interests. For example, as Twitter expanded into more countries, it confronted an increasingly large number of demands to remove content. Initially, Twitter responded by removing Tweets worldwide. If a user in a single country demanded removal of a Tweet, Twitter would do that for all users, in all markets. However, Twitter eventually decided it would be better to use geolocation to remove Tweets only for users from the country that made the legal demand.Sometimes friction, in the form of low levels of legal interoperability, may be desirable from a public policy viewpoint. For example, one of the most important considerations for governments is cybersecurity. Leaders are extremely concerned about the security implications of highly interconnected systems. Interoperability means that viruses and targeted cyber-attacks can have damaging consequences. Government-created friction at the technical and data layers may be controversial.58 But friction in terms of low levels of legal interoperability across countries may be beneficial if it encourages greater diversity of non-interoperable systems that may serve certain defensive purposes.Legal interoperability is a complex and critical issue, in part because it has the ability to either enable upward mobility in the global economy or to reinforce existing power structures, depending on the choices made. 59 4.7 Important issues for the future Interoperability is not an end in itself. And interoperability doesn’t always have to be maximized. Instead, private actors and regulators must work carefully to optimize the level of interop necessary to meet their objectives, even though the process is neither easy nor simple. This chapter has described an interop framework, the potential costs and benefits of increased levels of interop, and a variety of approaches for encouraging interop. Thus, when determining the optimum level of interop, all of these factors must be carefully weighed and balanced.With emerging and profound new systems and technologies, it is important to bear in mind some of the big questions and challenges that confront the Internet of Things and other, future interoperable technologies. Although these big questions do not yet have good answers, anticipating and considering them now may help regulators and policy-makers deal with them going forward. Some of these big questions are:• How does society address a proliferation of standards? In several areas, notably e-health, there is now a seemingly ceaseless release of new standards – some of which, no doubt, add value. Many others, however, are conflicting efforts. If standards are meant to bring interoperability to otherwise incompatible approaches, what happens when the sheer number of conflicting standards fragments the market, undermining the original goal of interoperability? In the consumer goods market, the solution to this problem is generally to let the standards compete and let the invisible hand of the market choose among the competing standards. But that approach can be expensive in time and money. It may also delay in innovation as potential market participants wait to invest in the winning standard. For these emerging new technologies and systems, are such costs acceptable? Or, is there a way to accelerate the process of choosing the optimal standards?• How can interop better manage complexity and scale? As described above, higher levels of interop can lead to highly complex systems. This complexity and scale, however, has many costs. It can make it hard to identify and correct failures. It can create security risks and magnify the impact of vulnerabilities. At certain scales, it can even represent a form of lock-in, as network effects become predominant. In many ways, successful interop can be its own worst enemy. How can interop better mitigate these problems in order to capitalize fully on the societal gains of large-scale interop?• How can highly interoperable systems better communicate with end users? As described in this chapter, end users often do not know where to turn when something goes wrong in a highly interoperable system. If a system behaves like a single, cohesive unit, that is both a success for interop and an obfuscation to the end user. This presents a challenge for interop even under the best of circumstances, where the end-user might 118 Trends in Telecommunication Reform 2016