By Mark Darlow
Investing in media asset management is still not top priority for most broadcasters because they think it is complex and expensive. Mark Darlow of Harris Broadcast Communications Division disproves this assumption and instead, shows how digital asset management can actually simplify content management while also bringing in cost benefits.
Digital Asset Management (DAM) has been on the agenda for some years now, but ask most broadcast buyers and they will still tell you that they are ‘going to investigate it seriously next year.’ So far, it hasn’t reached the top of their list of priorities. This is because it is perceived as complex, disruptive and expensive. This paper will show that, when implemented properly, DAM is none of these things and indeed, simplifies the business process of content management and brings clearly identified cost and revenue benefits. Part of the problem is the proliferation of new terms flying around. We have had asset management, just as we have had ingest and metadata, since the beginning of broadcasting. We had never used those terms.
When a programme or a commercial arrived at a broadcaster, it was ‘ingested’ by a librarian writing all the details (the ‘metadata’) on an index card, which was then stored in a database (or Rolodex as we used to call it). On a shoot, the production assistant captured the metadata by writing notes on the tape box or film can. The dynamic scheduling system was usually on a whiteboard.
For the first 50 years of the television industry, this method worked well, and people became very comfortable with it. When computers came along, the temptation was to take the same workflow and make it electronic. Fifteen years ago, one of the world’s biggest broadcasters spent several million dollars implementing a computer archive system that was a representation of its card index. But 15 years ago, the multi-media business did not exist; we operated in an analogue world with single channel distribution mechanisms. Going digital and the explosion of distribution channels has proved to be the catalyst for change.
Now, content resides not on tapes stacked on shelves, but as files in computer storage. That content needs to be accessed many times by many people, often simultaneously. For instance, a large playout centre may have 100 channels or more on air, 24/7, all accessing the same stock of commercials and probably, sharing some programming.
The access control systems have to change; they must approach the problem from a completely different angle. Now, the limitation is not the physical media, but the privileges of the individual user. If a user is entitled to view, process or transmit a piece of content, then it should be made available, instantly (or as nearly so as possible).
The user can reasonably expect to see the content as well as its database entry (its metadata). A frame-accurate, low-resolution proxy should be delivered over the office network to allow the content to be reviewed.
Some users will need to work on the content — it may be that a programme needs to be edited to meet local expectations or regulations, or it may be that a trailer needs to be created. In some cases, edit decisions can be made on the browse resolution copy and automatically conformed on the online material. In other instances, full-resolution content can be transferred over the network for desktop editing and graphics creation. This is one of the benefits DAM brings to an organisation, empowering a broader range of employees to manage content from their desktop, rather than a few staff with dedicated VTR machines.
We are all familiar with the principles of the tapeless workflow. Content is ingested into the system and sits on a server. It is checked for quality control, transcoded (if necessary) to the house compression standards and copied at low resolution to the browse server. The online version is then treated as a computer file, moved from spinning disks to near-line storage in archives, protected by backups and RAID schemes and transferred over Ethernet or Fibre Channel.
To make such architecture work, the controlling computer network must be capable of tracking the content at all times. In the old days, when you needed a programme, you simply picked the tape off the shelf; now, you select it from a database. But the operator shouldn’t need to worry where the content is – that is the job of the computer. Again, a large physical library and the staff to run it have been replaced by an ingest process and disc storage, providing more benefit to the larger user community at lower cost.
So, no modern workflow can exist without digital asset management of some sort. Of course, these digital workflows will inevitably work across multiple systems, responsible for discrete functionality sets (playout, etc.). One of the key developments of the H-Class DAM application is integration with other Harris applications, ADC, D-Series, Programme Scheduling, such that data can be entered in a users primary application but passed to other applications in the workflow chain, providing further cost and security advantages by eliminating the need to repeatedly key information into computers.
The DAM system you choose should be modelled on the business imperatives of the broadcaster, not its technical operations. It should ensure seamless linkage between the business phases and the applications responsible for those phases.
The first of these is commissioning and creation. Some broadcasters will make some or all of their own programmes; some will buy everything. Either way, this is where the schedule starts, so this is where DAM must start. Scheduling should be a dynamic process. Rights are attached to every piece of content, and it makes commercial sense to exploit them to the maximum. If you have bought six showings of a programme in a year, then the rights management system should be prompting you to schedule them.
Scheduling content should prompt DAM to manage movement of content to the necessary location.
The mechanics of ingest will depend on the nature of the content and whether it is produced in-house or acquired as a complete package. News content, for example, will be ingested as raw rushes, with completed packages automatically re-ingested internally to provide broadcast-ready content. Use of GPS information exported as metadata from cameras can automatically be catalogued with the content giving instant reference of “do we have any footage of …”
Earning revenue is at the heart of any broadcast operation, so sales and traffic must be linked to the whole system, tracking commercial sales from initiation through the as-run log to revenue receipt. From a DAM perspective, maintaining the stock of commercials and ensuring the correct one is available for each slot are vital. Of increasing importance will be the DAM systems ability to alert the sales system of product placement or product references within programmes.
For linear channels, the next stage is playout — delivering the programmes, commercials, trailers, promos and other content in a seamless, professional flow. In the multi-platform world, there may be multiple versions of the same single channel going to cable, satellite and terrestrial transmission. There may also be another for IPTV and a different variant again for delivery to mobile receivers.
Away from the linear channel, the content may be available online so it will need to be re-purposed, converted to a new format and delivered to an Internet or video-on-demand server. Again, rights management is critical here; a broadcaster may have clearance to broadcast content, but may not have clearance to put it on the Internet.
One of the benefits of the tapeless workflow is that increased storage capacities now allow broadcasters to keep everything, so the concept of moving content to an archive is becoming outdated. It is unlikely, though, that all content would be kept on an online server, so an application is needed to look ahead in the schedule, move unneeded content to near-line storage and pull material back to the playout server in comfortable time before it is broadcast.
All of these actions, and more, go to form the Digital Asset Management system. While it should be seen by users as a single system, it is likely to be supported by a number of different applications, probably from different suppliers. Scheduling systems, for example, or sales, are specialist tools. They they need to provide seamless interfaces into the DAM.
Achieving this means the consistent application of standardised metadata so that all the information is available where it is needed. Schedulers should be warned if they try to put a 4:3 programme into a 16:9 channel. Sales departments should know the genre and stars of a movie before they sell spots into it. And fair news access use of other broadcasters’ content should be carefully tracked.