In our experience, just about every community considering building a community network considers open access. They want to enable new choices for services and often would prefer the local government avoid directly competing with existing service providers, for a variety of reasons. However, we are only tracking 30 open access networks on our just-released Open Access resource page.
Many of the communities that start off enthusiastic about open access ultimately decide to have a single service provider (themselves or a contractor) to have more certainty over the revenues needed to pay operating expenses and debt. We believe this will change as the technology matures and more communities embrace software-defined networks (SDN) -- but before tackling that topic, we think it is important to discuss the meaning of open access.
On a regular basis, I get an email from one deep-thinking person or another that says, "That network isn't really open access." They almost always make good points. The problem is that different people embrace open access for different reasons - they often have different expectations of outcomes. Understanding that is key to evaluating open access.
How Many ISPs?
One of the key questions centers on how many providers a household is likely to be able to choose from. Various factors, including the network architecture and economics of becoming a service provider, will influence this outcome.
Some communities simply seek to avoid a monopoly network - they are focused on the idea of potential competition. For instance, we believe Huntsville's model and agreement with Google can be considered open access because any party could lease fiber from the utility to compete with Google. However, we believe the costs of doing so by using that network architecture make robust competition unlikely.
If Google is a strong competitor in Huntsville, they will likely not face significant competition from other ISPs on the utility fiber though AT&T and Comcast will still use their networks to compete. But in the event that Google is not a strong competitor, the door will be open to other ISPs to give people a better choice. It is extremely unlikely that this arrangement would give residents many choices for Internet access, but it is an improvement over the one or two pathetic incumbent options most of us face. Google is left with an incentive to meet user expectations, knowing that it could face competition if people are unsatisfied.
The UTOPIA model has resulted in many more choices for both businesses and residents, but most of those businesses are offering similar services at similar prices. The fact that it does not carry a "marquee" provider like Google or a national cable company on it may make brand awareness (and therefore marketing) more difficult, but it also provides opportunities for excellent local firms like XMission to thrive.
This leads into a second question: can a premise subscribe to multiple service providers simultaneously or do they have to choose one? This may sound like a dumb question at first - why would you want to subscribe to two different ISPs? Aside from perhaps wanting video or phone services from one and Internet access from another, many are hoping to see more innovation on this front. We have written frequently on Ammon, Idaho, because they are doing some of the best work in this regard.
The ability to offer simultaneous services depends greatly on the underlying technology. Not all FTTH networks can give ISPs the tools they need to have confidence in delivering a high quality product reliably to their subscribers. Communities that want to ensure they have this capacity should pick a consultant that deeply understands these issues and has worked previously on open access.
The Holy Grail among those who prioritize this flavor of open access is to make it very easy for network subscribers to manage their own subscriptions - changing providers on the fly (and again, see Ammon for a model). This approach would allow ISPs to specialize and greatly encourage innovation, particularly for niche services. You might subscribe to an ISP that specializes in great connections for video games while also having a part of your connection dedicated to a home alarm system and still be able to initiate a high quality teleconference for health care that wasn't transported on the public Internet.
Market Entry Costs and Consequences
A key question about open access comes down to market entry costs. How much will it cost an ISP to serve potential subscribers? In Huntsville, the costs of building drops suggests it will still cost hundreds of dollars per sub, which is less than the $1,000+ per sub that it would likely cost to build a network from scratch.
We would generally expect that the lower the cost for an ISP to connect subscribers, the more ISPs would be on the network. However, there is an initially surprising problem that can arise when the cost to offer services is very low, something occasionally called "ruinous competition." This is used to various levels of seriousness but represents a common economic problem: if a product has little differentiation (like an ISP offering only Internet access), then subscribers are likely to decide on the ISP based solely on cost. Over time, ISPs will cut prices until the margin all but disappears, which runs most of the providers out of business (or they consolidate) and the competition effectively disappears.
One of the key points of the “ruinous competition” problem is whether ISPs are effectively providing the same thing (generic Internet access) or services (home security, remote backup, help desk, telemedicine, etc.). To the extent that they are offering different kinds of services, we can avoid that problem. However, it is not clear that most networks today are technically capable of allowing service providers to differentiate their services in any significant way, which is again why Ammon's forward-thinking software-defined networking approach is so important.
Like other aspects of technology, open access will evolve with innovation. For now, open access means different things to different people who are often seeking different outcomes.