Metadata - or the ability to tag information digitally within a data stream - has emerged as a vital tool for broadcasters looking to best manage and leverage archived programming content.
For broadcasters, the transition to digital broadcasting isn't a simple case of investing in new technology - it requires the implementation of new processes that apply to the commissioning, scheduling, production, reporting and archiving of content.
This has resulted in significantly altered workflow, and - thanks to the features offered by desktop editing systems - a redefining of some job descriptions.
With this is mind, the term ‘asset' has taken on a new meaning as well, observes Steve Jupe, head of data DMI (digital media initiative) at the BBC in London. ‘Essence' and ‘metadata' are no longer separate entities, as they were in the analogue environment.
Jupe refers to this coexistence as an ‘enabler,' but admits that it adds to the complexity associated with asset management.
"It is an enabler in the sense that metadata can be added throughout the asset lifecycle and not be lost or need to be re-keyed across business support systems," he explains, adding that the creation of metadata becomes more automated, and can be used to not only drive workflow, but promote a better level of accuracy and consistency throughout the facility.
Jupe notes that the flexibility of the digital environment allows assets to benefit from a ‘snowball effect': metadata can be accumulated over the asset's entire lifecycle, rather than just at the end.
"This accumulation alters not only the volume of metadata which can be captured but also ensures data is captured at the point of creation - first hand from the expert - rather than re-translated at a later date," he says.
"Now you have all kinds of metadata being passed back and forth through the system - not only beginning with the original traffic schedule, but also the automation schedule, which then talks to servers and archive systems to make sure that the servers have the right content that's been pulled from the archive," explains Lewis Zager, a broadcast technology consultant based in Arlington, Virginia.
There are also record lists that tell the systems what to record for future air - and it's the metadata that ‘talks' to the servers to determine whether a programme sits on a hard drive or the server, and when it should be moved.
Because the digital environment supports multiple platforms, the separation of assets by media type or delivery mechanism isn't applicable.
"Assets are created and managed purely as content (that is) available and applicable to multiple delivery platforms or for easier editorial manipulation," Jupe says.
There are two sides to every story, however, and all of the new features that digital asset management offers are accompanied by their own set of challenges.
Jupe recounts that the BBC underwent a far-reaching cultural change that emphasised more locally controlled data creation, greater accountability and a widespread governance and support mechanism.
"Support processes for managing assets all change - media asset management becomes an essential element of all BBC business processes," he explains. "This role drives a wholesale reworking of all archive processes and puts the digital archive firmly at the heart of the digital BBC."
This dictated an increase in the volume of material requiring management, Jupe notes, adding that the 'granularity' of the assets also becomes a crucial point.
"Programmes or publications become an association of assets, relationships and metadata links - all of which add to the complexity of the environment to be managed," he says.
He adds that there are also issues surrounding the creation of unique identities for assets, as well as the ease at which digital files can become hidden and, therefore, lost.
"In a digital environment, all you've got is the file name to look at. You've got absolutely nothing to go on unless you have metadata. - Ian Baker, managing director, Metaglue."
Ian Baker, managing director of Metaglue, a consultancy and products provider based in Buckinghamshire in the UK, notes that not only can files be lost - they can be accidentally copied as well, which leads to problems if the wrong file is being modified.
"In a digital environment, all you've got is the file name to look at," he says. "You've got absolutely nothing to go on unless you have metadata."
Jean-Pierre Evain, who works in the technical department at the European Broadcasting Union (EBU) in Grand Saconnex, Switzerland, explains that the EBU's position on lost files is if they can't be found, it's as if they didn't exist in the first place. He notes that the organisation is working on a number of metadata specifications to help people catalogue their assets.
As Jupe stresses, in the digital environment, programmes become an ‘association' of assets. Evain explains that this requires the establishment of a content group, which is made up of episodes, which are part of a programme.
"This programme is going to be made up of different sequences, which are going to made up of different shots, which are going to be constituted of different media objects, like audio, video, subtitles and other data," he says.
"All of these components have to be organised together. This is the overall data model."
From there, another layer of description develops: the name of the group, how the group is described, how a programme is tagged, and how specific segments, shots and even media objects - such as video and audio - are labelled.
"Once you have defined the overall model, then you decide the level of granularity that you want to have to describe these different elements," Evain says.
Zager points out that metadata play a key role in managing - or, depending on how it's configured, mismanaging - different versions of the same programme according to geographical or cultural location, or language, for example.
"If somebody makes a mistake and inputs the wrong metadata, then the wrong version of the programme will go to air," he says.
To avoid these problems - and to ensure that assets are managed efficiently - a facility's metadata strategy must incorporate some foresight: how might the asset be used further down the line? Baker observes that unless everyone is on board, and therefore committed to entering metadata when they ‘touch' an asset, gathering useful information about a specific piece of material is difficult.
"Production personnel won't create metadata - or at least, good metadata - unless they see a point in it for themselves," he says. "You want to make sure that you are creating good metadata, and therefore, people should know why they need to create it."
In order for broadcasting organisations to capitalise on the capabilities that the digital environment has to offer, they must employ a standard metadata schema throughout the entire company.
SMPTE has issued guidelines suggesting how organisations can classify their assets, however because each broadcaster is different, there is a need to create much of the metadata backbone in house.
"A lot of the standard stuff is very useful, however, everyone has their own business, so you will want your own fields," Baker says.
"For example, if when you receive a programme, you want to ensure that it's tagged with your programme number and not the BBC's programme number instead.
"Equally, if you get a file from the BBC, you don't want to destroy the BBC's programme number, because you may want to go back and use it at a later date," Baker continues.
You really want an item labelled 'my programme number,' and then you can be absolutely sure that you are using your programme number and not somebody else's, and that, obviously, extends to a whole lot of other bits of business-specific metadata.
There are also different types of metadata - not just with reference to technical descriptive and digital rights-related fields, but also within areas of the organisation, Evain explains.
"That is where the difference between the main data model and the information that is managed at the database level, and all of the information that flows within the production environment and also between providers," he says.
The EBU applies three sets of metadata: one for production, another for news exchange, and a third for archive exchange.
Zager notes that companies often underestimate how labour intensive this can become.
"It's not simple, and it requires a tremendous amount of human input," he says.
"Who is going to sit down and input all of this material that you have into an asset management system? It requires a tremendous amount of manual input to get it right."
Baker advises facilities that are implementing new metadata strategies to start small and to build in flexibility.
"You will learn an awful lot if you just put a few items of metadata in," he says, suggesting that a good start is to create fields for basic items, such as programme ID, the location the material was shot in, who shot it, and how the material arrived in the building.
"As you learn more, and people start to learn how to use the metadata, then you understand much more about the fields that you need and the fields that you don't need. Nothing kills a metadata scheme as quickly as people becoming bored with entering data."
When a metadata strategy is rolled out to affect every piece of a broadcaster's business, the sheer interconnection of it all may produce problems as the organisation evolves.
"If you improve your business processes, quite often that make the metadata schema a bit out of date because your business processes have changed," Baker explains.
"Immediately, in the success of having your own metadata schema, you sort of create your own destruction, because the business processes have changed thanks to the new metadata."
Metadata is structured data which describes the characteristics of a resource. It shares many similar characteristics to the cataloguing that takes place in libraries, museums and archives.
The term "meta" derives from the Greek word denoting a nature of a higher order or more fundamental kind.
A metadata record consists of a number of pre-defined elements representing specific attributes of a resource, and each element can have one or more values.
Typically, the semantics is descriptive of the contents, location, physical attributes, type (e.g. text or image, map or model) and form (e.g. print copy, electronic file).
Key metadata elements supporting access to published documents include the originator of a work, its title, when and where it was published and the subject areas it covers.
Where the information is issued in analogue form, such as print material, additional metadata is provided to assist in the location of the information, e.g. call numbers used in libraries. The resource community may also define some logical grouping of the elements or leave it to the encoding scheme.