Share this story

Share on X Share on Linkedin Share via email

By Julian Wright

Metadata New Year to all

The driest aspect of media operations management is content metadata but well-structured and organised metadata schemas can pay significant dividends in driving operational efficiency. January is the perfect time of year to look at how this is structured, utilised and managed in your operation. Like Dry January it probably isn’t as onerous as it first appears and will equally deliver health benefits – well, operationally.

January, the Monday of the year. The festivities and over-indulgence of December are already a distant memory and, for those of us in the Northern Hemisphere, several long cold, dark months lay ahead. How very miserable. So universal has this sentiment become that Blue Monday (not New Order’s seminal 1980s synthpop record of the same name) is recognised as the most depressing day of the year

The more optimistic amongst us use January to make resolutions, declutter, stop drinking (for a while) or lose weight. These organised optimists use this time to tidy up and arrange so that the year starts off on the right footing. The theory being that the rest of the year will follow the organised structure defined at the outset. In media management operations the equivalent of the painful-yet-rewarding trip to the gym is to organise metadata schemas.

Why is video metadata important?

Still reading? good. Metadata is commonly, and rather glibly, described as “data about data” but its importance to a media organisation is second only to the assets themselves. Metadata helps media companies organise, manage and find their content – it’s the foundation on which media asset management systems and slick operations are built. It is hard to get excited about metadata, though. It’s not particularly interesting. But structuring, generating and tracking metadata can increase your media business’ efficiency exponentially – as well as saving and generating bundles of time and money.

6 Key Types of Video Metadata

The subject of metadata is an expansive one, metadata can be divided into countless types and categories. Some metadata is automatically generated by the equipment or software used to create or process the assets, while other metadata needs to be added manually. The types and amount of metadata that is required to support an operation will depend on a variety of factors including the size and type of the organisation. Metadata is a vast subject area but – akin to keeping those New Year resolutions realistic – we’ve limited this piece to, a very manageable, six of the most common, key metadata categories.

  1. Catalogue
    Catalogue metadata is typically the highest level of metadata. It describes the asset as a whole and helps identify the object, typically from a synopsis metadata field. For finished long form material, catalogue metadata tends to be quite structured and have a hierarchy such as Show > Series (Season) > Episode. For rushes this metadata is typically less structured and details high level information, for example ‘ 2017 Presidential Inauguration.’
    Beneficial use: Descriptive metadata is useful when searching for clips or complete programmes.
  2. Temporal
    Temporal metadata is most commonly used for linear assets, such as video clips, and describes what can be seen at various points or sections within the object. It is metadata which is time specific within an asset. Temporal metadata is typically generated by manually logging the material and, with the exception of sports, is generally free-text or uncontrolled. Loggers create temporal metadata by typing a commentary description of what they can see in a clip, and this metadata is tracked against the time-code of the object to create a frame-accurate, referenced data-set. Temporal metadata sometimes includes transcriptions of what is said in a piece of video which may be machine generated or derived.
    Beneficial use: Temporal metadata makes it easier to find specific areas of interest within a clip or a programme by searching for keywords. It is particularly useful in news and sports applications which tend to have very long rushes.
  3. Rights
    All media assets, howsoever sourced, should include metadata that identifies the rights owner and provide details regarding copyright and usage allowances. This is becoming an increasingly important aspect in an era in which almost anyone with a mobile phone can produce material. The creator is automatically the rights owner and individuals are increasingly asserting those rights.
    In professional production, rights tend to be more complex, and are increasingly so. Typical restrictions of geographies, play-out repeat intervals and date based expiry are being extended to platforms and editing limitations. Requirements to credit also fall into this category.
    Beneficial use: Rights metadata is critical to ensuring that content distributors are aware of, and comply with, any rights restrictions. This is becoming increasingly important – particularly in news organisations.
  4. Structural
    Structural or Technical metadata relates to the structural attributes of an electronically stored asset. These typically include video file format, codec, bit-rate and wrapper, frame-rate as well as the audio arrangement. This category also includes aspects such as file size and running duration for video assets.
    This technical information is mostly created automatically when the video asset is generated, and the amount of available data is increasing with even basic camera equipment capturing vast amounts of data. Devices capture everything from aperture and shutter speed to GPS (Global Positioning Service) information which, if coupled with Time of Day (ToD) data, provides a powerful core reference from which much information may be derived. So technical data isn’t just for the techies.
    Beneficial use: Technical metadata is critical to ensuring content is format compliant with the systems on which it is going to be processed. This is increasingly important for the ever-increasing number of media distribution platforms.
  5. Administrative
    Administrative metadata typically carries production information about an asset or assets. This might be information about a shoot which is not automatically generated by the equipment – such as the names of key crew, location details or weather conditions.
    Beneficial use: Administrative metadata is typically used in post-production operations to manage rushes. This is particularly beneficial in productions with high shoot ratios such as wildlife or reality programming
  6. Relational
    Understanding the relationships between assets can be extremely powerful in many operations. The ability to create and view the relationships between assets is import right through the production process, but especially for finished assets. Typically, a completed piece of content will have a number of child assets such as subtitles, audio tracks (potentially in different languages) as well as derivative objects such as promos. Features and big budget productions from the major studios are delivered with Electronic Press Kits (EPKs) which contain an array of promos, stills and other marketing material. EPKs can be vast, and being able to view how the assets are related is extremely important, particularly in the case of episodic productions.
    Beneficial use: Relationship metadata is especially important in multiformat distribution or playout environments, especially those servicing multiple languages and regions.
    Modern media management systems like BLAM provide the capability for operators to simply define their own metadata schemas and ensure data is properly captured using controlled vocabularies and other techniques. So, after the morning gym session, a few hours looking at metadata use in your organisation can really help improve your operational fitness in readiness for a successful 2017.

Happy New Year

 

By Julian Wright

Share this story

Share on X Share on Linkedin Share via email