Future-Proofing Texas Schools with Better Internet Infrastructure- Episode 639 of the Community Broadband Bits Podcast

Michael Keough (00:07):
I don't think schools that are running poor or low speed Internet services are going to realize how important it is to run higher level stuff until it's too late and now they're trying to catch up and catching up. It's going to take several years.

Christopher Mitchell (00:20):
Welcome to another episode of the Community Broadband Bits podcast. I'm Christopher Mitchell at the Institute for Local [00:00:30] in St. Paul, Minnesota, and I'm speaking with Michael Keough, who is the CIO, Chief Information Officer, for Region 16 Education Service Center. Welcome, Michael.

Michael Keough (00:40):
Hey, thanks for having me. It's great to be here.

Christopher Mitchell (00:42):
I'm excited to talk to you and a few people that are listening to this show might have just heard you on a Shelby call, the School's Healthcare Library Broadband Coalition, and brought to my attention that you had this very interesting story about connecting Community Anchor Institutions there in, I [00:01:00] want to say North Texas, right? Is that the right way to describe it?

Michael Keough (01:02):
Yeah, so it's actually the Panhandle of Texas, but we always call it North Texas at the top, most of Texas. So yeah, it's the panhandle, but it can be referred to as North.

Christopher Mitchell (01:11):
Excellent. So Region 16, when I hear regions, I almost always think Colorado. I'm most familiar with their region system, but what does Region 16 encompass?

Michael Keough (01:21):
So most states have education service centers or education service agencies. We're basically just an entity created by the state that is distinctly created to serve [00:01:30] K through 12 schools. And so in Texas, the mandate is to serve education and to foster that through public and private need. And so we get to do a lot of work not only working with the schools, but working solutions that would be a general community benefit that ultimately tie back to school directly or indirectly.

Christopher Mitchell (01:47):
About how many schools are we talking about over what geographic area?

Michael Keough (01:51):
We serve 60 schools over 26,000 square miles.

Christopher Mitchell (01:55):
Wow,

Michael Keough (01:56):
That's a big region.

Christopher Mitchell (01:58):
And then your responsibility goes well beyond [00:02:00] connectivity, right? This is a small part. You have to worry about all kinds of information, technology type things, I'm guessing.

Michael Keough (02:06):
I do. It seems like this is eating up quite a bit of my job. But yes, we do everything on the technology side from managed services, help desk, cybersecurity. We are an Internet provider, which is part of this, which we're actually moving out of that provider space to more of a collaborative space. And then we actually have a programming initiative where we build products that we sell throughout the country.

Christopher Mitchell (02:26):
Let's talk about the Internet access stuff. And I think for people who [00:02:30] aren't familiar, let me guess that you have kind of a system in which you have a network that connects the schools to each other so they can exchange information with some kind of a private ring, and then all the Internet access kind of goes into that ring from a couple of access points. I would guess maybe one or maybe a few.

Michael Keough (02:47):
That's accurate. It's actually not even quite that good. So at this point today we have what I would more call a Hub and Spoke type network. So we use various different providers and we're a central location and we'll [00:03:00] do a spoke to a school, and sometimes those spokes are kind of, they route around to several schools. So one spoke might hit three or five or 10 schools in the line. So if that fiber gets cut, all 10 of those schools lose Internet access. So there is no resiliency in the panhandle, and there actually is quite a bit of fiber cuts. Most of our schools run one Gig, a few of them run as low as a hundred to 200 meg.

Christopher Mitchell (03:22):
Wow. I've heard this because Shelby collects the statistics at this point. Is that a significant challenge [00:03:30] for the school to be offering a modern education if they have that level of connectivity? When it works even?

Michael Keough (03:35):
It's not only drastic now, it's going to be drastic as they come. And I don't think schools that are running poor or low speed Internet services are going to realize how important it is to run higher level stuff until it's too late and now they're trying to catch up and catching up. It's going to take several years to do the right construction, do the right processes. So we're trying to get well ahead of this with some of these modern infrastructure things that are coming.

Christopher Mitchell (03:59):
Okay. [00:04:00] So you said that this is taking up an increasing amount of your time, and I would hope that in five years that's not the case. But then you listed something and I'm curious, is it mostly cybersecurity that is growing or is it also the communications, the physical communications that's taking up your brain space?

Michael Keough (04:17):
It's all of it. So I do a lot of presentations. I travel significantly probably once or twice a month. I'm in a different state or I'm somewhere in Texas. The project also is taking a lot of information to get [00:04:30] out the idea of what we're doing so that not only our state and local government, but even our federal government and the FCC, of course Shelby all knows what we're doing, so that not only can it be possibly replicated, but it brings up a level of transparency to see what we're doing. So the communication side, yes, I spent a significant amount of time creating material, speaking at events, doing podcasts, things like that to talk about what we're doing On the cybersecurity side. It's one of the most critical things facing our schools today, and [00:05:00] our schools are nowhere near prepared. So we have a cybersecurity department that manages cybersecurity for our schools, but we are trying to get out as much advocacy as we can to our schools to fully understand that they are a key threat. One of the top, I believe, three to five threats in the nation is public school systems because you have kids up to 18 years old who have raw social security numbers that are never used, and if you can compromise that data early on, they get out of high school and they don't even know they've been compromised. So centralizing some [00:05:30] of this as part of a regional security operation center is definitely part of the initiatives.

Christopher Mitchell (05:35):
So when you're talking about documenting what you're doing, what are you doing? And I think the next question then is going to be, I'll warn you for what are the challenges. So what are you doing?

Michael Keough (05:45):
Let's start at the beginning. So we have been an Internet vendor, just like any other company that provides Internet to school. We buy it at a price, we sell it at a price and we make a revenue. And for us, it's been a significant markup and it's been a revenue [00:06:00] stream. We've been able to, as an organization, just as any organization would do, we take that revenue and we put it back into an organization that helps serve our customers. So whether it be our school districts or other services, we find ways to do that. What we found was just especially as the type of organization we are, that process just didn't seem right to the new leadership. So the new leadership being me, my executive director, our CFO, we stepped back and said, okay, just because done this for 15 years and just because it's been about [00:06:30] a $2 million revenue for us, doesn't mean this is the right thing to do for schools and it's not necessarily even the right thing to do for the E-Rate program.

(06:39):
So we stepped back and said, okay, how can we do this differently? And we found, we looked at several different models, but the model we landed on was the best for schools. It's best for the E-Rate program, it's best for every kid we serve. It's not so great for us. So we lose $2 million of revenue a year to do this process, [00:07:00] and we also lose a $10 million investment, which is our portion that we have to invest on the E-Rate side of things to make this happen. So why would you do that? It's a completely negative business model, but we looked back at it and said, this is the right thing to make sure our kids in these 60 school districts are empowered to do everything they need to do and have the same quality of education and resources with these modern needs that any school district could have in Dallas or Austin or anywhere else in the country. [00:07:30] So that's the purpose and that's why we're doing it.

Christopher Mitchell (07:33):
For people who don't live and breathe this, I think they might be confused because they hear that you're getting Internet access and distributing it to the schools and selling it to them. And I think the question is, okay, well what is the new plan that makes it so different from the old plan?

Michael Keough (07:48):
So with the new plan, we're actually going to give them 10 Gigs of resiliency. So resiliency, just meaning that before they had one path to the school in the resilient model, it's a path [00:08:00] around the panhandle where multiple schools are connected on a loop. So if the lion's cut, it just goes the opposite direction. A school district doesn't go down. So even if a town loses Internet access, the school district stays up. But even this project will actually help the town indirectly as well.

Christopher Mitchell (08:16):
My friend Mary St. John from Calix is fond of saying, we do these events at tribal broadband bootcamps, and she's been at several, and she always says, when it comes to your data, if you like it, put it on a ring, like a Beyonce song, put a ring on it, and I always [00:08:30] brutalize it. She does it much better, but if you like it, put it on a ring.

Michael Keough (08:35):
That's right. And there's so much truth in that. One of the other things that go along with this, and one of the conflicts we've had is that we will publish. We're doing this, well, every city in the panhandle of Texas has some form of Internet access. There's nobody that's just, there's marginalized communities and there's people that don't have the level of access, but they have access to the Internet. And so you say, Hey, we're putting a 38 million project. When it's all said and done, it'd be about 45 [00:09:00] or so million dollars that's going in the ground of fiber and we're doing all this connectivity. And people are like, well, you're just overbuilding. We already have fiber. What they don't understand is that the fiber that we have today in the ground is not the fiber we need in the future. So when we start thinking infrastructure 4.0, quantum networking, even modern cities, we're living in an era where even the whole United States lives in an infrastructure that's almost 80 years old [00:09:30] back in the 1940s.

(09:32):
And so in order to live in a modern infrastructure and keep up with China and other countries, especially Japan, Korea, we have to start putting infrastructure in the ground and building things in such a manner that not only prepares for the current need, which actually isn't being met, but also the future need with things like AI and quantum and things like that. With the current need, people get their 200 megs of Internet at their house, then they get cuts three or four times a year [00:10:00] and they're like, ah, it's okay. We deal with it. I go get a cup of coffee or do something else on Internet comes back up. What they don't understand is that businesses, especially businesses that want to come into the panhandle, they can't afford 10 minutes of downtime, let alone three days. I recently had a fiber cut that the cut happened on a Friday afternoon, and the company that owns that fiber, a well-known company, you probably know the name, they didn't actually get to fixing that fiber until Tuesday.

Christopher Mitchell (10:29):
If you have a [00:10:30] company that's doing manufacturing or people don't always realize this, and I've used it before as an example, but if you have a company that's in the trucking business,

Ry Marcattilio (10:39):
If

Christopher Mitchell (10:39):
They lose Internet for 10 minutes, they can literally be missing out on hundreds of thousands of dollars of RFPs, RFPs request for proposals where the bidding process for getting trucking jobs and things like that. It is not a matter of being like, oh, now my employees aren't as productive. It's everything grinds to a halt in a number of cases.

Michael Keough (10:58):
That's right. And that intrinsically [00:11:00] damages economic development opportunities for growth. And even from a school model, and this is where my interaction with the schools come in, you build business in the area, it brings kids, the kids grow schools, that creates more funding opportunities and more workforce development, and it just feeds itself.

Christopher Mitchell (11:18):
Yeah, it's a common thing that I've heard, which is, oh, there's already fiber in the area, and I have a negative reaction to it, but it's one tempered [00:11:30] by knowing that there are many different circumstances. One is there's possibly a private company that's there and they're doing the best they can. And at a certain point I feel like we want to figure out how to help those businesses succeed, but sometimes they're not in a position to deliver the connectivity that the institutions need public safety needs and others that need. And so when it comes to this sort of public services, to me, it really makes sense for counties [00:12:00] and organizations that are larger than counties to look at what they own. And I'll just use it as an example, like Dakota County here in Minnesota, I've talked about frequently more than 30 years ago, probably on the order of 30 years ago, they put 12 count fiber to a lot of the schools.

(12:18):
A lot of the schools were involved in that. They put it in conduit and then over a weekend they would go to a school each weekend and they'd rip out the 12 fibers and put in 144 back like 15 years ago. [00:12:30] That's probably still enough for today, but the fact that they owned it and they could do that, they didn't have to pay anyone. They paid a couple people to go out there and do it. They didn't pay hundreds of thousands of dollars for an upgrade because they controlled it. And now every intersection, I mean almost every intersection in what is soon to be the second most populous county in Minnesota has Internet access. The Department of Transportation basically doesn't even pay for because they've built this and they've made sure that they can provide these public services. The schools have great access, [00:13:00] and so that's the level of infrastructure that we need. And you can't always get it from a private company or even a bunch of private companies.

Michael Keough (13:10):
That's right. The fiber we're putting in, and I think this is very important, and there are people that will vehemently disagree with me on this, and they have very much so, and most of these are private companies, but we are putting the fiber in open access, and I think especially as a government entity or even a nonprofit or whatever you are, if you can put in fiber [00:13:30] and open access, it forces the tier two provider or tier one provider. I think in this case it's tier two to sell it to all the private companies, a wholesale rate. So what we're seeing in our area in select positions, we have people paying as much as 150 to $200 a month for one Gig Internet that's higher than most of the country. And then sitting in Amarillo, I pay 70 bucks. [00:14:00] There's some people that pay $45. So there's a big discrepancy there, but when you do this type of model, it builds a cohesion where pricing can go down and competition can go up.

Christopher Mitchell (14:12):
Yeah, that's the hope. Now from a reliability standpoint, I'm curious, have you had situations where, let me just ask directly. Is there a horror story of the schools going down for some period of time or

Michael Keough (14:26):
Not just one multiple times a year, [00:14:30] and it's not just one school. Like I said, many of these are daisy chains, so it's like one fiber serves, and when I say one fiber, I'm talking, it doesn't just serve the school. It's in the same tubing or whatever that connects five cities. So I have at least two to three times a year where a quarter of that 26,000 mile area literally goes dark. I'm not talking just the school, I'm talking cities, businesses, homes lose Internet for 12 hours, and that's a common scenario.

Christopher Mitchell (14:59):
I [00:15:00] remember when we spoke years ago with the guy who was running the St Vrain school district, St Vrain Valley, up in the front range of Colorado, and I asked him, they were moving from a situation where they were leasing fiber and getting Internet service from a major international provider to where they controlled it all themselves and were their own ISP effectively. And I asked whether he was concerned about additional costs of operating [00:15:30] to do that, and he said he estimated it would actually result in significant cost savings because when you set these networks up, I mean, it's not to say it's easy, but it doesn't require a lot of staff time to keep it running to generally things are working well and if they break, you can fix it. But his main point that I'm getting to was that he said, I would rather have my people fixing the problem than sitting there on hold waiting for another company to respond.

Michael Keough (15:59):
Right, a hundred percent.

Christopher Mitchell (16:00):
[00:16:00] So that's something that resonates with you. This is something that I hope would save funds in the long term then?

Michael Keough (16:06):
Oh, I believe it will. Number one, assuming all things hold well with the Universal Services Fund and the E-Rate program, this will save that program millions. It'll save our school districts locally roughly $320,000 collectively a year. And then really more than that, that's what it would save them today's dollars. But keep in mind, most of them are running between [00:16:30] a hundred meg and a Gig of Internet, and we're converting everybody. It doesn't matter if the school has a hundred kids or if it has 10,000 kids, everybody's getting a minimum of 10 Gigs, which is way overkill for a 50 kid school or even a 500 kid school, but we don't care. We're going to make sure they have that level of speed. And then at our core, we have anywhere from 200 to 600 gauge whatever we need to support that build out. So yeah.

Christopher Mitchell (16:55):
So if I understand the E-Rate program correctly, the way that you're likely [00:17:00] working through it is that E-Rate pays for building out the fibers that are needed for the schools, and then you're able to, through the Region 16 education service center, use other funds that you have to pay for additional fibers that you're then going to make available on an open access basis as well as using for other needs, I'm sure. Right?

Michael Keough (17:23):
Kind of Let me refine that a bit. So we are forming the consortium and we're building dedicated fiber in what we're calling a distance [00:17:30] learning network to connect each school as a campus of the Region 16 entity, which is centered directly in the middle. So with that model, every campus gets a 10 Gig fiber directly to our building, and that is fully funded by E-Rate. And so what E-Rate carries about there is that my two strand or my pair of, or two pair of fiber that's going to those individual schools that is dedicated to the schools. I can't put any more fiber in than what I've paid for with that, but the majority [00:18:00] of the cost to build fiber is not the two pair of fiber that we're putting in or even a bundle of 600 pair of fiber. It's the construction and the right of ways and all that time and effort and the manpower to actually get that fiber in the ground.

(18:14):
What E-Rate allows is that if you're doing that construction for a school, the fiber company can put more fiber in the ground as you're doing the construction. So that's where I say it's a $38 million project, but in the end it's probably going to be a 45 or 50 million project because the company Fiberlight [00:18:30] is going to have it. They've already expressed to us they're going to have a significantly bigger bundle of fiber than what we're paying for. And that's the indirect benefit where communities will get fiber. They never could have paid for no private company, no city or municipality or nonprofit could have purchased this network had we not been doing it for schools in the first place.

Christopher Mitchell (18:53):
Right. It's hard to justify putting that much fiber in the areas of such low [00:19:00] density where there's a lot of people because there's a very large area, but they're spread out. So is there anything else we haven't talked about that we should cover in terms of the modern challenges of making sure that the schools are well connected?

Michael Keough (19:14):
So there's another component of this that we didn't know would be a part of it, but it did end up being a critical part of it. Obviously making a loop is important, so we know that a loop is better than a hub and spoke, which is why we designed the network that way. But loops still [00:19:30] aren't perfect. You still have laterals. You still have some people that are off on a deeper lateral where they may not be as close to the loop as others.

Christopher Mitchell (19:39):
So if I could just clarify, so if people are on the loop or the ring, then they're covered if there's a cut, if there's one problem because they're still connected, but if you're off a lateral off of that, a single cut could still isolate you without connectivity.

Michael Keough (19:53):
It could. Absolutely. So that obviously is a bit of a challenge. And then on top of that, there's a latency [00:20:00] issue, so you have all this fiber and you have trying to get it back to the HubSpots in Dallas or Denver, Houston, there's a latency issue there as well. So we found a company, we partnered with them actually, it's called Duos Edge AI, and they put in these pod or edge data centers. In fact, we just put one behind our building. These are a million dollar data centers, and we didn't pay a dime for it. The company put it in. We traded some land to get a mutual benefit there, but our goal is to put these [00:20:30] all around the panhandle, all along our fiber in small towns and big towns and what it's going to do, it's going to sit on the fiber we're putting in, so it's going to be a direct benefit to the schools from a disaster recovery emergency response.

(20:43):
These are all SOC two compliant, but where it really gets interesting is when you get providers, small companies in the area, whether it's something as big as a big massive telco or whether it's just a small hometown, we serve three communities, telco and all those telcos, [00:21:00] colo, their fiber in these data centers we're putting around the panhandle, and now you have not a loop network, but more of a mesh network. You create interoperability between providers where if one provider goes down, it's not unfeasible to say, Hey, we're down over here, but I could go this direction on your network and we can get back home. And we say up, and by the way, we're going to give you the same benefit. If your fiber goes down, you can come down our network. As long as you have some collaborative [00:21:30] spirit and probably a little bit of money changing hands, there's a way to make that happen where it's even more impactful than a loop and it's costing us nothing to make that happen. Just the right partnership with the right company.

Christopher Mitchell (21:42):
Does that also likely include the ability to put in some Internet exchange type where you put in a Netflix box and AWS box and things like that so they don't have to go out to the wider Internet for those resources they can keep [00:22:00] 'em local then within Texas?

Michael Keough (22:02):
Yeah. It's funny you say that because I actually have a meeting tomorrow with AWS and one of our goals with that meeting, and we've told 'em about the data center project. We've told 'em about the fiber project and we want to see an AWS, and they're very interested in AWS being in our core pod right here in Amarillo. So all the pods would connect back to the core pod, and then we would have a sort of AWS backbone presence where definitely our schools could carve off of the main [00:22:30] network and then they could get into the AWS network for AWS specific things. But we want to do the same thing with other providers, probably not for the school side on Netflix, but still love to see them in there. Google, we have a great relationship with Google, of course, Microsoft, so there's a lot of hyperscaler providers that could be in these pods and create a benefit in the panhandle, and most people don't really understand what it means to be riding on an Internet and connect it directly to one of these big cloud-based companies, [00:23:00] but it would be a game changer. You would have significantly more responsive Internet to those services that are hosted by those providers,

Christopher Mitchell (23:08):
And hopefully you'd be able to help those providers out that are not thrilled about your investment and cut their operating costs and hopefully help them be more competitive so they would still get something to gain, I hope.

Michael Keough (23:20):
Absolutely. This is the panhandle of Texas. We all support each other, and I think that that's a model that can be replicated, and what we want people to know is that, hey, this is a project that is meant [00:23:30] to be, not everybody's going to win big, but everybody we're hoping will win. We're not trying to give one benefit to one organization, and at the end of the day, the only reason we are doing this is to support schools, but we're designing in such a way that in that support of the school, that intrinsic indirect benefit to a community is going to be something that builds our community, builds school districts, and builds our ability to generate some economic and workforce development.

Christopher Mitchell (23:56):
Great. Anything else then to still add on to that?

Michael Keough (24:00):
[00:24:00] I think that's the big focus right now. The next step for us once this is in is trying to work on that security piece and trying to get some centralized security operations that are sitting right in our house where all those schools that connect to us can also get an even deeper security benefit.

Christopher Mitchell (24:17):
Good. Well, I'm glad that you're not so overloaded with the cybersecurity challenges that you're still able to focus on these long-term investments that will really pay off. Thank you so much for sharing it with us today.

Michael Keough (24:29):
Truly my pleasure. Thank [00:24:30] you so much.

Ry Marcattilio (24:31):
We have transcripts for this and other podcasts available@communitynets.org slash broadbandbits. Email us at podcast@communitynets.org with your ideas For the show, follow Chris on Blue Sky. His handle is at Sport Shot Chris. Follow communitynets.org stories on Blue Sky. The handle is at communitynets. Subscribe to this and other podcasts from ILSR, including building Local Power, local Energy Rules, and the Composting [00:25:00] for Community Podcast. You can access them anywhere you get your podcasts. You can catch the latest important research from all of our initiatives if you subscribe to our monthly newsletter@ilsr.org. While you're there, please take a moment to donate your support in any amount. Keeps us going. Thank you to Arnie Sby for the song Warm Duck Shuffle, licensed through Creative Commons.