So why is it that nobody noticed until yesterday (or at least called to my attention) that the second segment of the meeting (the second twenty-minute segment) was missing?
As some of you know, our fancy new camera's native output format is the not-immediately-usable MXF. In order to have the camera also produce quick-and-dirty MP4 files, we needed a "proxy card." Detcon 1, the 2014 NASFiC, partially underwrote the cost of us buying that card. I made a point of mentioning their sponsorship at the start of each of the four days' meetings.
What we did not realize until we were at the convention is that the MP4 files that the proxy card generates are limited to 20 minutes per file. Now it's a perfect break; that is, the second file starts on the next frame after the first one. We just didn't realize it was going to happen because we hadn't actually recorded anything more than 20 minutes long with the proxy card before.
Those of you who were there know that we had "technical time outs" about every 30 minutes. That's because the 16 GB P2 cards the camera uses can only hold around 50 minutes of video at the resolution we were using, and we knew we couldn't stop at exactly 50 minutes, so we needed overlap time. (The camera came with five 16 GB cards, thank goodness. Bigger cards are very expensive.) So approximately every 30 minutes Lisa needed to swap cards. If we'd been able to afford a newer model of the camera, with more slots, we could have gotten around this (the cards are "hot-swappable"), but the camera we have has only two slots, and one of them is for the proxy card, so the only way to swap cards is to stop recording, turn the camera off, pull the active card, insert a new card, and turn the camera back on. Lisa got pretty good at this, but it still takes a minute or so to do it.
The higher-quality MXF files will eventually be used to generate higher-quality versions of the recordings.
With approximately 30-40 minutes between technical time outs but 20 minute segment limitations (about which we didn't actually know) on the MP4 files, we ended up setting ourselves up for confusion on the first day.
As it happens, I had us take our first technical time out about twenty seconds short of twenty minutes into recording. Consequently, the first cartridge and the first segment were the same, and we still didn't realize that there was a potential problem.
By the time we got through card #3, the video team had figured out that the camera was splitting the video into pieces. The initial attempt to fix this required using the video-editing software to splice the two pieces together. But that's not the right solution; it takes almost as much time to do the editing and production as it did to shoot the video in the first place, and defeats the purpose of the quick-and-dirty uploads. Because of these delays, parts 3-5 of the Thursday meeting didn't get online until later that afternoon.
As the videography team came to grips with how the camera was behaving, they concluded that the best thing to do was to upload the segments "raw" as they came off the camera, rather than trying to do any of the time-consuming editing. This is why the pieces from Friday-onward are labeled 01a, 01b, and so forth. The 01 is the card on which it was recorded, and the a, b, c are the segment within that card.
Thanks to the team figuring out how the segments of the recording were coming off the camera, and thanks to the 20-minute segments being relatively quick to upload, we continued to run on roughly a 30-minute tape delay for the rest of the week. Every segment (except the last one, of course) was online before the meeting ended, with the final segment of each day being online soon thereafter.
Yesterday,
I'm merely surprised that nobody pointed out the missing 20 minutes before yesterday. Surely someone noticed the gap? Heck, I'm sort of surprised that Certain Paranoid Sociopaths didn't seize upon our honest mistake and declare that large amounts of Nefarious Things happened during the 20 minute gap.