Rambla CDN and the power of metadata

As you may know, Rambla is currently involved in a research project - funded by the Flemish government agency for Innovation by Science and Technology (= IWT) - regarding support for metadata through our CDN. The goal of this project is to find a way in which our customers can benefit from having a context-aware CDN. This post will try to explain what we mean by metadata support and how this could benefit you.

Kinds of metadata

Metadata is simply data about data. For multimedia purposes, we can divide metadata into a number of basic categories :

• Content Attributes : Describes the basic form of the actual multimedia file or stream, including things like the length and format of a video.
• Content Description : Covers all information about the actual content, like title, genre, date and location at which the content was created. This information can either be generated by your production software or it can be human-authored (e.g. summaries for playlist or program guides).
• Timed Metadata : This is information that applies to a subset of the content. At a basic level this is content such as closed captioning or even lyrics for music, but it can be much more sophisticated including listings of who or what appears in a scene or geospatial information such as where a scene was shot.
• Hotspots : This is a more detailed level of information than timed metadata, allowing identification of exactly which portion of a frame or sequence of frames an object, person or user interface element exists on.
• Context Information : This provides a context in which the multimedia may be consumed, such as an allowed platform, access restrictions or monetization related data.

Value of metadata

The availability of metadata allows you to improve your multimedia management and can drive growth in viewership, discovery and monetization opportunities. These opportunities range from contextual advertising, interactive playlists and content recommendations to advanced use cases such as personalization and targeting, social networking, video syndication and dynamic packaging.

The integration of metadata at CDN level offers customers a direct way to harness this potential, without them having to deploy a software solution to store, manage and disseminate their metadata. Moreover, keeping the metadata on the CDN together with the multimedia files will unlock a number of extra benefits. Here are some examples:

Metadata Capturing

• Automatically capture and store content attributes for multimedia files that are uploaded and/or encoded using our services. For instance, if a source file has been transcoded into different formats, you will be able to distinguish the transcoded files by looking at their metadata.
• Create or update metadata programmatically, from multiple sites or applications. Create your own vocabularies or use existing ones. Allow third parties or end users to collaborate in generating metadata (e.g. ratings).
• Future development of advanced metadata processing services is also possible, like converting speech to text or tracking timed metadata or hotspots. This could lead to related services being deployed at CDN level: e.g. video-level or scene-level searching based on metadata indexing.

Media Accessibility

• Easily identify your files from any location on the web, using content description data (e.g. title, description, date). Don't be tied to a particular CMS or application.
• Let your applications retrieve dynamic file listings (feeds, playlists, ...) or provide such listings yourself for end-user consumption or video syndication (e.g. based on genre, container format).
• Export metadata in different formats. This should also make it possible to re-encode multimedia files, while storing existing metadata in a container-specific format.

Content Monetization

• Dynamically serve ads or content suggestions, or let others ingest preferred content from your account.
• Get and compare statistics based on metadata queries (e.g. get all multimedia traffic for a certain event or genre).
• Future development of an API for context information is possible and would allow you to control who can receive your (live-)streams, based on user credentials or credits.

Your Opinion matters

Since this is a research project, we are very much interested in your opinion. Would you consider storing your metadata at the CDN level? What benefits or features would you be interested in? Which kinds of metadata are important to your company or organization? Do you see any problems concerning your current workflow or infrastructure? Just leave a comment, or send us an email.