By Fred Dawson
May 23, 2013 – A new approach to using the cloud for collaborative management of new and archived video assets marks another step toward producers’ ability to leverage entire libraries of content for new TV programs and business models.
The advance is embodied in the new Stage component of the Reach Engine content management platform developed by Levels Beyond, which is working with a wide range of studios, networks and other entities to facilitate better use of assets. “We believe a hybrid cloud video model with locally managed assets offers content producers a firm foundation to optimize the unprecedented ROI that`s becoming available to them,” says Levels Beyond CEO Art Raymond.
Digitization of valuable video assets has become a top priority across the spectrum of entities that see ways to drive new revenues from repurposing of content, ranging from snippets captured from old files and in live events to distribution of complete productions into multiple outlets. Some initiatives are on the scale of NBC Sports’ recent library conversion, which as previously reported, is utilizing the services of Deluxe Entertainment to digitize, catalog and store some 400,000 hours of archived content. Others entail conversion of specialized libraries such as the University of Southern California’s eight petabyte catalog spanning 50 years of commercial and academic content.
“Today’s libraries are dynamic living vehicles, not just resting places for static content,” Raymond says. “It’s critical to allow users easier upload, access, editing, distribution and integration across all video platforms.”
The cloud offers a way to do this to maximum effect, he adds. “Firms will need to adapt quickly to cloud-based video production and distribution to prosper and thrive,” he says. “It untethers local content libraries so producers can take full advantage of the paradigm shift currently under way on how people interact with and publish video content.”
The range of applications for elements pulled from these digital libraries is endless. One interesting example can be found at the San Francisco Giants’ home park where the people in charge of generating video on the stadium’s big screens now have access to digitized content from games going back decades. When a retired player shows up at the stadium, producers honor the guest by throwing up a selection of sequences from the star’s playing days, says Giants executive producer Paul Hodges.
The team is using the Avid ISIS disc storage archive managed through Avid Interplay to have content ready at hand for instantaneous access in website, advertising, TV and scoreboard applications while relying on less costly LTO digital tape storage to archive the rest of its files. Optimizing content placement across these repositories is a challenge, Hodges acknowledges. “We have to make the decision as to which material stays online and which gets sent to the [LTO] archive,” he notes.
Elsewhere, another digitally inclined baseball franchise is taking the manipulation of content assets to a new level through use of the Levels Beyond Reach Engine to interact with the EVS live production facilities the team uses at game time. Here, says Danny Gold, co-founder and head of strategy and solutions at Levels Beyond, the idea is to capture into manifest files all the tags producers use to mark video segments for replays and other purposes during the broadcast. By catching tags that would otherwise be lost, the team is creating a more comprehensive metadata repository as their game videos go into storage, Gold says, adding he is not at liberty to name the team.
The two teams’ contrasting approaches illustrate how the new capabilities in digital post-production management represent a generational sea change beyond what so far has been associated with digitization of content libraries. In one case, a proprietary disc-based archive system, namely Avid’s ISIS, is accessed through that vendor’s management system; in the other instance a system designed to work across multiple types of archives is positioned as the front end compiler that allows participants in any given post-production project to quickly find and access whatever they need.
“The growing amount of digital content and shorter turnaround time for repurposing and monetizing those assets is a significant data management challenge for companies that work with video,” says Lance Hukill, vice president for big data market development at Quantum Corp., a provider of storage appliances and data management solutions to media outlets and other enterprises. To help customers meet this challenge, Hukill says, his group has paired the Levels Beyond Reach Engine with Qantum appliances and an archive management portfolio that leverages LTO tape as well as disc-based storage.
The resulting media asset management system automates media processes such as conversion of video formats and transfers while managing content across multiple tiers of storage, Hukill says, noting the solution is applicable to libraries storing as little as tens of thousands of assets or as much as multiple petabytes. All assets managed through the system are accessible either locally or remotely for timely collaboration on media projects, he adds.
The Cloud Advantage
Such integrations simplify and lower the cost of deployment for any business that needs a solution to manage large archives of video, audio and rich media assets. By integrating its platform with multiple vendors’ editing, transcoding and delivery solutions, Levels Beyond has made it possible for content owners’ production workflows to access digital content, no matter where and how it’s stored, Gold explains.
“We designed Reach Engine to be compatible with the most popular content editors and media tools in use by businesses today,” Gold says. Through the Reach Engine Production interface, customers can ingest media, create clips from raw materials, send those clips to NLEs (non-linear editing systems) such as Avid, FCP, Premiere and XML, import and modify rough cut timelines and export final outputs using Telestream’s Episode Engine.
The Stage cloud enhancement has extended access to multi-site production teams wherever they might be, without requiring them to create duplicate cloud storage of assets for use with whatever proprietary systems they might be using at their locations, Gold notes. “If you have an Avid site with Interplay working you’ll be able to sync your Interplay data base with the [content] proxies in our cloud service,” he says. “It doesn’t matter what you have on site. Your collaborators will be able to interact with Reach Engine through Stage in the cloud.”
The Stage solution supports rules-based rights management to define what various parties to the cloud-hosted platform are entitled to access, including metadata as well as core content. “You can set up rules on a workflow or engineer the system to require manual approval for various types of usage,” Gold says. “We’re seeing a lot of our customers using Reach Engine as a rights broker.”
Similarly, the Reach Engine workflow allows customers to automate interactions with components such as transcoders and file transfer systems that are used in the distribution process. “We have a lot of these actions prebuilt into the workflow,” Gold notes. “We write plugins for all those tools to allow you to automate your processes through one-click selections of whichever tools you want to use.”
The range of players working with Levels Beyond at this point offers a picture of how digital content automation is altering the landscape. “What we’re seeing is a perfect storm of all these things people can do with video,” Gold says. “We’re going after a lot of sub-verticals that are able to expand on their opportunities with Reach Engine. We’re working with news organizations, education institutions, ad agencies – a lot of entities beyond traditional TV networks and studios.”
For example, the USC Digital Repository referenced earlier plans to use the Levels Beyond platform to provide streamlined media management, distribution and monetization capabilities for video producers who host their material with the repository. This arrangement will provide an opportunity for content owners to use the repository to digitize, preserve and store a video library and then use Web-based tools to manage, edit, package and distribute to any content delivery network, website or individual, says Sam Gustman, executive director of the USC Digital Repository and associate dean of USC Libraries.
“The explosion of commercial and academic digital content and the need for greater access is driving demand for highly efficient, integrated and accessible archiving solutions,” Gustman notes. “The Reach Engine Studio platform seamlessly provides customers with the flexibility and performance to cost effectively package high-value content for worldwide distribution.”
The USC Digital Repository, built from the experience and infrastructure created to house the USC’s Shoah Foundation video archives, is one of the first such repositories to open its services to all commercial and educational users, providing professional expertise and technological resources to allow users to cost-effectively preserve and manage complex collections of film, TV and video tape assets. “They built a very robust system for digitizing content, including batch digitizing capabilities with the Front Porch auto loader,” Gold says.
On the ad agency side, Levels Beyond is involved with two major firms, Gold says. “They need to be able to use ad content in ways that go far beyond the 30-second spot,” he notes. “For example, car manufacturers are putting together second-screen experiences to run with sponsored events. We’re seeing a lot of immersive brand advertising with tie-ins to local retailers and markets.”
As illustrated in the baseball example, a vital function performed by Reach Engine is the compilation of metadata from multiple sources to support the repurposing of content at very deep levels of granularity. “The first step,” Gold says, “is to identify where we can get information that’s already in existence somewhere. It can be embedded in the content or in an entertainment registry of information that somebody has captured. We’re thorough in this because you should never waste effort replicating metadata that’s already out there.
“Our workflow makes it easy to extract data from external sources,” he continues. “We map to the field where data is being ingested. With live events we tap into the EVS system where the person operating the console in the production truck is marking content with data that would otherwise be lost.”
All of this requires integration into production and asset management systems so that descriptive data and tags can be extracted and compiled into the metadata manifest for a given suppliers’ digital library. “We have a base of system integration partners we work with here and in the U.K., which allows us to perform integration with our customers’ systems in a couple of days,” Gold says. “Reach Engine lets you create a flexible workflow system so that when the integration is complete you can capture all that production metadata as it’s ingested.”
Once the system is set up to capture everything available to help with identifying every piece of content there are other steps customers can take to enhance the metadata repository. One involves performing frame-by-frame luminance detection in order to mark frames where scene changes occur. Another involves use of speech-to-text technology to enable phonetic dialog searches in conjunction with keywords the customer wants to use to expedite finding content. “I can type in homerun, and the system will throw markers onto frames anytime it detects somebody saying homerun,” Gold says.
Where commercial content has been augmented with closed captioning the process is much more direct, although it requires that the logging system be able to communicate with the 24 different closed captioning formats now in use, Gold notes. “If you give us a caption file, we’ll capture it in our search engine so that you can search on that and capture any scene in the library,” he says.
A final process rounding out development of the comprehensive metadata repository involves crowd sourcing, where fans of a particular sport, program series or movie can be given access to library content and tasked with subjective assessments such as selecting the best shot of a race. “Say you have a five-year archive of racing footage and you open access to a select group of fans who log into view the content through Stage,” Gold explains. “Now you can compile tags on subjective parameters that give you another dimension to use in describing content. We think people will start adopting this with the use of Stage.”
Flexible Business Models
Indeed, the possibility of using Stage to open access to consumers represents a potential new direction in developing business models which may, in the long run, allow direct-to-consumer distribution from archived content. “There’s nothing keeping Reach Engine Stage from being brokered to content libraries where you can expose your deep archive for access by end users,” Gold says.
This isn’t as farfetched as it may sound when seen in the light of near-term trends emerging with the capabilities embodied in Reach Engine and Stage, he adds. “We’re seeing more and more content owners taking steps to cut out the aggregators,” he says.
“They want to avoid going through a Hulu where Hulu takes a big piece of the advertising revenue,” he explains. “Initially they had to rely on aggregators because they couldn’t do everything themselves. In this new environment it’s all about building and nourishing the relationship with the consumer and controlling the advertising.”
Reach Engine and Stage allows content owners to build out branded online consumer experiences with their own software players in place to support advertising and new features. “Your ability to deliver highly tailored user experiences allows you to think about new business models,” Gold says. “And by going directly to consumers you’re able to collect data on what consumers are interested in.
“Does it make sense to launch another 24-hour linear channel to serve a large niche that’s taking shape around some of your content? How can I engage with consumers to deliver them more targeted advertising? These are questions content suppliers can answer and act on with these new tools.”
The choice of revenue models isn’t an either-or proposition insofar as the Reach Engine workflow allows content suppliers to support multiple models whether they’re based on direct-to-consumer relationships or affiliations with aggregators and service provider distributors. “The content is there and organized, so you can start flowing it to all these outlets depending on whatever policies you want to apply,” Gold says.
“And you can shift models quickly with various mixes of advertising and subscription options,” he adds. “For example, you might want to charge $5 a month for a mobile app as an enhancement with ad-supported content. You can control the models on the fly where you have the option to go with subscription or advertising depending on what fleshes out over time.”
As these capabilities take hold content suppliers will be able to weigh how far to go in pursuit of direct relationships with consumers. Gold cites the HBO Go model as one to watch, given the network’s consideration of a direct-to-consumer subscription model as a way to capture users who aren’t subscribing to cable or satellite services (see February issue, p. 8).
“That’s the last domino,” Gold says. “When we’re in the middle of all this you can extract a lot more value out of your content. It will be interesting to see how it evolves.”