It's hard to talk about software in 2012 without mentioning "the cloud". You've seen the ads from sofware vendors asking what your cloud strategy is? With all this hype, it's worth taking a look at the state of cloud video editing and the promise it could hold for Professional Video Editors.
First the dream. Imagine having instant access to all of your material without having to plug in a hard drive. Even better if you could access it on any computer, from anywhere in the world, with the ability to easily collaborate with others, right? No more backups, because this would done securely, automatically, and redundantly. How about transparently using clusters of computers to instantly render video, output it into multiple formats, and deliver it as easily next door as the other side of the world. All of this without having to invest in racks of equipment or worrying about upgrades.
Sounds like a game changer right?
While cloud video editing is improving day by day, I am not suggesting that the current state of cloud based editing even remotely approaches the offerings of professional desktop applications like Final Cut Pro, Media Composer, or Premiere Pro. There are some significant barriers that must be overcome before pro-video editing can move into the cloud. So, rather than compare apple and oranges, I think it needs to be seen for what it is – a promising beginning.
Cloud video editing could be a good fit for consumers looking to quickly create and share videos with minimal effort. Yet for video professionals, the story is more complicated. A fully web-based service could work well for some users, while hybrid solutions that blend a desktop program with online collaboration features may work better for others.
At the end of the day, the real answer to whether editing in the cloud makes sense depends on the particulars of your workflow... But for now, lets take a look at where things are at, and where they may be headed.
Cloud Editing Has Been Pioneered in the Consumer Market
The idea of moving video editing into the cloud isn't a new one. However, until recently, most of the activity was geared at the consumer market.
Yahoo! acquired JumpCut, one of the earliest attempts at online editing, back in 2006. It dropped the service in 2009. This was less to due with the quality of the product (it did have a loyal following), and more to do Yahoo's history of failed acquisitions.
Since then, others have sprung up in JumpCut's place and are pushing the ball forward. Magisto promises to automatically edit iPhone videos in a couple of clicks. Google offers the bare bones YouTube "Video Editor", and Animoto now allows users to upload video that can be automatically cut together with photos and set to music.
The general approach in this market is to make video editing and distribution faster and easier than it is with desktop applications. The name of the game is push button editing, so there isn't much that will bubble up into the professional markets.
The Challenges of Bringing Cloud Video Editing to Pros
Much of the experimentation with online editing has taken place in the consumer market because projects are relatively simple and because it has a different set of value drivers than the professional market.
While consumers are willing to trade off features for ease of use, professionals, in most cases, can't make the same trade off because editing is their livelihood. Witness the reaction to the perceived oversimplification and missing features of Final Cut Pro X.
Web-based editing software for high-end editors would need to offer all the power and flexibility of the best of breed desktop software, while upping the ante with the benefits of online storage, encoding, and collaboration features. This is a very tall order.
Aside from the immense challenge of duplicating the functionality and flexibility of desktop software on the web, there are several other major challenges to making online editing software work.
The Upload Bottleneck
The biggest barrier is probably the uploading bottleneck. If you are going to edit online and truly take advantage of using multiple computers to speed rendering, then your source material needs to be uploaded into the cloud. Due to the large file sizes created by pro video codecs, upload times could be prohibitive for all but the simplest of projects. Ingesting footage already takes long enough without having to wait for hours for it to upload.
Many smaller production companies are using cable Internet or ADSL connections that offer fast download speeds, but have upload speeds in the 1-2 Mbps range. At these speeds, it would take over an hour per GB of footage uploaded. All this before getting down to work.
Are upload speeds going to increase on a $/mo. basis to a great enough extent to support full cloud editing? Not anytime soon. In fact, the challenge of upload speeds will be compounded as people move to 2K and 4K workflows.
While large production companies and post-facilities have enterprise Internet connections with speeds that can reach into the gigabit per second range, they may be slower to move to new software and workflows. Why? Because they are the ones that have the biggest investments in shared storage and custom collaboration software. They typically work on the largest and most complex projects around. The decision to try an untested workflow in these environments wouldn't be taken lightly.
A Flaky Internet Connection Can Bring Your Business is Down
A challenge associated with moving to cloud services is the question of what to do when you have problems with you Internet connection.
Most consumer and small business Internet services are best effort, meaning that disruptions and outages do occur. Internet service providers rarely offer SLAs to anyone but their enterprise clients.
For many non-core software applications, downtime due to an Internet connection isn't a huge issue. People can either tether via smartphones for desktop email in the event of an outage. Sending invoices to clients can wait a day if things aren't working. Online task lists can be updated later.
However, if you are an editor and you can't access your system because of a glitch with your Internet connection, this is lost money. There are backup options like using a smartphone as a modem, working in a cafe with WiFi (you can edit from anywhere), or investing in multiple Internet connections. However, these options are potentially costly and may not be practical for everyone.
The bottom line is that care must be taken before moving the core of your business online.
Proxy-Files and Encoding
If editing is going to occur in the cloud, then the workflow will likely have to use proxy files to increase responsiveness and accommodate differences in Internet connection speeds. This introduces a couple of wrinkles — quality and cost.
Using lower resolution proxy files is essentially a step backwards compared to today's workflows where editors work directly with native files and formats. Things like advanced color correction would become more of a multi-step workflow, negating the benefits of the cloud.
The second aspect to using proxy files is cost. At 5-10 cents per minute, the cost of processing video for a large project could add up. While this cost would be a minuscule part of a production budget, as discussed in more detail below, this variable cost means that a different type of business model is required for a web video editing service.
A big opportunity with a cloud based workflow is the potential to move files off local hard drives and eliminate the mess of backup and archiving systems that most of us use, potentially with rich metadata for advanced categorization and retrieval.
However, while online storage can be virtually limitless, it certainly isn't free. For example, Amazon Web Services, the infrastructure provider that powers many of the top web applications, charges 5-10 cents per GB of data stored per month. At these rates, 1 TB of data is $50-$100 per month.
While online storage is redundantly backed up, highly durable, and accessible from anywhere, it makes archiving to LTO seem relatively cheap.
Given these costs, online storage really only make sense for work in progress, finished work, and archival footage that has a high commercial value. Source footage may be cost prohibitive to archive in this manner.
New Pricing Models Needed
Encoding and storage costs highlight the fact that cloud video editing software needs a different pricing model than traditional one-time licensing.
In traditional licensing, the vendor has next to no incremental cost for each additional copy of the software sold. Ongoing usage of the software does not generate costs (aside from things like support) for the vendor.
In a cloud model, the service provider's costs are recurring. The cost of things like bandwidth, encoding, and storage increases in proportion to the number of customers and to the intensity of their usage. These costs mean that subscription based pricing with tiers based on usage makes more sense than a one time charge.
While the total cost of ownership with cloud video editing could ultimately be lower than with shrink wrapped software due to things like hardware savings, it would be a paradigm shift that people would have to wrap their heads around. As we've seen with the reaction to Adobe's launch of subscription based pricing, this could be a challenge.
Cloud Editing Applications With Sights on Pros
In spite of the challenges listed above, both Novacut and WeVideo have set their sites on meeting more than just the needs of hobbyists. Novacut is going after DSLR filmmakers, and WeVideo is going after everyone from consumers to event videographers, marketing departments, and online news organizations.
Novacut is an open source project that used Kickstarter to raise funding to develop a real-time collaborative video editor. The product, which is still in development, is being designed from the ground up to meet the needs of DSLR filmmakers. Laser like focus on this passionate and price conscious group makes sense in terms of keeping the feature set manageable.
One of the things that I find most interesting about Novacut is the approach it takes to combining local computers and the cloud. Each user will have to download and install the editing application, and they will work with the project files locally. Novacut will use the cloud to sync meta-data changes that describe the edit between collaborators. Essentially it's a synced edit decision list.
The benefit of this approach is that syncing edits will be instant and will take very little bandwidth once the media files are distributed. Since at this point, the files reside on each participant's hard drive, editors will be able to work directly with high quality source footage rather than the low-res proxy files required by a fully online workflow.
The downside to this approach is that each user needs their own copy of the project media. With remote collaborators, this means that files will either have to be uploaded or shared with a hard drive. To speed uploads, Novacut will offer an option where users can exchange low-res proxy files, but this has its own set of challenges.
Other downsides to Novacut's approach are that it will require software installation (Ubuntu will be the first operating system, with Mac support following later), distributed media management could get complicated, and since the media resides locally, the power of cloud computing won't be leveraged to speed up tasks like video rendering.
A further challenge may be keeping the momentum of the open source project going. EditShare's Lightworks Open Source editing project means that there may be competition for talented developers.
To support development and reward the hard work of the people that are bringing Novacut to life, I expect them to develop a business model like Kaltura or other open source projects. In this model, users would be given the choice between downloading the open source software and managing their own Amazon server instances, or paying a monthly fee for a turn-key version that has customer support.
WeVideo introduced a cloud-based collaborative video application in the fall of 2011, and subsequently signed a deal to work with Google to offer a free version for YouTube customers. Since then, the company has rolled out a paid version that is targeted to consumers, businesses, bloggers, and journalists.
Unlike Novacut, WeVideo's service is completely browser-based, meaning that editing can be done on any computer and that it can be accessed on mobile devices and tablets. Since it is fully web-based, all source video must be uploaded to WeVideo's servers. Once this is done, the video is encoded into a proxy-file format that makes it faster to work with over the web.
Since WeVideo recently launched, I thought it would be a good idea to take it for a spin. I signed up and had the application up and running fairly quickly. I shot some video on my iPhone, uploaded it to my new account and cut 3 shots together to create a short scene. It was intuitive, responsive, easy-to-use. The free version that I tested had minimal output options (360p, dicey but passable quality for web video).
Paid versions from $7-$80 / month will get you higher quality outputs, local downloads, more storage, and priority support. An enterprise version (pricing is only available on request) provides more granular control over how people participate in a project (Administrators, editors, reviewers, and contributors), and it allows the service to be branded and tightly integrated into other products and services.
The main strength of WeVideo's approach is that it makes it easier for different people to work together on a project. In the paid versions of the service, participants can upload media, view videos, and edit project timelines. The service also makes it easy to push videos to social media sites like YouTube, Vimeo, Twitter, and facebook. Render times for video are also fairly good, as a local machine isn't tied up while videos are processed.
For many video professional editors, there are several downsides to the service. The first of these is that footage must be uploaded before editing can begin. This weak link is compounded by the fact that the maximum upload file size that WeVideo supports is 500 MB. While this likely helps with the reliability of uploads, it could be a problem with HD source footage that would likely have to be cut down into smaller clips prior to uploading.
A not so readily apparent issue is that WeVideo doesn't allow multiple parties to edit the same timeline. Each person on a project essentially gets their own version of the timeline to work with. While this helps prevent conflicts, there doesn't appear to be an easy way to merge the different timelines together down the line.
The other issues come down to control and flexibility. While WeVideo offers 30 built in effects, that may not be enough for many professional editors. Likewise, color correction is pretty basic, titling is relatively simple, the application lacks things like chroma keying, and export options are limited to the preselected services and quality settings.
As mentioned at the beginning, the practicality of WeVideo or other cloud video solutions will largely depend on your individual workflow and output requirements. If you want to take a deeper look at WeVideo, here is a really detailed review.
Professional Applications Where Cloud Editing Currently Makes Sense
So where would the early cloud video editing programs make sense? Here are some of the project characteristics that come to mind:
- Simple productions where speed is more important than custom effects
- Videos are relatively short
- Digital source material
- Multiple people need to contribute source footage
- Multiple people need to review videos and provide feedback
- Remote access from different locations is required
- Video distribution is going to be online
- High speed Internet connections are available.
This might be the a fit for video bloggers who are interested in telling a story quickly rather than spending too much time on unique transitions and custom graphics. Other applications would be people covering trade shows and events where footage is contributed by many people and then cut together quickly for immediate online consumption.
In fact, WeVideo offers up TechZulu as a case study of an independent news organization that used the service to produce professional, high-definition coverage at this years SXSW festival.
I had my entire team submit their video interviews and coverage into our WeVideo accounts. No matter where I was or how fast of an Internet connection I had, I was able to edit all our footage and export it directly to our YouTube account at 720p HD resolution. Not only did I not have to sit and wait for my machine to compress and export our footage, but being able to continue editing while one job was being exported in the cloud with WeVideo saved me hours of precious editorial time.
— Efren Toscano, TechZulu founder and CEO
The main caution in this use case, is that you might want to arrange for a wired Internet connection to mitigate risk of notoriously spotty convention center WiFI (I've witnessed a parade of fellow geeks searching desperately for power and Internet connections at South by Southwest).
A Hybrid Solution - The Cloud Doesn't Have to be All or Nothing
For most professional editors, it doesn't look like the whole edit suite is going to be moved into the cloud anytime soon. It may never be the perfect fit.
A less black and white option is to keep the editing software on the desktop and use web services to address pain points in different parts of post-production workflow. Right now that means using several different services. Over time, desktop editing applications will likely become more tightly integrated with online services to give editors the best of both worlds. So what areas make sense to move online?
Video collaboration for review and approval. A big pain point in the production process is sharing video with people outside the edit suite. Web service like ScreenLight provide simple video collaboration, review and approval over the Internet regardless of where people are or what device they are viewing it on. A natural evolution would be to integrate the service with editing software so that you could send a clip out for review directly from the timeline.
Encoding into multiple formats. Sending batch encoding jobs (whether it be a large number of files, or files that need to be output into multiple formats) to a web service makes sense, as all the files can be encoded simultaneously. Using a service like Zencoder would free up local machines for more productive uses. Upload times may still be an issue, but in this case it's just the final files that are uploaded rather than all of source footage.
Video distribution. The opportunity here goes beyond distribution to social media sites. A web service could be used to replace tape delivery of broadcast material. This type of service could encode video according to the unique needs of each broadcaster and deliver the downloads in a secure way that can be audited.
Archiving video files. The cloud can be a good fit for this, since lengthy upload times are less of an issue than with online editing. It's an application that could be done in the background, or if the files are too large, a hard drive could be shipped to a depot for direct transfer to a cloud storage service. The main issue with this application is cost.
The bottom line, is that the there are plenty of opportunities for the Avid, Adobe and Apple to integrate web services rather than moving the full editing applications to the web.
Cloud video editing services aren't going to spell doom for Media Composer, Final Cut Pro and Premiere Pro anytime soon. That said, upstarts like NovaCut and WeVideo highlight the potential benefits that can be had by shifting parts of the workflow online and they signal the direction that the rest of the industry will move towards.
Does cloud editing makes sense for pro video is a bit of a loaded question. Professional video editors aren't some monolithic group, and no software vendor is going to satisfy all segments of the market. Each vendor builds its product with a particular target segment in mind. So the answer is whether it makes sense to you on an individual basis.
If you were at all concerned that Final Cut abandoned the Pro market, then the current cloud editing products are not for you. However, if you live and work online under tight timelines, then it might be worth checking out further.
2012/09/24 - For an update on the potential of cloud video editing, check out our recent post on Adobe Anywhwere.