The proprietary protocols have led to another problem in IM technology -- the centralized IM networks. Most IM clients connect to a centralized server or server pool which is owned and operated by one authority. This basically means that an IM service provider "owns" the entire IM network, and clients participating in the network must bow to the policies and wishes of the providers.
While having a centralized authority is good for ensuring that IM standards are well-defined, the world will never be able to agree on a single company to be in charge of owning and operating the world's IM network. Hence, various competing networks have sprung into existence, each one luring as many clients as possible into its proprietary network and trying to become the world standard in IM technology.
Ironically, the fierce competition between IM providers has produced some problems as well. As people become interested in IM technology, they usually percieve a need to choose between the various IM networks. In most cases, however, the choice is already made for them, because to the new user, the value of a given network is not determined by its technological features, but rather by the number of friends and family that are already participating in the network. For an IM network to have any value at all, the people with whom a client intends to communicate must already be registered on that network.
In addition to these "social" problems, there is at least one great technological problem with centralized networks. If the central server fails, everyone loses their ability to communicate. Users of every major IM service have experienced this frustrating phenomenon. IM networks should not have a single point of failure.
Because IM is really nothing more than accelerated, interactive email, it makes no more sense for a single authority to own an IM network than it does for that authority to own the internet's email system. What the world really needs is a distributed IM system operated by multiple authorities with only local points of failure.