 Since we are now looking at what services are and how they should be seamlessly provided across different stakeholders in an NGN or other kinds of next generation networks. So we must understand that the services are not only going to talk to the end users, they also would be talking to other services. So interoperability is an important aspect that needs to be catered for while considering these services. In this module, we shall look at what interoperability issues arise and how they should be best handled as per the NGN recommendations in Y-Series. First, let's look at the assumptions. It is widely understood that whether it is the exercise or the core or even the distribution side, we are going to experience and we are going to look at networks with which are heterogeneous in their design and in their architecture. So are these services cognizant to that fact? Then the services are heterogeneous themselves. Services which are being offered from one kind of legacy system once integrated into NGN would have a unique kind of intrinsic behavior. So it means we would have some similar kinds of services which would be offering some similar kind of functionalities but they would have some non-uniformity or heterogeneity within them. And lastly, when we look at the user equipment, the capability of the user equipment is not consistent across all NGN. Depending upon the constraints and depending upon the feature set of the user equipment, some services may be well compatible and some services may not be smoothly running. Then begins with the concept of keeping everything interoperable. It is fairly theoretical but it's a utopian goal that must be at least tried. For that, the first and foremost feature of services should be that they all should be interoperable through some kind of open interface. But as the reality strikes, the de facto standard of IP is already there and when the services as in messages and the protocol data units are encapsulated within the message, eventually it gets carried in an IP packet. So IP is the de facto standard. So that makes the design of services quite flexible, quite adaptable but at the same time it is closely tied to the availability and the constraint of the IP. From there, we deduce that if a service has to be designed and interoperability has to be kept intact, then the appreciation of the behavior and the operation of IP has to be considered. And for that, the first and foremost requirement is to understand that IP version 4 is already suffering from the address depletion. The addresses which were assigned by IANA are almost over except for certain regional internet registries where some very sporadic IP address pools are available. Otherwise, IP version 4 addresses are increasingly being replaced with IP version 6. So this natural transition has to be catered for and has to be dealt with in interoperability from the services point of view. Then in another important service design consideration from the interoperability point of view is that there are certain non-NGN networks which are integrated into NGN and then there are some default or I would say by nature inherent NGN based networks. So if there is some kind of incompatibility between these networks, it is also reflected in the services which are provided between these networks. So that's an important consideration that interoperability has to be ensured between non-NGN networks and NGN networks and their subsequent services. When we look at the services which are once offered and while these services are being provisioned, they actually have to be continued when some kind of migration takes place from one network to the other. For instance, if the caller is initiating the call from PSTN and while the call is on, the user moves to an area where the user equipment supports voice over IP what's going to happen and in the telecom world it is with regards to the interoperability of establishing a service and keeping that service alive for instance while moving from a GSM network to LTE network and vice versa.