Splice Now

When I helped start the editing program at the American Film Institute, the idea of teaching post production in an academic setting seemed a little nutty. But the idea that students would someday enter the program already familiar with digital tools? Had it occurred to us, we would have thought that was ridiculous.

Today, most film students enter graduate school with knowledge of several digital media applications, not just one, and Final Cut is usually among them. That’s partly because it’s cheap, easy to pirate, and you get the suite. But it’s also because it follows a drag-and-drop, desktop-publishing approach to editing. For young people, that makes the learning curve less steep. But it doesn’t necessarily provide the best toolset for professional editing. What I’m hearing from faculty at AFI and USC is that after a few months, most students end up preferring Media Composer. They like the precise trimming, the media management and the effects interface among other things. (Chris Hocking recently blogged about FCP vs. MC and came to some of the same conclusions.)

When Avid’s segment mode debuted in the early ’90s very few editors had ever touched Pagemaker or Quark, but there was still an internal debate in Tewksbury about whether drag and drop should be the foundation of the UI. The question comes down feedback. Every computer application has to supply feedback to the user, has to show you what you’ve done. The more responsive, fine-grained and intuitively presented that feedback is, the more control you have.

Imagine that as you typed in a word processor, the text arrived on the screen a second or two after you keyed it in. Even that small delay would drive you crazy, because it would interrupt the feedback loop. Regardless of your medium, if the controls are intuitive and feedback is fast and precise the interface seems to disappear, letting you think about creating and shaping the material rather than the machine itself.

Drag and drop offers good visual feedback, but it’s only telling you about the size and shape of little rectangles on the screen. I would argue that in editing, it’s more important to provide feedback about the film itself. You want to get the editor as close to the film as possible and permit him or her to make every editorial decision based on moving video. That’s why in the MC you see frame images in segment mode, why you trim with JKL, why you can slip and slide with JKL, as well.

An easy learning curve is important, sure, but it’s not equivalent to power, nor does it help you use the system all day in the trenches without fatigue. Fast and precise often means “some training required.” There’s a lot of overlap between FCP and MC — both give you JKL trimming, both let you drag and drop clips in the timeline. But the finesse with which they do it — the tightness of the feedback loop and the elegance of the controls — makes a big difference. There’s still plenty of room for improvement and each can learn from the other. Media Composer Version 3 included much faster timeline performance as recently as last year, something editors tend to notice almost instantly.

Avid has done a lot of internal work lately, and people are starting to notice. Apple will presumably hit back soon. I’m as eager as anybody to see what they have in store for FCS3, but while we wait for the Cupertino marketing juggernaut to ramp up it’s wise to remember that a good UI is many things, some of which are pretty subtle and hard to explain in marketing materials. It takes time in front of a system to find its power, and it takes many iterations to refine an interface.