We filmed several takes with two angles, and I put all of them into a Multicam clip, with a third “angle” for the album version of the song. Was it the most elegant idea? I dunno, I haven’t done this before. (That’s not true; I made music videos in high school, before video was digital. In fact, we were using cassettes instead of CDs to sing along to. The lip synching wasn’t very good because the boombox’s batteries would slowly die.)
Then I watched all the takes of Laser’s great face and marked my favorite performance bits. But what I couldn’t tell by looking at this collection of green ranges was whether the whole song was covered.
Also, if I were to dump all of my Favorites into a timeline at this point, it would be like listening to a song while intoxicated on the wrong amount of marijuana. You’d miss half a verse, and it would seem like it takes half an hour just for one song, and you’d wonder if you’d just said something embarrassing. Listen, I’m not good at THC.
I wanted to get all of my Favorites into the timeline, but in order of the song lyrics, so that I wouldn’t have to rearrange things back and forth. I decided to assign numbers to each line to make them unique. So I put the lyrics into my favorite text editor (Textmate), which shows me line numbers. Now that I think of it, pasting into a spreadsheet would’ve been just as easy. Go figure.
So, for every Favorite that included "the bandwidth is throttled," which is the first line of the song, I renamed the Favorite "01", and "I hate this dumb couch" Favorites were named "13", etc.
As is common with songs, some of the lines happened more than once. I got lucky here, because the video that Laser made for the projector on their face didn’t repeat any images, so instead of trying to memorize the nuances in the performance of three instances of “I’ve got nothing to fight with,” instance #2 appeared on my numbered lyric sheet as ”(hexagons) I've got nothing to fight with” and instance #3 appeared as “(fjord) i've got nothing to fight with”. I don’t know if the image in question is literally a fjord, but it’s more fjord than hexagon.
Compared to typical editing, this system seems backwards. Normally, I’d describe content with Keywords and then identify Favorites, but I’m glad I did it this way, because, for example, sometimes the lip synching was off (sorry for putting you on blast, Laser!*), or Camera B wasn’t focused, and since I didn’t mark those as Favorites, I never had to bother with them again.
I renamed the Favorites with their corresponding lyric numbers.
Then it was just selecting one lyric-number after another and Appending them to the timeline (I think there’s probably a way to get them to sort themselves numerically, but as I was trying to think of how, I felt like it would take the same amount of time to do it the dumb way), narrowing down the Favorites where duplicates existed, filling in any gaps, and tweaking a frame here and a frame there so that they actually appeared to be singing the song.
*Get it? Laser blast? Anyway, enjoy the video here: