The first round of Capital Budget Amendments are out, and included is a proposal from Scott Resnick to start a Madison Co-op Internet Utility, and a proposal from Paul Soglin for expanding Internet connectivity at “community and cultural” centers The budget amendments don’t get a lot of space to really explain the background information and details, so I thought I’d round up a few documents and put things in one place.
Preliminaries
It’s sort of universally agreed that Internet access is important to fully participate in today’s society. Granted, not everyone wants to and we’re not ready to legally insist on it, but by and large it’s tough to avoid. We’re not satisfied with the connections to the Internet we’ve got now: in 2010, Google’s Gigabit Fiber program sparked the imagination of Madison, and resulted in a nice community response, developed by residents and city staff together.
However, like far too many things in our society, not everyone has equal access to the Internet at even current speeds, much less at speeds far higher than what anyone gets today. The “Digital Divide” is the gulf that exists, for geographic or socio-economic reasons, between those who have easy access to computers and the Internet, and those who do not.
This is especially worrisome in Madison for children in families who are on the wrong side of the Digital Divide and are at risk of falling even farther behind their better-connected peers. To start to address that, last year’s Capital Budget included $150,000 in funding to provide infrastructure to increase Internet access for low-income families and neighborhoods.
In the Capital Budget amendments that were released on Friday, there are two amendments that touch on Internet Access and the Digital Divide. Amendment #4 is from Paul Soglin for $200,000, which would “provide a facility, in several low income neighborhoods, where citizens could go to get access to high-speed Internet service.” Amendment #5 is from Scott Resnick, Mark Clear, and Joe Clausius (and Chris Schmidt, but as Council President he co-sponsors things that he may not necessarily support) that would plan for and create a Madison Co-op Internet Utility. $100,000 in 2015 would fund planning and to carry out several legal requirements State Law insists are done before starting a utility, and then provide $250K in funding in 2016 would provide for some of the Capital requirements of the network.
So, that’s up to potentially 3 projects that work to address the Digital Divide:
- 2014 Project for $150K (reauthorized)
- 2015 “Amendment 4” from Soglin for $200K
- 2015 “Amendment 5” from Resnick et al. for $350K over two years
I don’t want to write much about the 2014 Project because I serve on the committee that’s trying to make recommendations on how to spend that money, so I’ll save my thoughts for committee discussions. (There’s also a lot of great information that the committee has been collecting on the Digital Divide on the city’s Legistar site for the item) The only thing I’ll say is the committee is almost certainly not going to be ready to make recommendations in time for the money to actually be spent in 2014. The proposed 2015 budget re-authorizes the money and there was no amendment proposed to remove it on Friday, so as far as I can tell, this project is still on target to happen. Now, it’s possible that as part of the debate on Tuesday at the Board of Estimates that the re-authorization gets deleted or one of Soglin’s or Resnick’s amendments (or both) gets combined with the 2014 project, so I’ll be watching with great interest on Tuesday to see what happens there.
Soglin’s Proposal
I don’t know much (and so don’t have much to say) about Soglin’s Amendment #4. Here’s the full text:
The purpose of this project is to provide fiber-optic or high-speed point-to-point radio connections to community and cultural centers. The connections will allow the centers to gain access to higher speeds of broadband for community member use. Many families, in the areas where these centers are located, do not have access to high-speed broadband. This project would provide a facility, in several low income neighborhoods, where citizens could go to get access to high-speed Internet service.
It’s a little confusing because my understanding is that most, if not all, neighborhood centers already have (or will have later this year) good Internet Access through MUFN (much more on MUFN later), so I assume he’s talking about things other than the 14 official “Neighborhood Centers” in Madison – ie things like neighborhood churches, clinics, YWCAs, job centers, etc. I know that not all schools are connected yet, so if we’re considering what community assets should we connect, we may want to put some resources there.
(A total aside: every school should be a neighborhood center, open from early in the morning to late at night, for people ages 4 to 104, and busy all the time. We’ve way overreacted on the security front in keeping people out of the buildings. There should be a public library branch on one end, and a public health clinic on the other. More to the point, every school should have a guest wireless network available to anyone, just like the UW, if they don’t already.)
Overall, Soglin’s proposed a fairly straightforward project: identify places people go, provide those places with better Internet access, and as part of the deal make those places share access. Depending on how far the City’s willing to go, there’s no reason the City couldn’t go to local businesses and make the same deal – certainly there are people who use coffeeshops as their primary Internet access, so why not help some of the businesses that are in neighborhoods where it’s tough to make ends meet to offer the same services?
The downside to Soglin’s proposal is that it doesn’t address the core of the Digital Divide: access at home, on their own schedule. People will still have to go to another location to get on the Internet. That’s already the reality for many people today, who may have to go to a Library to use the Internet, or if they’re fortunate enough to have a laptop or other device with WiFi, they can go somewhere with Internet access like Target or McDonalds and get online there, potentially even by just pulling up in the parking lot and getting close to the building if it’s after-hours. Soglin’s plan would bring more of this sort of access to more neighborhoods, which might not even have a full library or a McDonald’s or anything that has complimentary Internet access. However, having to travel means that tasks that might take someone with home Internet access just 30 seconds are still major endeavors, or outright impossible. And as for students, well, I know I usually did my homework in high school outside of the working hours of the public library, and would be at a real disadvantage today without home Internet access.
I’m wondering/hoping if there’s a companion project in the operating budget to support these expanded connections, both in keeping them running, and to keep paying for the Internet connection once the network is installed. Such a project could be helpful for the 2014 project and Resnick’s amendment, too. (As we’ll see, Resnick’s amendment doesn’t need matching operating funds for 2015, though it might in 2016) Finally, there’s also the worry about what resources are available at the facilities that we connect to the Internet to actually use the Internet. Is this just a network connection, or will there be new (or new-ish) computers installed at the facilities for people to use.
All of these concerns doesn’t mean that I think Soglin’s proposal is bad, it’s just important to understand the limitations and what still needs to be addressed. Assuming that the infrastructure spending is done sensibly and can be re-used for a more ubiquitous network in future years, I think this project is worth funding.
The Madison Co-op Internet Utility, Part I
The other proposal, from Resnick, Clear, and Clausius, is more ambitious. It could be complementary to Soglin’s proposal, or it could be a complete alternative. Here’s the full text:
Fund this new project with General Obligation Debt of $100,000 in 2015 and $250,000 in 2016. Add the following narrative for the project: The purpose of this project is to study the feasibility of a Madison cooperative internet utility to provide wireless internet services to low income neighborhoods and families. The study must meet the objectives outlined within Wisconsin Statute §66.0422 to start a municipal internet utility, gather input from existing service providers on cooperative models of internet service, facilitate meetings among non-profit partners such as the University of Wisconsin-Madison and Madison Metropolitan School District, explore relevant businesses models, and index existing resources and infrastructure within the City of Madison to better serve the public beyond the Isthmus. The study should explore LTE and LTE- Advanced options and provide a scalable solution. In 2016, $100,000 of capital funding is allocated for operational software, licensing, back-end network systems, and authentication software. The remaining $150,000 is used for required infrastructure, installation fees, and radio base stations.
There’s a lot going on in there, so let’s try to break it down.
First, in a sense, this does propose to create a public, city-owned, Internet Utility. It’s just not an ISP as most people would think of one.
One thing to address right off the bat – isn’t it illegal for cities to get into the Broadband business? It turns out, no. In 2003, the legislature passed a law (the section 66.0422 bit referenced in the amendment) that set some conditions around a city getting into the broadband business, but didn’t outright ban it. The City Attorney has looked into the rules and is confident that we can meet all of the procedural requirements, though it will take an official study and a set of public hearings that take a while to perform. That’s what much of the 2015 budget would be spent on and why the actual network is pushed out to 2016.
MUFN- The Metropolitan Unified Fiber Network
To explain the co-operative Internet Utility part, I want to talk just a bit about MUFN, the “Metropolitan Unified Fiber Network”, which you have probably heard of but likely don’t know much about. My description is a simplification in many ways, and this presentation might help too.
MUFN is, as the name suggests, a Fiber Network that spans much of the Madison area. MUFN is not something through which you can get directly get Internet access . Instead, it’s a way to get a fiber optic connection from point A to point B. I mean that very literally: if you get a MUFN connection and you shine your laser in one end, a beam of light comes out on the other end, miles across the city. What you do with your laser is up to you.
You can think of MUFN as being able to buy a pipe across the city, and to stretch the analogy a bit, you can put whatever you want into your pipe: water, oil, sewer service, beer, whatever. The other pipes don’t interfere with your pipes. Obviously, it only makes sense for some people to get a pipe across the city, and there’s a big caveat: the pipes don’t directly connect to any larger network of other pipes -it only has two ends. So, for example, Great Dane might be big enough to warrant a pipe of beer from their location downtown to Hilldale, but it’s not a way for the Great Dane to pipe beer directly to consumers.
So too with MUFN: a fiber optic connection from one building to another doesn’t get you on the Internet, it just helps you build your own cross-city network. This makes a lot of sense for something like the City or the School District, which needs to connect multiple buildings together (City Hall, the Libraries, the police and fire stations for the City’s network; the different schools for MMSD). To actually connect to the Internet, data might go over MUFN from a Library to City Hall, where it switches on to the Internet through the City’s existing Internet connection. Similarly, School Internet traffic might go from West High to the Doyle building, where it then goes on to the Internet. School Internet traffic doesn’t mix with City internet traffic on MUFN, they just sometimes go by each other on different strands of fiber in the network.
MUFN is a collection of conduit under the streets, fiber optic cables running through that conduit, and some electronics to help tie everything together. MUFN was largely funded by the Federal Government as part of the 2009 stimulus, but it got a big boost at the start from the City of Madison when the City contributed most of the conduit and fiber that it had already installed over the years. The Federal grant paid for additional conduit and fiber installations around the city, both as part of its “backbone” and as lines to specific sites around the city. The City throws some money into expanding the network each year (though at current rates, it will be decades until everywhere in the city is reachable) and makes a point to install conduit and fiber during street reconstruction: Johnson Street will carry a lot more fiber later this fall. Here’s a map:
What makes MUFN cool is its governance and membership model. Again, it’s not directly for home users, but it’s also not just the city and the schools and other big institutional users. There are commercial partners using it, and using it to provide Internet access to consumers and businesses.
If you’re in the business of being an Internet provider in a city, you don’t run a cable from your headquarters to every house you connect to the Internet. Instead, you build up layer by layer: all the houses in neighborhood connect to a neighborhood office, and all the neighborhood offices connect to a central office, and from there you can connect out to the world. It’s the same concept as the road system: neighborhood streets go to arterials, arterials go highways and Interstates. The exact number of layers and offices is not important – what is important is that MUFN provides the arterials for networks in the city. If you want to start an ISP business, you can sign an agreement with MUFN to provide fiber between your offices around the city. There’s enough fiber in MUFN that multiple ISPs can use MUFN, and it’s way cheaper to pay MUFN than to install your own fiber around the city.
However, MUFN doesn’t go everywhere, and this is where the MUFN governance gets really interesting. As part of the agreement to use MUFN, if a company needs to install extra fiber somewhere, the City gets part of the fiber that’s installed. This is no big deal cost-wise: the expensive part is digging the trench to install the fiber: that’s hundreds of thousands of dollars a mile. The fiber itself is just a few thousand dollars, so if you’re going to dig a trench for fiber, it always makes sense to install a lot of fiber. The MUFN agreement also says that beyond the free fiber the City gets, the City can buy extra fiber when a company using MUFN decides to lay more fiber. (again, that’s essentially at no extra cost to the company) The MUFN agreement also says that the city owns the conduit that gets installed in the public right-of-way. (Which means if the city wants to later add fiber, it can.) It’s a great deal for the city (cheap expansion) and a great deal for the companies using MUFN (cheap access to a very big fiber network with no construction delays.) There’s some extra language in the original MUFN agreement that extends this model to everything that originally got bought with the Federal grant, so if any of the partners go out of business the City and the other partners don’t lose access to that part of the network.
So, MUFN is a great model for the bigger players, and companies can use MUFN as a way to build a network around the city, and connect that network to the Internet. What it doesn’t solve, and why you probably haven’t heard of it, is how to connect houses to the Internet. That brings us to Resnick’s proposal.
The Madison Co-op Internet Utility, Part II
MUFN was one of two projects the city proposed for Stimulus funding. Unfortunately, the companion project, MBI – the Madison Broadband Initiative (or here), was not funded. MUFN was what was called a “middle-mile” network: you used it to get to a different network. To use the streets analogy again, you drive on a big highway to a “middle mile” street like East Washington. Some people live on it, but most people live on a street that connects to East Washington – the so-called “last-mile” streets. (Or, sometimes, “first-mile” streets). MBI was a “last-mile” proposal: how could we get people, businesses, and other organizations on to the Internet at higher speeds for lower costs? MBI would have used a mix of MUFN and additional fiber optic connections to many more buildings/anchor institutions around the city, and deployed a WiFi network attached to that fiber network to cover some of the low-income parts of the city. People in those neighborhoods could have directly connected to that WiFi network and used it to access the Internet. MBI also asked for funding for connectivity for schools and to build out a WiFi network at every school.
Alas, MBI wasn’t funded, and so over the past few years we’ve been using local tax dollars to slowly connect some of the community institutions that MBI might have already connected, and to start rolling out WiFi in the schools. We’re still no where near as connected as we’d be if MBI had been funded 5 years ago, and there’s been no real approach considered for how to connect any home users.
While we’re not as connected as we could have been, in the intervening five years technology has changed and there are some new approaches to consider that weren’t really on the radar in 2009. To be clear, Resnick’s Madison Co-op Internet Utility (MCIU from here on out) isn’t as ambitious as the $20M MBI project, but combined with MUFN expansions and new technology it gets us much of the benefit of MBI for considerably lower cost.
The network that the MCIU is exploring would be more like MUFN than MBI, in that would be a common network that multiple companies and organizations shared, and each company provided access to different users. End-users wouldn’t know that they were somehow sharing a network.
The big difference between MCIU and MUFN, of course, is that MCIU is wireless for its connection. However, unlike MBI, which used WiFi as its wireless connectivity, MCIU will use some variation of LTE. This likely means that users won’t directly connect to the network from their laptops like they do with MadCityBroadband or would have with MBI, but will instead connect through a sort of “MiFi” mobile hotspot that talks WiFi to the computers in the house, and LTE to the MCIU network. That should help give a much better experience than neighborhood WiFi networks have fared. A reasonable goal should be a 20-40 megabit connection for about a $50 end-user connection cost, plus whatever we need to charge for monthly costs.
The “Co-op” part is how MCIU is a lot like MUFN. No one will buy Internet access directly from MCIU/the City of Madison. Instead, companies will become members of the MCIU, and use the wireless network to connect to their customers. A person might buy their Internet access from Purgolder Wireless or Regent Internet (both fictional), both of which use the MCIU signal. However, Purgolder Wireless might turn around and use AT&T to connect to the main Internet backbones in Chicago, whereas Regent Internet might use Verizon to connect to an Internet backbone in Minneapolis. The MCIU will take care of ensuring that data from Purgolder Customers gets to Purgolder, and Regent customers to Regent – that’s the $100,000 software as part of the 2016 funding.
MCIU will at the start probably not cover the entire city, only places where there is both a compelling need for Internet access and there’s already infrastructure in place to support installing the LTE base stations – think libraries or water towers the City already owns, has a fiber connection to, and doesn’t have to pay any leasing costs. It could start small and then scale up. This isn’t entirely pie-in-the-sky, there are people at the UW who have some experience in running such a network.
However, there are some opportunities to expand coverage and bring in additional partners. The UW could get involved, and use the MCIU network to extend the UW network to students off campus. The school district could get involved, and use MCIU to give students access to the MMSD network (and through it, the Internet) for fairly low cost. In the city, there could be some cost savings as the city uses the MCIU as one of the networks on its police and fire service (for safety, the city uses multiple networks in case any of them goes down)
The MCIU proposal is somewhat complicated, which is part of why it needs a year of study and the resources to actually do the study. The hardware costs are fairly straightforward and aren’t too expensive, but what business model works best is something that has to be explored. Should MCIU charge co-op members a flat rate per subscriber, or should it be usage-based? Should the co-op members charge their users for a set amount of data like cell phone plans, or should they say “unlimited” but throttle when the network gets busy? Are there area businesses that would like to offer service over the network? Do we need additional money in the operating budget to support it? Can we work with MATC or the MMSD to build an apprenticeship helpdesk to assist in running the network day-to-day? Running the network itself isn’t that expensive – power’s cheap, the software mostly just works, but it doesn’t take more than a couple of people working full time answering phone support to quickly drive up the cost.
So, could there be some kind of hybrid model, where the wireless cost is low but includes no support, and we encourage very small resellers to take on the support costs? Imagine an MCIU cost of only $10 a subscriber, but is resold at $20 and handled by someone who lives in your neighborhood, with many, many neighborhood-sized ISPs. You could easily imagine that the MCIU only offers “second-level” support that costs money per call, and only takes calls from certified partners. My guess is that something like 80% or 90% of problems can be solved without needing to call MCIU (it’s just problems with Windows 8 that are easy to fix) you could easily imagine that people could earn a living wage even just getting started and could later parlay the experience into another, more in-depth job later.
Resnick’s proposal also faces a problem that Soglin’s proposal may face: getting Internet access is all well and good, but just as if there’s no working computers at the neighborhood center, if people don’t have a computer at home, a MiFi connection doesn’t do them any good.
The MCIU proposal is one of the most interesting things to come along in Madison broadband in a while, and is an opportunity to make an impact on many Madisonians who are currently left out. I hope the Council funds the study, and that the study says there’s a solid path forward that we can deploy in the next few months and years.
