The idea of ‘gatekeepers’ is used in community work, especially rural communities or others where there is a stronger sense of community. Sometimes indirect work with gatekeepers can miss crucial groups of people so direct contact is a good method, just very expensive and labour intensive; but often it is a very powerful way to spread a message. I see parallels with small worlds theory. I wonder whether network scientists can use their ken to find or borrow/unify existing theory on striking this balance of between manageable in-boxes, dismissing spam but having good-guy accessibility, between low order and high order connections, between complusive helpfulness/quest of approval-seeking/overcompensating schema (psychology) and healthy self-protection, for the benefit of progress and their own happiness, nevermind sanity! ]]>

p.s. Thanks for the shout out. One of my favorite papers from the “scale free” network hoopla is R. Tanaka “Scale-Rich Metabolic Networks.” PRL 94 (2005), and not just for the catchy title.

]]>I can’t vouch for either the truth or usefulness of the claims put forward in the book; we all know that power laws can be slippery things.

And, of course, “It’s a power law!” is hardly the end of any story, even if it’s true (that is, even if you haven’t fooled yourself by doing sloppy statistics). For example, everybody loves “scale-free networks”: collections of nodes and links in which the probability that a node has *k* connections falls off as a power law function of *k.* In the jargon, the “degree” of a node is the number of links it has, so a “scale-free” network has a power-law degree distribution. But the degree distribution does not by itself characterize a network! Two networks can be quite different but have identical degree distributions. For example, consider the “clustering coefficient”, defined as the probability that two neighbours of a node will themselves be directly connected. (It measures the “cliquishness” of the network, in a way.) One can build networks with indistinguishable power-law degree distributions but arbitrarily different clustering coefficients.

The NetworkX Python module has a built-in function to do just this: powerlaw_cluster_graph().

]]>However, perhaps laziness is relative – Einstein did seem to achieve quite a bit.

]]>I think *we* would make Einstein feel like a slacker in terms of responding to letters/emails etc. Although I am not sure if I want to actually beat him in terms of correspondence, I mean, there is this matter of actually spending time doing science…

]]>…

…

…sorry. ]]>