One of our goals here at the Tribune is to give you a better idea of the thought and creative processes behind our journalism. We debuted the Stump Interrupted series this week and I've received some emails about how the work is done. The short answer=not quickly. Giving the Stump treatment to an individual speech takes dozens of man hours and at least three professional design programs.
We'll start at the beginning. When we shoot with Stump Interrupted in mind, it means rolling for the entirety of a speech and getting cutaway shots later. We frame the shot with enough room around the subject so that pop up bubbles or other motion graphics can appear. Those of you who work in this kind of medium know that crisp, clean audio is critical, so we do our best to get our mics as close to the subject as we can.
After the reporters involved (in the KBH case, it was myself and Morgan Smith) begin marking up the speech and researching the claims for bubble content, the art gets underway.
My friend Todd, who is the man behind the bubbles, is responsible for bubble-making and bubble-animating. When we started, we created two standard bubble designs (rectangular and square) in Adobe Illustrator. Two sizes gave us options depending on what the inside text would read. We also spent about two hours making sure the bubble typeface was just right, and that the off-white background color of the bubbles matched our Texas Tribune style. At the end of the typeface decision summit, I could barely tell the difference between "Tahoma" and "Myriad Pro".)
The process of animating the bubbles happens in another program - Adobe After Effects. I could lie and say I have passing knowledge of the richness of that motion graphics software, but I won't. That's all Todd. Each individual bubble (in the KBH case, the count was 37) is animated into the video, and then paired with the mouthpop bubble sound. Because After Effects is not a video program, like say, Final Cut Pro, we couldn't watch the bubbles and the video together for the entirety of the speech unless the project was finished and then rendered out into a movie file.
The rendering takes about four hours. That means it took half a day to make any individual changes and get to see them play out in the video.
The third program we deployed was the reliable video editing software, Final Cut Pro. Final Cut comes in occassionally with sound design (if it wasn't already accomplished in After Effects) and it's used to be able to watch and mess with the final product, i.e., add the TT watermark, lower thirds, etc. We originally tried to do more with Final Cut and less with After Effects, but using After Effects for the bulk of the work proved more efficient.
Not that any of this is really efficient at all. So thank you for your great feedback and questions. We'll continue to work on Stump as we can get to various candidate speeches.