1. There is not a single definition for metadata
Although metadata literally means ‘data about data‘ from its Greek etymological root, various organizations have defined the term more precisely, such as the National Information Standards Organization (NISO), which defines it as ‘structured information that describes, explains, locates, or simplifies the retrieval, use, or management of an information source’.
Other organizations such as the American Library Association (ALA) opted for a more descriptive definition: ‘Metadata is structured and coded data that describes the characteristics of information entities of interest to assist in the identification, discovery, assessment, and management of the described entities’.
Finally, the Dublin Core Metadata Initiative (DCMI) explains metadata as ‘structured data that includes data associated with information about a system or object for the purpose of providing a description’. Although all definitions are similar at one or more points, it seems clear that metadata is an essential component to help us locate, classify and understand the objects-whether physical or non-physical-that they are attached to.
2. The conceptual origin of the metadata is in libraries
Already in Ancient Egypt (280 BC), the Great Library of Alexandria offered all its scrolls with a small label attached to each one. They contained the title, subject and author of the work, allowing readers to locate their search more easily and librarians to organize the catalog without wasting hours of work in the process. These labels would later become bibliographic records, and their digital version would be mentioned for the first time by two MIT engineers in a report entitled ADMINS, where they mention ‘the need to maintain a record (metadata) of the data records’.
3. Metadata travels more than you might think
For example, every time you take a photo with your smartphone, it generates data about the exact date and time it was taken, the resolution, exposure, focal length, file dimensions…and also the exact coordinates from where you took it. Although most social networks will remove this data if you publish it, in the digital age it is important to be informed and cautious about the information you share with others on the Internet.
4. Google would not be Google without them
In 2007, Google irrevocably changed its search engine. Until then, it was based on a list of relevant links, whether paid or unpaid. But Google expanded the scope of its searches to news, images and video, and as a result, countless metadata began to be relevant in its search results, which would also determine its importance when creating web pages and facilitating their search as part of optimization processes. Today, the success of digital marketing depends significantly on accurate and well-organized metadata.
5. Metadata automation came in 2015
Similar to the creators of YouTube, who came up with the idea for their creation when they couldn’t find a way to share a video among friends, Amnon Drori, Gal Ziton and Itai Kahalani, three Israeli software engineers, were frustrated by having to manually track the origin of the different data they needed in error-prone sessions that could take hours or weeks of work. So they decided to found Octopai, a Machine Learning technology company that automatically maps and manages metadata collectively in a single search interface. The start-up has so far raised more than five million dollars from different level investors and alliances with giants like Microsoft.
6. Its importance as a business asset is growing every day
In the words of Amazon’s CTO, Werner Vogel, ‘Data is at the core of value creation, while physical assets are losing their importance in business models’. To get an idea of the certainty of this sentence, it is only necessary to take a look at the common element of most of the 10 most valuable companies in the world: how much would Alphabet, Apple, Microsoft, Amazon, Facebook, Tencent or Alibaba be worth without the data they manage, organize and use to improve their offer day by day?
This importance is not limited to large corporations. According to the recent study Streaming Wars: Raise or Fold conducted by Capgemini, although content is the king and first key differentiating element in the supply of content, the collection and accumulation of data that helps improve the user experience, brand recognition and other key elements in business profitability make it increasingly an essential strategic element, a reality that can be extrapolated to an entire broadcast sector that is increasingly closer to the consumer. So, how can industry professionals take advantage of its full potential? We dedicate the last points to explain it.
7. The recording of metadata should not be saved for last
One of the most common mistakes made by Broadcast and Media professionals is to only include metadata in the last phase of production, archiving. However, metadata should be recorded throughout the production process, from the moment it is generated and with special emphasis on the production and planning phases, where most of the relevant user metadata is generated.
If this task is followed up properly, it will be much easier in the future to locate and keep our metadata organized, information whose potential, as we have seen, is on the rise. We can even use the templates offered by some databases for each phase of production, increasing our productivity and efficiency.
Everything you need, we have! Request your demo today
8. Metadata management is easier if you apply standards
To convey meaningful information, metadata must be combined into groups. In the same way that our respective languages group words together to form meaningful sentences, it is more than advisable to follow rules of grouping and structuring metadata so that the information is also passed on meaningfully.
A metadata structure is defined as the general organizational structure that all assets of a database follow. It can be a list of fields that act as a grouper, or a schema that shows the relationships between these data. The various Broadcast, library or archive communities have developed numerous standards or norms to structure them (SMPTE ST 335, EBU, MPEG-7, MPEG-21, PBCore, etc.), giving users several options to choose the one that best suits their purposes or even the necessary initiative to generate their own standard.
9. Cataloging systems are essential
Fortunately, as we have seen throughout this article, technology has evolved to assist us in carrying out successful content cataloging tasks and today, the cameras themselves can automatically record many of the technical metadata that can be useful in the medium and long term. Similarly, this information can also be automatically imported into MAM systems and there already are, in fact, some advanced cataloging functions, such as speech-to-text, or system integrations, such as VSNExplorer MAM with Online Thesaurus, to help us solve this task in a matter of seconds.
10. There is no magic formula
Users must continue to manually add consistent, clear and accurate data. Most video metadata is created automatically, but manuals are becoming increasingly important. These have great potential for marketing our productions and from the very moment we ingest material into our MAM, it is a good practice to add them by hand to preserve much more information about our productions.
Although the lack of time to prioritize more urgent tasks may be pressing, a good documentation job can save us a lot of time in the near future…or maybe not so near. As Mike Cox pointed out in Descriptive Metadata for Television, ‘metadata is not only included for us or for our most immediate needs, but also for users who will be using it 100 years from now’.
Subscribe to our newsletter to stay updated about all our news