1) Buyers have been disaggregated into purchasing firms and countries. This creates many more nodes, but better reflects the buy-sell dynamic. Note that now a single firm (e.g. ecosecurities) can be represented as two nodes (as a buyer and a consultant).2) Validators have been added as well (as red nodes, consultants are now blue).3) I normalized the network matrix in a way that is relatively standard and helps eliminate great disparities in weights.4) The size of the nodes is proportional to a statistic called betweenness centrality. This is one measure that might be useful in studying the oligopoly/oligopsony issue.5) The layout is constructed using the weights and hence is a reasonable good two dimensional representation of the relationships between the various actors. I've included it as a pdf - you should be able to zoom in as far as you like and still have legible labels.Two initial observations:1) The host countries are grouped geographically. 2) There is (obviously) a main clump and a peripheral one (as well as many peripheral small clumps). This invites some further study.