http://www.utdallas.edu/~liebowit/palgrave/network.html
In our study of Networks we came across the topic of Network Effects, which was so similar to Network Externalities that I began to use the two terms interchangeably. Little did I know, there is a distinction between the terms, “Network effects should not properly be called network externalities unless the participants in the market fail to internalize these effects…When the owner of a network (or technology) is able to internalize such network effects, they are no longer externalities.” The next topic discussed is with regards that Network Effects alone aren’t enough to explain natural monopolies; these effects have to be greater than the production costs for this situation to occur. After this, common network effects models are challenged, including the stipulation that each additional user provides additional value to participants. It is noted that if additional value is not provided then it would be possible for multiple networks to compete on an even basis. Another way for multiple Networks to exist would be if you drop the homogenous tastes constraint, because if some users much prefer the technology of one network to another then the benefits to them might outweigh the Network Effects. Eventually, the limitations of Network Effects are explored, and it is noted that while it is important for a small group of collaborator to be using compatible software, it would be much less important that the software is compatible with the rest of the world. In industries where Network Effects aren’t that important the main reason a firm comes to dominate is that they simply provide the best product.
This resource goes into greater depth about Network Effects, and begins to drop the conventions of our simplified model. This in effect, begins to explain how certain previously inexplicable cases such as, Wintel Vs Apple, why the Apple network was able to co-exist. Additionally, in many networks, once they achieve a certain size, each new user adds very little additional value to a network. By dropping this constraint we begin to see why positive feedback doesn’t take over in all network situations; a network can saturate before it has captured a whole market. If lecture stimulated your interest in Network Effects and you are looking to explore the subtleties of this topic you would find this paper to be a worthwhile read.