One of the ways a hub based network can be grown is through “preferential attachment”. When a new node is added to a network, the links from the new node will have a higher likelihood of being attached to nodes that already have many links. For example, if the new node initially starts out with two links, these links will have a higher probability of being attached to nodes with many links. This does not mean that a link will not be attached to a node with only a few existing links, just that it is less probable.
Preferential attachment will favor older nodes, since they will have had an opportunity to collect links. One of the examples of a naturally occurring preferential attachment network in Linked is in networks formed from journal article citations. Early journal articles on a given topic more likely to be cited. Once cited, this material is more likely to be cited again in new articles, so original articles in a field have a higher likelihood of becoming hubs in a network of references.
Barabási-Albert network model
Pingback: More Richness | Networked Media