Well, I asked for details of that Hezbollah converged telecoms network, and some appeared via the comments at Abu Muqawama. First of all, there's a map. My first reaction on seeing this was that it looked a lot like a rather underdeveloped, dated official backbone network - there's not much redundancy anywhere, and there's only one path between Beirut and the Beka'a, or between either place and the South, which goes far too close to the southern border for comfort. The long lines follow precisely two routes, each of them along the highway right-of-way; there's more redundancy down south, but it doesn't look like you'd need more than three cuts to thoroughly partition the whole thing. However, breaking up useful comms within the southern area itself would be considerably more difficult; and this does seem to fit with their tactics.
And here's an interesting point - there appear to be some very critical nodes in the upper Litani valley, not far at all from the main Israeli incursion in 2006, and for that matter from where the UNTSO post was destroyed. It's also clear that they have put in several loops right along this sector, close to the border, which should come as no surprise. But the backbone goes that way, too.
Of course, it's far from obvious that the map is honest, accurate, or comprehensive, or that there aren't radio links that thicken it up but aren't shown on the map. "Salem" provides the answer, which is that this network at least is the old official one before Oger took over running it, and Hezbollah took it over. This doesn't mean, however, its capabilities are necessarily the same or even remotely similar.
The single biggest cost in a telco deployment is always the rights-of-way, cell sites, and digging the holes. Essentially, it's Baumol's cost disease, seen from the opposite end - rather than some labour-intensive goods becoming relatively more expensive than capital-intensive ones because they are excluded from technological progress, it's more that the inputs with the greatest manufactured content - switches, routers, Node-Bs - get cheaper due to Moore's law and economies of scale, while the civil engineering doesn't display any such trend. Further, the electronic kit goes obsolete quickly, the bearer - fibre or copper - not for 20 years at least. And the holes in the ground don't - ever.
Pat Lang posted an article of his from the 1970s at the time of the 2006 war entitled "The Best Defense is..." As I recall, there was quite an emphasis on wired comms in that, too. This sort of investment implies a strategy based on the control of land, and an intention to command the informational terrain as well as they do the physical and human sort.