Why Tapeless Isn't Painless

As someone who spends a great deal of his workday just organizing media that's already been digitized, I appreciated Terence Curren's assessment that tapeless workflows "blend [...] the worst parts of the film workflow with the worst parts of the videotape mindset." In other words: lots of footage to sort through, and you still have to marry the sound.

Nonlinear is the 'NL' in NLE. And that was such a leap forward from the flatbed days that it seemed random access in the recording format (allowing you to jump nonlinearly through clips) was going to be a similar game changer. The problem with the tapeless formats like P2 and XDCam is that they require a different skill set, one that even some assistant editors, who are the most organizational of the anal retentives, can't even get a handle on. Media named with strings of letters and numbers that are essentially meaningless is not for the human brain. It is for the machine brain. You need to know that going in.

Continue reading about why tapeless isn't painless...For example, I think it is painful to shoot a documentary tapeless if the documentary also makes use of archival tape-based footage. Why mess with two workflows when you could just have a tape-based workflow? And what about archiving your new footage afterward? Cloud storage for video is just a half-baked idea.

Well, assuming you are shooting P2 and don't have tons of cards, you probably won't be gathering too much footage. Most people seem to be just bringing the P2 in at full rez (i.e. DVCProHD resolution) and editing with that. Bene! Great!

But let's say you're dealing with the .r3d files from a RED camera. Editing real-time in 4k is not practical or even yet realistic. So you'll have to manage your files in a way that your proxies will relink to the full rez when you're ready to online.

Some editors (and I mean you, Doug) are paralyzed if they can't rename the files descriptively. The key, I think, is to think in terms of metadata. (What is metadata ? It's just information about files -- what they are, when they were created, what the legal restrictions on their use are, etc. -- Data about data.) Filename is no longer also the description. The filename is just one piece of metadata.

When we have more files than can be managed by a single human brain, we use machine-readable codes. You know, like barcodes. Think of the weird string of letters and numbers that names that P2 clip as a barcode for your computer.

With machine codes, we can quickly skip through a database. In the first all-tapeless movie, Zodiac, the post team used Final Cut Pro's XML abilities and Filemaker Pro's XML abilities to have Filemaker manage Final Cut's media. Since then, Apple introduced Final Cut Server (which can also work with Avid media, by the way, though not as seamlessly).

Final Cut Server is just one recent attempt to get all this metadata together and synchronized. I'm told in its present state that it's good for small companies, but it doesn't scale well. There's a much less user-friendly program that's getting some traction in LA, the poorly-named CatDV, which is more flexible and, because a producer's paper cut can be exported as an EDL, really aids with reality TV (where you spend most of your time 'frankenbiting' -- editing to make it seem like people uttered soundbites they simply didn't).

Producers doing the work of editors -- it's a crazy mixed-up world. What hasn't changed is the essential nature of editing: A collection of clips that adds up to more than the sum of its parts. Tapeless or not, the process will never be painless.

Photo CC-licensed by zimpenfish.