[David Lawrence]

Hey All,

This is a super interesting conversation! How about we move it to it’s own topic? I feel like there’s a lot of rich ground we’re just starting to cover and we’re way way off this thread’s original subject. Wouldn’t want people to miss out!

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Chris Harlan] “Again, constructing a trailer or promo is, for me, very much like scoring. Not just music and SFX, but speech patterns and rhythms, VO breaks and bridges. All very musical. David, we definitely think alike in our editorial approaches.”

Chris — you, me and thousands of others in our industry. Think about the branding for Avid’s flagship NLEs… They’re named that way for a reason 🙂

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

Great post, Jeremy. I appreciate the thoughtful reply.

I think I finally understand why you and I have such a different POV on the value of tracks:

[Jeremy Garchow] “I don’t think a magnetic timeline would make sense when working on the audio of a video based project as there’s no reason for the timeline to fluctuate that much during a locked edit.

I think video editing and audio post are two totally different processes and therefore might require different interfaces, but if the software is smart enough, I don’t see why tracks are totally necessary. Really, I’d leave that to the audio folks. “

That’s our key difference, right there. I think of video editing and audio post as one and the same.

Even though my primary work involves video post-production services, I have a background in music and sound design and sound is the foundation for everything I post, whether a music video or documentary style interview. In effect, I use the NLE as both an NLE and a DAW in a fluid, back and forth process to create rhythm, pacing and beats with both picture and sound interchangeably. For me, it’s all the same during the editorial phase.

Now I know your thoughts on FCP7’s audio capabilities 😉 and for the most part I agree. But here’s an example —

On the project I just finished, I posted a four and a half minute segment cut from a ~1 hour interview with a single subject. The speaker is interesting and engaging, but his speaking style is full of pauses and tangential thoughts. His final ~4 minute actuality has close to 300 edits. I craft the pauses, add or pull breaths and even construct entirely new sentences to make him succinct and editorially on point. It sounds perfect. You accept it as his actual speech without question. This is pretty typical of the work I do.

I usually go thru this audio workflow before I even begin to think about picture. My efficiently is high because I checkerboard and use FCP7s timeline tools with surgical precision. Often, the difference of a single frame will make or break the transparency of a constructed acutality. If I need even more precision, I use volume curves to get sub-frame accuracy, or two frame audio dissolves with start, center, or end positioning. This gets me about 80-90% done with audio. I don’t need the real-time effects at this point. After locking picture, I round trip to STP to do any final tweaks, sweetening and mix. Everything is organized because I’ve been organizing track layout as I go. It’s fast and efficient because I’ve already done most of the work in FCP.

This is why I need tracks — they are my scaffolding. I build on top of them and need them to be stable as I edit. Working on the trackless magnetic timeline (especially with audio) with everything rippling and moving up and down by itself whether I want it to or not, it feels like trying to build a house on quicksand!

For me, it’s a very musical process, like scoring a symphony. That’s why when I look at something like Shawn Federline’s timeline, I don’t see clutter, I see a musical score built on a common language that can easily be read, understood and shared across a range of applications and disciplines. And with audio especially, track arrangements, waveforms and clip patterns often make it possible to identify audio type instantly, just by looking. This at-a-glance feedback is a huge timesaver when sharing projects.

On the other hand, if I wasn’t thinking in terms of audio, I can easily imagine why tracks might not seem so essential and even appear as clutter. As long as playback is correct and you can output what you need, what difference would the vertical positioning make? It wouldn’t. Makes sense.

I think I finally get where you’re coming from on tracks. Thank you! I hope this gives you a better understanding of my skepticism and problems with the trackless, ripple-only timeline.

[Jeremy Garchow] “FCPX still needs work, obviously, but under the surface I really appreciate what is trying to be done, and that is catalog and manage a massive amount of data, have it follow you through the post process, and then use it to your advantage on output. There is not one NLE that I know of on the market today that is attempting this inside the editor.”

On this point we’re in complete agreement. The database driven, metadata engine under the hood is impressive and filled with potential. It’s why I’m investigating FCPX as a DAM even if I don’t ultimately use it to cut.

That’s the most frustrating thing about this whole debacle. All that power under the hood and no clear way for editors who work like me to take advantage of it. I hope Apple will address our needs in future updates but I’m not counting on it. Time will tell.

Anyway, thanks for an interesting conversation!

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Jeremy Garchow] “I guess I can’t explain well enough for people who haven’t tried it to understand. “

You’re doing a fine job explaining, and I think we get it. The problem is roles are a work-in-progress, so whether or not they wind up being more efficient in scenarios like Herb’s depends on features we don’t even know will be implemented. What we do know is that tracks have worked well as a common language for representing multiple channels of related, time-based information for decades. So the burden is on Apple to demonstrate that their new paradigm is actually better.

Let me ask you this – assuming you’re correct and a trackless, metadata driven, timeline paradigm is actually a better UI for multi-channel, time-based object manipulation, how long do you think it will take for the audio industry to adopt it as a standard for DAWs? Do you expect the next release of Logic X will be trackless?

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

Interesting sub-thread, really enjoying the ideas here.

[Jeremy Garchow] ”
You don’t see any use for a role based vertically sortable timeline? Meaning for any given primary storyline clips and everything connected to it, if you want your audio to always stack as dialogue, then effects, then music? I can think of some really practical reasons, especially when there’s no tracks. “

[Jeremy Garchow] “And it’s not there yet. Roles are just the beginning the jumping off point, of how a database driven design will help manage data.”

Agree that roles have a lot of potential, especially for data management. What I don’t see is how they make it easier to edit. Cutting and data management are two very different tasks. They may be related, but the actual work and the tools I need to be efficient are completely different. One of the thing I need is structured visual organization, something that tracks intrinsically provide.

It might eventually be possible to assign a lane index to roles to simulate tracks, but I think things things will start getting messy fast. I’d be happier if the limit on a single primary storyline went away and became user definable. Let me set up my sequence to fit my organization needs, then add data management tools to that.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Jeremy Garchow] “This means no matter how the timeline is sorted vertically that it will play as intended without the ‘top layer’ being necessarily on top, however the timeline is sorted visually. Time (I think) would be the only constant of the visual timeline (or I guess it doesn’t have to be with the timeline index). Why else would there be a transfer mode that layers clips with alpha channel information “on top” when the alpha clip is actually underneath the clip you are trying to composite over? Down is now up. It might also explain why horizontal clip relationships are now user denied instead of track defined.”

This sounds completely unusable to me. Maybe something like this makes sense in a nodal compositing system but I don’t see any value in a time-based system. What would the benefit be? How would editors be able to share projects and have any idea what they’re looking at?

[Jeremy Garchow] “All of that is pretty damn complex, and also does not resemble any Apple app that I am aware of. It also attempts to explain a trackless timeline, and the parent/child relationship in the new XML. If all of this is what is ultimately intended by Apple there is probably really good reason that Apple did not unleash this level of complexity at once.”

I agree that it’s complex. I think that’s one of the main reasons it’s difficult to use. But I’m not so sure there’s a grand yet-to-be-revealed plan. A simpler explanation is that it’s just bad design. Remember, Apple doesn’t always hit home runs (Ping, MobileMe, hockey-puck shaped mouse, etc.) Keep in mind the team that brought us this UI is now apparently having heated debates on whether persistent in/out points are a good idea.

I mean seriously — they’re debating the value of persistent in/out points in a professional NLE… and we’re supposed to believe they’ve got a well-thought out plan for this thing? Seems like quite a stretch to me. I’d love to be proven wrong.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Lemur Hayop] “Just when I was considering buying PEFCP to export FCPX OMF to Pro Tools. We’ll see what happens.”

I’m not following. Was there some news today?

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Mark Morache] “I think it was just something about the project and/or event files that went pooch. “

Since these files are database, it’s easy to imagine that some kind of database corruption happened when you lost the drive. Some kind of autosave vault sounds like a good idea.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Craig Seeman] “Import folder as keyword collections (bins) or create empty keyword collections and drop the clips in and they’re keyworded.

JKL and use Markers with comments as part of the marker name. Double tap M and add your name/note.

drop in timeline and start cutting.

Your Marker names are visible and searchable in the timeline index”

I’m not trying to exactly duplicate my FCP7 logging workflow, I’m more interested in the parts of FCPX that seem better. I can quickly skim/JKL in the event browser, then hit I and O to set a range and F to mark the range as a favorite. Then using list view, I enter a comment in the notes column. This is a fast and efficient workflow for logging and works great for me. At that point, I would want to move to a more flexible editorial environment to cut.

FCPX doesn’t have to be an all-in-one solution to be useful. If there’s a clean way to get work out, then it can find a place in many workflows as part of a larger set of tools. Mix and match the tools using just what you need to fit your style and maximize your efficiency.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Steve Connor] “What do you need keywords to be able to do?”

Here’s an example –

My typical FCP7 logging workflow is:

1) import clips to appropriate bins
2) skim/JKL thru each interview or dialogue clip in browser adding color-coded markers with log comments
3) drop marked clips on timeline and begin cutting.

I can imagine a FCPX preflight workflow something like this:

1) import clips to event
2) separate interviews/dialogue and b-roll with keywords
3) skim/JKL thru each interview or dialogue clip marking favorite clip ranges with log comments

Then I’d just need a translation utility that would covert keyword groupings into bins and favorite clip ranges into markers with duration and export as XML so I can cut with tools I’m faster and more comfortable with.

This would be especially useful in field-based and quick turn-around situations, but I can also imagine using it as a general logging scenario. I think it plays to FCPX’s current strengths and benefits.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl