Federated discovery
A guide to how television and video services can be discovered at scale using distributed systems.
The importance of federation in finding services
Television service discovery is becoming more complex. Services are no longer delivered through a single network or platform. They are provided by many organisations, across different countries, using a mix of broadcast and online delivery. Devices and applications need to discover these services without maintaining individual relationships with every provider.
At a national level, this can be managed through fixed configurations or direct integrations. It does not work at internet scale.
Federated discovery applies a distributed approach to this problem, allowing services to be found through a network of registries and resources rather than a single point of control. This reflects how the internet itself has scaled.
What is federated discovery?
Federated discovery is a distributed approach to finding services, in which multiple independent parties provide structured references to services, enabling devices and applications to discover services at scale without relying on a single central directory or controlling authority.
In a federated architecture, devices and applications can query one or more registries to discover available services. Registries may represent sources such as national services, platform offerings, or global aggregations.
This form of discovery federation mirrors established internet patterns, where no single authority maintains a complete catalogue, yet discovery remains consistent and interoperable.
What are the problems with scaling service discovery?
Service discovery becomes difficult to scale when every participant must integrate directly with every other participant. In this direct model:
- Manufacturers must integrate with each service provider
- Service providers must negotiate with each manufacturer
- Platforms must maintain multiple relationships over time
This creates unnecessary complexity and limits growth.
A federated model uses a structured registry layer for service discovery, reducing the number of relationships required and enabling service discovery to scale across a growing ecosystem.
What are the requirements for service discovery at scale?
A service discovery system must meet a number of requirements to operate at scale across devices, services, and markets. It must:
- Allow discovery without direct integration between clients and services
- Support multiple independent service providers and registry operators
- Remain interoperable across different devices, applications, and environments
- Avoid dependence on a single controlling authority or point of control
- Depend on no single administrative root, namespace, or domain under unified control
- Ensure that failure, withdrawal, or policy change of any one operator does not disrupt the system as a whole
- Enable independent evolution across different markets and participants while preserving interoperability
- Operate reliably over the lifetime of deployed products
These requirements reflect the realities of a distributed and evolving ecosystem, in which technical interoperability must be maintained without imposing centralised operational control.
A federated discovery model satisfies these conditions by design, allowing multiple parties to participate independently while maintaining a consistent and interoperable approach to discovery.
What are the limitations of a central service registry model?
A central service registry model depends on a single domain, directory, or point of reference to represent all services.
This creates structural limitations. It:
- Risks a single point of failure
- Concentrates control in one organisation
- Imposes a potential gatekeeper
- Limits the growth of the ecosystem
This gatekeeper risk arises wherever discovery depends on a single authority, whether implemented as a single domain name, a single registry, or a single aggregated directory. Decisions about which services are included, represented, and prioritised may become centralised.
In an open and evolving ecosystem, this creates operational and market constraints. It may limit innovation, introduce barriers to entry, and raise questions around neutrality and governance.
A federated discovery model avoids these issues by distributing responsibility across multiple parties and registries. Discovery federation allows different participants to contribute, while maintaining interoperability without reliance on a single controlling authority.
How does federated discovery work?
Federated discovery introduces a structured process for finding services.
In a federated architecture:
- Service providers publish metadata describing their services
- Registries aggregate and organise service offerings
- Devices query registries using standard parameters
- References enable direct access to services
A registry functions as a directory that references service offerings, not a single catalogue of all services. This separation allows discovery federation to remain flexible, while service delivery evolves independently.
How do devices choose which service registry to use?
In a federated discovery system, a device or application may be able to access more than one service registry.
A client must therefore determine which registry, or registries, to use when performing service discovery. This selection may be based on context and implementation choices.
A client may:
- Adopt a registry provided or configured by the device manufacturer or application
- Select a registry based on user context, such as country or region
- Support multiple registries to reflect different sources and requirements
- Apply preferences based on compatibility or supported delivery methods
In addition to technical considerations, registry selection may reflect broader requirements. A client may favour registries that are:
- Trusted and well-governed
- Comprehensive in service coverage
- Reliable and performant in operation
- Maintained over the device lifetime
This allows service discovery to operate within a competitive and open ecosystem, where registries may coexist and clients select those that best meet their requirements.
Discovery federation therefore supports both interoperability and choice, without requiring a single mandated registry or central point of control.
How do devices find service registries?
Once a registry has been selected, a federated discovery system relies on standard internet mechanisms to locate it.
Devices and applications use domain names to identify discovery endpoints. These domain names resolve dynamically to registry infrastructure using the internet domain name system.
This approach allows discovery federation to adapt over time:
- Maintain stable entry points through domain names
- Delegate registry operation to different providers
- Redirect discovery to alternative endpoints as required
This indirection is a defining characteristic of a federated architecture and enables long-term flexibility.
How does federated discovery improve resilience?
Federated discovery improves resilience by removing dependency on any single component.
A federated architecture:
- Allows multiple registries to operate independently
- Enables fallback to alternative discovery endpoints
- Supports dynamic redirection through domain resolution
This ensures that service discovery can continue if a registry becomes unavailable or there is a change of operator. Discovery federation limits the impact of failure and supports long-term continuity.
Who runs registries in a federated service discovery system?
A federated discovery ecosystem allows different organisations to operate registries for different purposes.
Participants may include:
- Service providers managing their own offerings
- National bodies representing regulated services
- Manufacturers supporting device-specific discovery
- Independent operators providing global aggregation
This diversity is a feature of discovery federation, not a limitation. A federated architecture allows multiple registries to coexist while remaining interoperable.
How does DVB-I support federated discovery?
DVB-I defines how devices query for service lists, but does not specify how a registry is discovered. This enables federated discovery by design, rather than enforcing a single central service registry.
Discovery federation builds on DVB-I by using standard internet technologies such as DNS and HTTPS to locate registries dynamically. Devices can query multiple registries using a consistent interface.
This federated architecture works within the DVB-I specification and does not require any changes to it.
What role does the Service List Registry play in federated discovery?
The Service List Registry is designed to operate within a federated discovery model.
It does not attempt to act as a complete catalogue of all services, or to control which services may be discovered.
Instead, it provides an authoritative and structured means to discover service lists across different sources, acting as a reference point, rather than a point of control.
Within a discovery federation, a registry can:
- Reference service lists from multiple providers
- Validate and curate service information
- Ensure reliable discovery endpoints for clients
This supports a federated architecture without introducing central control.
What does federated discovery enable?
Federated discovery enables services to be discovered at internet scale.
A federated architecture can:
- Enable new services without requiring platform updates
- Support multiple markets and regulatory environments
- Maintain interoperability across devices and applications
Discovery federation provides the foundation for a scalable and adaptable television ecosystem.
Conclusion
Federated discovery applies established internet principles to television service discovery.
By distributing discovery across multiple parties and registries, it reduces complexity, improves resilience, and avoids central control. A federated architecture allows discovery infrastructure to evolve over time while maintaining stable interfaces for devices.
As television continues to move into an environment based on the internet, discovery federation is not simply an implementation choice. It is a response to the fundamental requirements of service discovery at scale.
The Service List Registry is built on this principle, providing a structured and authoritative way to discover services within a federated ecosystem.