Jedidiah,

Jason (lead developer) here. This is a long answer, but this question keeps coming up all over the place.

We love PiTiVi and it's what we currently edit our own videos with. And to be clear, we're *reusing* what the vast majority of the PiTiVi development time has been spent on... GStreamer and GNonLin. By using the same multimedia backend as PiTiVi, any improvements we make there benefit PiTiVi users too. The multimedia backend is where most of the work is for *any* video editor. If we had to write our own backend, we wouldn't attempt this project! Too much work!

However, we're focused on *very* different use-cases. PiTiVi is designed to be a simple video editor aimed at home users, which we think it excels at. Novacut is designed for professional TV and movie production.

Our gold standard is the over-the-shoulder two-camera shot used for dialog between two characters. This shot is used over and *over* again in TV because, well, it's just good storytelling. This shot has two cameras filming simultaneously, while audio is recorded on a third device. A boom mike is held above each actor, and each actor has their own audio channel. When editing, the audio and two cameras all have to be synced up, and then the editor switches back and forth between cameras as the dialog progresses from one character to the other. For each take, this shot producing two video files (one from each camera) and an audio file. And remember, a single TV episode will have dozens of scenes, sometimes dozens of shots per scene, and dozens of takes per shot. That's a *lot* of data and complexity to deal with. But this is the use-case editors will spend most of their time on. We want to make their lives easier, so our priority with Novacut is to make this use-case as fast and pleasant as possible.

We're going to make Novacut as easy to use as possible, but by definition it will always be more complicated than PiTiVi because it's dealing with a more complicated use-case. We strongly feel that these two drastically different uses-cases need drastically different user interfaces, otherwise you end up with a confused compromise that does't serve either use-case well. Although I haven't yet talked to Edward Hervey (the PiTiVi and GNonLin developer) about this yet, I'm quite certain he would agree that this is the right approach.

So we're not re-inventing the wheel. We're re-using existing open-source components wherever possible. All the really hard work has been done for us already, for which we are very thankful. At the same time, I don't know of another editor that has the same use-case priorities that we have. Different video editors *should* have different priorities because there are distinct use-cases worth optimizing for.

And one place we *certainly* can't be accused of re-inventing the wheel is the distributed workflow... you can't re-invent a wheel that hasn't been invented yet! No video editor, free or proprietary, has yet tackled this problem. It will take a lot of trial and error to get this right, so we need a simple architecture designed from the ground up to be distributed. The UI is also unknown territory, needs to communicate things no other video editor has needed. So by starting with a blank slate (in this limited respect), it will be far easier for us to take quick evolutionary steps toward the right user experience, the right semantics.

So in summary, we're really lazy and wont be doing much! GStreamer is our savior for the multimedia backend, and CouchDB is our savior for the distributed features. We're just writing glue code and focusing on the UI.

Jedidiah, you have a good point, but it is a mistake to assume any new Free Software project will reinvent the wheel. Yes, there are some solid free video editors out there. Will Novacut leverage the best of existing Free Software? Absolutely. Novacut will take the best of what is available, build it into a distributed architecture harnessing the benefits of everything from collaborative worklfow to computation and storage in the cloud.