly disk-full situations) - but these can be serious enough to cause data-loss, so we can’t release without fixing those (even though they are edge-cases).
