3 ms·
Municipal broadband really isn't the solution here. Sure, it's miles better than private infra, but we down here in NZ have a much better solution. (Not claimin
by ripdog 5y ago
Municipal broadband really isn't the solution here. Sure, it's miles better than private infra, but we down here in NZ have a much better solution. (Not claiming we invented it, of course.)
In 2011, Central government started a big infrastructure project to roll fibre out to ~80% of the population. (Called UFB - ultra fast broadband.) However instead of making this just be state owned, the money was paid as cheap loans with a competitive bid system. Companies bidding had to agree to a significant number of regulations, including a requirement to not act as an ISP and to offer wholesale access to the network for a set fee. They could also offer their own services above and beyond the regulated maximum.
So to start an ISP in NZ, all you need is some national and international transit, and at least one server to do billing and core routing. As a result, we have dozens of major, actually good ISPs and many more niche ones. Most of them service the whole country - at least the UFB service area.
I actually like my ISP. I have IPv6 and a static public IPv4. Peering is excellent and international bandwidth plentiful. Gigabit bandwidth with no usage call and no fair use policy (at least no policy regarding days usage).
Municipal internet is great until you realize you just turned a monopoly into a duopoly. Without competition, what drive your new ISP to do better? If you're super lucky, the goodness of their hearts, I guess. Because you know the old incumbent will just match the new ISP and do not better.
EDIT: Folks in other threads have pointed out how bad US government bodies have historically been at investing in infrastructure maintenance. I'd like to note that the NZ model neatly deals with those concerns too. The companies which build out the networks are made the owners presumably when the loans are paid off. These companies become very profitable even though they can't become ISPs themselves, and are thus heavily incentivized to invest their profits back into their networks to keep them working well and to come up with new products to sell. Already a NZ fibre company has deployed 10gbit fibre commercially, with no extra government support.
- erentz 5y agoThe UFB program was a gift to the incumbent Telecom NZ. Their design was expensive and overwrought as you’d expect from a legacy telco. While it is nice that NZ has semi-decent fiber it’s still worse than my experience of FTTH overseas. Note that there is nothing inherent about muni fiber that prevents it providing the same unbundled layer 2 and dark fiber services. But in your example of Chorus they backhaul customers over layer 2 to your location for you. That’s how your example of one guy with a router works. He puts it in Auckland and everything is backhauled hundreds of miles to it. Munis do not operate backhaul networks, these aren’t state wide networks where a guy with his router can drop it in one pop in the biggest city and serve customers from across the state. This kind of design also becomes incredibly inefficient. The provider Chorus here is really doing most of the heavy lifting and doing it in the least efficient way possible. (Back hauling layer two to POPs hundreds of miles away.) Today all services live at layer 3 and are provided over the top. There’s a good case for just simplifying this and munis just providing the layer 3.
- throwaway2037 5y agoYou wrote: "it’s still worse than my experience of FTTH overseas". Can you be more specific?
- iptrans 5y agoI assume the grandparents refers to the tromboning inherent in the NZ design.
- 650REDHAIR 5y agolol...Nothing drives competition here and the upstarts are squashed by large telcoms who have historically gobbled up government funding and then did very little with it. In the heart of Silicon Valley there are only a couple real options for high speed, reliable internet.