Commentary

Cloud-Based Video Editing And Collaboration

Digital Nonlinear Editing (DNLE) systems have existed for more than 20 years as hardware/software alternatives to dedicated standalone silos. The progression of DNLEs is straightforward: first, DNLEs replaced dedicated tape-based editing systems in standalone configurations. Next, DNLEs assumed LAN-based sharing capabilities, enabling a limited number of editing stations to share content. Simultaneously, compressed picture quality (1989 at 250:1 software JPEG compression) improved and, ultimately, DNLEs were able to work in uncompressed video (and film) resolutions.

Flash-forward to 2009. DNLEs exist in standalone and sharing configurations. The number of DNLEs that can be networked in a sharing environment can easily reach 100-200 and can perform real-time operations on X number of tracks. The more dedicated the DNLE is, and connected to DAS, Striped-DAS, High-Performance SAN, etc., the higher the number of tracks that can be played back in real-time (RT) without rendering. Typically, for a shared environment (that is, when we are talking about 50, 100, or even 200 workstations), the RT playback will be 2-4 video tracks and 4-8 audio tracks. Naturally, as the number of workstations decreases, the number of simultaneous playback tracks increase. Again, this number varies depending upon storage, the network switch (2-10 gigabit), resolution, frame rate, and number of seats.

Today, both the offline process and online process still exist. It is also, of course, quite normal to digitize content at the highest resolution required for finishing and both the offline/online process is combined into a single stage -- and thus, there is no redigitizing content at the finished resolution.

Again, this past and present form of digitization, manipulation (editing, compositing, etc.), and delivery only form a prologue for what is to come. Let's consider some of the problems and opportunities that exist in the world of creating software for the broadcast, media, and entertainment industry (we'll abbreviate this BME). First, while DNLEs exist in all-software models, almost all of them operate where content is either directly attached or available locally. This, naturally, either limits operational distance, or requires that content be moved from the origination point (acquisition point or digitizing point) to the editing point.

But the next evolution of the DNLE as well as the collaboration experience will be encountered as a result of what I term: editing on the wire. This kind of editing hasn't been possible for a number of reasons, among them: the focus of DNLE manufacturers, network connectivity issues, and quality of experience. This represents the first bona fide opportunity.

What is it and what does it represent for the BME industry? Specifically, editing on the wire refers to the desire to be connected to the media even if it is remote to the user. Content, streamed to the user, in the resolution that the user's connection can reliably support, means many things: first, staff can work remotely and work on media remotely. Content doesn't move to the user. Instead, the user logs into an application session -- which, of course, can easily be serviced from the cloud -- and is presented with a complete, browser-based editorial system.

In the corporate high-tech world, employees work virtually from their home office every day. In the BME world, the only workers who work from home are typically freelancers who have their own systems and redundant copies of content. But, being connected to the cloud and editing on the wire is the most natural evolution that we are about to experience.

Imagine the university that is faced with having to buy 25 more computer workstations to upgrade the digital media lab. This type of infrastructure-heavy, hardware-based purchase may no longer be necessary. Instead, that university can utilize a private or public cloud that provides 25 virtual, browser-based editing suites. Students can utilize their Macs or their PCs, and can either come into the lab or can work from anywhere in the world as long as they have an Internet connection.

Editing on the wire is the first step to the next fascinating decade of the DNLE evolution.

8 comments about "Cloud-Based Video Editing And Collaboration ".
Check to receive email when comments are posted.
  1. Rob Mottola from east2west media group, inc., August 27, 2009 at 11:22 a.m.

    this is huge for any editor - being able to access footage or elements that are hi res just might might take a big bite out of fed ex...who hasn't had to either wait til the next day to finish a cut, or been calling fed ex for tracking info for a non delivered tape!

  2. Deborah Brozina from Making Change, August 27, 2009 at 11:55 a.m.

    Except that accessing hi res footage to edit a true master isn't really possible over large distances. Right now, 4K is the top standard to edit a master. However, this is very challenging technically. What is being suggested is possible, if you wanted to go back to the model of a offline then online edit. Or, you give up having the highest quality of master produced.

  3. Stanford Crane from NewGuard Entertainment Corp, August 27, 2009 at 1:25 p.m.

    unfortunately, the digital switches and routers are stuck in the '90's with their interminably slow connections to the backplane, which is of course massively limiting real bandwidth, not specified bandwidth, so blame John Chambers for not wanting more bandwidth, then he figures he'd sell less switches, but I think he'd actually sell more. at present the AdvancedTCA is a joke, and should be renamed Antiquated Telecom Computer Architecture. I ought to know, I invented it.

  4. Jim Kaskade from eyespot Corporation, August 27, 2009 at 8:14 p.m.

    Couldn't agree with Tom more. The idea of using the Internet for the presentation layer only...and making the content available from a centralized source which does all the heavy lifting in terms of manipulation...expands on the notion that the Internet IS the operating system, and the Cloud replaces the traditional server farm with on-demand and configurable virtual machines.

    Where the art comes in....can you make light web applications which provide the same functionality and user experience as traditional media tools?

    I think so...

  5. rockstar babu, September 1, 2009 at 4:16 a.m.

    hi every body this is really very nice and useful for every one thanks for share it and good luck
    .........
    Rockstarbabu
    ----------------
    <a href="http://blogshubhinetwork.com">SEO</a>--SEO

  6. Pete Burns from Forbidden Technologies plc, January 14, 2010 at 4:38 a.m.

    Forbidden Technologies in the UK is already doing this. FORscene is a Cloud based video platform. It uses a proprietary video codec to deliver individual frames of video (ensuring accuracy) and works over any broadband connections.

  7. Jonathan Mirow from BroadbandVideo, Inc., January 15, 2010 at 1:33 p.m.

    Anybody who has actually used the public internet "cloud" for video applications knows that this is pie in the sky technononense. The backbone of the internet is set up to handle HTTP transfer - and does that quite well. Can any of you out there sustain a 500k connection for two hours without a hiccup? I think not - so this Flash Gordon stuff about colaborative HD non-linear editing using a broswer application is hoodoo voodoo. Don't ask me if I want to see the demo.

  8. Stephen Streater from Forbidden Technologies plc, March 2, 2010 at 10:56 a.m.

    Well written Cloud software glosses over temporary Internet glitches without users even being aware of them.

    Our customers seem quite happy to use the public Internet to run our video Cloud application. It is used for making television and professional Web videos around the world. We have around 10,000 hours a week going through our FORscene system and it has handled around 1,000,000 hours of professionally shot video so far.

    The software runs as a Java applet or (Java webstart) through a web browser over HTTP, which the Internet is perfectly capable of supporting.

Next story loading loading..