Broadband at a Crossroads: Evan Feinman on Getting BEAD Right - Episode 645 of the Community Broadband Bits Podcast

Evan Feinman (00:07):
The really unpleasant working conditions that my former folks are experiencing now is going to drive a bunch of them out of NTIA and NTIA is not going to be permitted to backfill their positions and instead the private sector is going to still have the same ticking clock that they had before, but they're not going to have any support in solving these government problems.

Christopher Mitchell (00:27):
Welcome to another episode [00:00:30] of the Community Broadband Bits podcast. I'm Christopher Mitchell at the Institute for Local. I'm in St. Paul, Minnesota. We might get rocked by thunderstorms at any second, but assuming I stay online, I'm going to be talking with Evan Feinman, who is the former BEAD director at NTIA. Welcome back Evan. Thanks for coming back.

Evan Feinman (00:51):
Thanks for having me, Chris. Really excited to be here to talk again.

Christopher Mitchell (00:54):
So we talked before and it was as fast to publication as we could make it. This one is a little [00:01:00] bit more of a retrospective and we're not going to rush it out the door, and so we're going to be talking about things that are a little less topical and more about lessons learned and responding to some criticism and just getting a sense of what actually goes into building a program like BEAD. So let me start off though by asking you, and this might be overtaken by the news, one of the things that I've seen from people is what we talked about before, which was just this idea that what you're talking [00:01:30] about is lunacy Evan and that we will not see the Trump administration significantly change BEAD. And so in the week since we talked last, have you seen any signs to give you more of a sense one way or another how you think they're going to end up?

Evan Feinman (01:46):
No, I have not seen new indications of what the administration is going to do. I mean, I will point out that as we said before, what we know the goal is, is to serve more locations with LEO and less locations with [00:02:00] fiber. That is explicitly what Secretary Lutnick directed NTIA to do was to redesign the program to do that. They are now going, they're now in the process of sort of figuring out where they're going to take away discretion from states. Right? Right now in the program, states get to decide what the right project is and the right technology for the right locations. The Trump administration is planning to take that discretion away from states and where they set [00:02:30] that economic breakpoint where they say you've got to go least cost above X, Y, or Z figure is the critical component. If they set it at a figure that is quite low, $5,000 was the driver of the analysis that we talked about before, which shifts huge percentages to lower orbit satellite that will have a dramatic impact on B, if they set it at 25,000, that will [00:03:00] have a very minimal impact funding and ultimately have burned half a year for no real purpose.

(03:07):
It is also worth noting though that even a minimal impact, right, if you've got millions and millions of households, yeah, sure, only shifting five 10% of those households off of fiber and on LEO that wouldn't otherwise have shifted that way is small in the context of the program, but we're still talking about tens of thousands of American families, multiple American communities that [00:03:30] are effectively going to be left out of future economic prosperity as a result of that change. So it's not nothing, it's just smaller than it might've been otherwise.

Christopher Mitchell (03:38):
As someone who works for the Institute for Local Self-Reliance, I am an expert on one thing perhaps, and that is watching as different people in power, different parties in power, we'll say This is a decision we're going to make ourselves, and this is a decision that we'll let communities make. In this case NTIA the way BEAD was designed, [00:04:00] you were not deciding the technology mix. I think people need to just reiterate that you said it, we've said it before, but the states were deciding and in order to put a thumb on the scale at the federal level, they have to take that decision away from the states and that's what we're watching for.

Evan Feinman (04:16):
Yes, and that's really concerning because the reason prior federal programs didn't work is because they were top down one size fits all folks in DC making decisions [00:04:30] for but not with communities and states. The whole point of BEAD, part of the reason it was slower to roll out is that this is a program that is implemented primarily by the states with federal guidance and sign off and the moving back to toward an RDOF model is a terrible idea. The biggest flaws in RDOF were their reverse auction, which rewards the lowest bids and their [00:05:00] one size fits all pick and projects out of DC approach that didn't include local governments and state governments. And so it poses a grave threat to sort of the proper outcomes that we want to see here.

Christopher Mitchell (05:13):
And just for people who are curious, Evan's dogs are more exercise than normal.

Evan Feinman (05:18):
Yes. I apologize

Christopher Mitchell (05:20):
For that out your window. I can see there's just masked men running around constantly. I think they're just terrorizing the dogs. So also if you disappear suddenly I'll have video of it.

Evan Feinman (05:28):
Oh, please do. No, it's actually [00:05:30] I believe my dog's diarist enemy and nemesis is approaching who is our mailman and day. He fights off an attempted invasion of our house and he's been successful today and I suspect he'll succeed again,

Christopher Mitchell (05:43):
A hundred percent success. So one of the things that we've seen is criticism, and I'll just say that I've heard this from Sascha Meinrath mostly and a few other people is around the state strategic plans. So I'm curious if you look back on this, [00:06:00] the states had to develop a five-year plan that then allowed them to, this was a plan to make sure everyone had the Internet access available and was able to take advantage of it, and then that led into their initial proposal, which has then led to their final proposal and this criticism that those five-year plans aren't really worth anything at this point. And I'm curious how you look back at it and think about that, which I think this was not your idea. I think this was spelled out in statute, but just let us know [00:06:30] how you got there and how you think in retrospect that went.

Evan Feinman (06:34):
I just be honest, I don't actually believe that the five-year action plans were a particularly useful exercise. To the extent they did anything worthwhile, it was for the states that had not run broadband programs previously. They required that those states then start thinking [00:07:00] about these complex questions and start to get smart on the issues. But there are better ways to do that, and I read every word and comma and piece of punctuation in state's initial proposals and final proposals because they mattered. If every state had written an amazing five-year action plan or if no state had written any five-year action plan, there'd be no difference in the way the BEAD program ran and there'd not be a single more or [00:07:30] fewer person connected. I think it was largely not a worthwhile enterprise.

Christopher Mitchell (07:35):
Now when you say that, I'm curious because let's go back to this period. That was one of the first things that the states had to do, I think, right?

Evan Feinman (07:43):
Yes. Well, it was something that states as soon as they got rolling they needed, they had a longer deadline to write their five-year action plans and they did set about doing it and there may be some value in having this big public facing document that allows people to see here's [00:08:00] the 500,000 foot look at the strategy that we as a state are going to undertake. But I will tell you, we reviewed them in a very cursory fashion because they were required, and so one must review things that are required, but again, it was gym class scoring. If you wrote something, we approved it. There's nothing actionable in them, right? A, B, C, D, E, F, G, and you're done, right? I mean there's just really very little value in those plans. That's not to say individual plans [00:08:30] don't contain good work and thinking, but I mean for the purposes of running the BEAD program, they were not useful or particularly paid attention to, I think by anybody.

Christopher Mitchell (08:39):
I definitely remember reading some that were much better than others and one of the things that I think is valued just giving people an early exercise to get going, if we put ourselves back into that place, the state broadband offices are being staffed up with people who probably had policy experience education or healthcare or things like that. And so it was a pretty decent [00:09:00] activity and then engaged some of the people in the community whose words were supposed to be reflected in the document. And so I think I might be a stronger definitive of it than you are in terms of doing some building capacity and that sort of a thing. But I am curious then when the ACP ran out and you all, because the ACP was figured prominently in every one of these plans in terms of how they were going to get this solved, this fundamentally big challenge of affordability, were you all watching this with horror [00:09:30] as that was happening then?

Evan Feinman (09:31):
We were pretty confident that the ACP was going to be renewed as were I think a lot of other people. It seemed to us that there was a deal to be made there and I think there were strategic errors on many sides of that negotiation in terms of how it went forward. That said, I wasn't in the room, so it's hard to criticize having not been a part of the perian thrust, I will say that it was everybody's belief that there was going to continue to be some [00:10:00] significant low income support program. We talked about both the ACP or successor programs. In a lot of ways people didn't fully reckon with what it meant to eliminate that program, particularly in the context of the requirement in the law that there'd be low cost service options provided by winners of BEAD Grant.

Christopher Mitchell (10:22):
I want to move on to the area that I feel I want to make sure we don't run off time on, which is the state restrictions and then we're going [00:10:30] to come back to permitting and probably a little bit of the mapping and data type questions. And I want to preface this just briefly by saying that Congress gave you a hard set of conflicting instructions. In some cases it feels like in retrospect, I wasn't paying attention when it actually happened, but the federal government created the 1996 Telecommunications Act leading up to it for years. The Congress was very clear they wanted to open up telecommunications to everybody [00:11:00] and then lo and behold, the Supreme Court finds that no, actually cities can be locked out of building their own networks by the state. And regardless of whether you think that's a good idea, there's a legal question about just whether or not that was going to happen.

(11:14):
And so we were excited to fast forward to BEAD. We were excited to see that you were told basically to make sure that states did not discriminate against cities. I forget the exact language, but it was pretty clear that states should [00:11:30] not be able to prevent cities or communities or tribal governments or others from seeking funds. At the same time, there was not the explicit grant of authority that this Supreme Court or district courts or other courts would interpret, I think as giving you the authority to change state laws. So I'm just hoping you can walk me back through how you looked at that and how you dealt it.

Evan Feinman (11:53):
Well, we split the baby, so what we said was any state [00:12:00] law that discriminates against a particular sort of ISP public, private or otherwise that was on the books prior to 2020 shall remain on the books. But if after this law has been passed, which is a very clear statement to the states, that this is a program that wants to award the best project regardless for the given purpose regardless of technology type or business model, [00:12:30] we said laws that are passed after the passage of the IIJA, which conflict with the IIJA'S rules will be a violation of those rules and adherence to those laws will preclude a state from moving forward. And we did have to multiple times say to one, two or another third state, if this current law which is being considered in your legislature passes into your code, you will not be permitted to move forward with the BEAD program. [00:13:00] And that was powerful in a lot of times it was also yet another job we had to do. We had every single one of our program officers was monitoring legislation in their state or states, which was complicated and hard, but another piece of the puzzle that we had to watch.

Christopher Mitchell (13:16):
Did you contemplate other approaches? I mean, was there a discussion about taking a hard line with states telling them, for instance, I feel like Ohio and Michigan, I don't remember exactly, you probably know better than me, but I feel like both [00:13:30] of them had, or at least one of them, but I'm forgetting, which had language that was basically like, look, we don't want to give money to the public sector. We just want these grants to go to private companies.

Evan Feinman (13:41):
It wasn't worth having a big fight with a state about why they should change a law that's already on their books. It also just sort of good federalist principles, right? Like the regulation of business entities within a given state is for that state to regulate.

(13:59):
It is not [00:14:00] properly for especially the Department of Commerce running an infrastructure program to tell a state how it ought to regulate various business entities that are state chartered corporations that exist within or doing business within that state. So we didn't seriously approach that. I'll also say for what it's worth, while I think there are a lot of shenanigans that are done on behalf of for-profit entities by their allies in government, I think ideological fellow travelers, mine on the left have [00:14:30] a rosier view of public and municipal broadband than really is warranted by its efficacy as a rural broadband solution. To the extent there are successful public broadband entities, by and large, they are in really valuable territory. They're in cities or dense counties and they're taking on the easier challenge. There are relatively few publicly operated [00:15:00] rural broadband networks that are operating sustainably and really managing to chip away at the problem.

(15:09):
An exception to that, to the extent they bridge the gap, I think is the rural electric and telephone cooperatives which aren't public in so far as people pool their funds via government and then operate the entity, but the same model. I mean it's [00:15:30] like everybody, they're a cooperative enterprise that is not for profit. And so I think to the extent people want to see those people powered networks out there pushing in rural areas, your best bets to find or work with your local cooperative rather than try to reinvent it as an authority or a part of your local or regional government.

Christopher Mitchell (15:52):
I appreciate you dangling the tiniest amount of red bait in front of me, and I want to say that I agree with you not [00:16:00] necessarily in exactly the way you would've phrased it. I would say that we have seen almost no evidence that absent state help one can create a rural cooperative Denovo to build a fiber network. It's very challenging, it's very expensive and the co-ops that have done it largely have decades of experience in another infrastructure that they were building. So I would just phrase it that way. The cities, this always gets into me what people define as cities, but absolutely the vast majority [00:16:30] of municipal networks that have been built are small towns or pockets of density. Sometimes they do go out and expand into rural areas, but it would be very difficult to build a brand new municipal system with just what's left over today in most areas. So for me, I always wanted to see the precedent, so I wanted to make sure that if a successful municipal network wanted to then expand into the rural areas around it, it would be allowed to. But I was not expecting, and I laughed at all of these people that were like, [00:17:00] be it is just going to build municipal networks. And I was like, no, it's not. It never was.

Evan Feinman (17:04):
And that was never really our belief about what we were going to do. Certainly to the extent there existed, like you said, a municipal network that then wanted to expand out from its stable base, they were eligible. Moreover, we weren't just requiring that counties and tribes and cities act on their own. What we said was local governments and tribal governments may select their own partner and by bring [00:17:30] that proposal into the mix to get score with anything else. And so even if a county government or a city government didn't have the ability to stand up their own entity, they could go pick their own chosen partner and then bring that partnership together with the BEAD program to see if there was a fit to see if it was going to be the high score, and folks were not allowed to discriminate against those entities. We did, for what it's worth, push in both directions.

(17:59):
So we [00:18:00] pushed back pretty hard on what was much more common, which was states saying trying to create new hurdles to public participation in the BEAD program. We did have one state that was pretty consistently trying to create hurdles for private participation because they already have a lot of publicly funded broadband in their state and we had to push them back too. And boy, they didn't like that they really, they were all in on their model, and I'll say that every time, there was no state that welcomed us [00:18:30] and said, we'd really like you to tell us how to do what we're doing internal at our state business. That was really never the way it went, and I respect that. I came from a background prior to this of having worked for multiple governors and we fairly jealously guarded our prerogatives and we always came with us. Just a matter of fact, we were not telling them what they should or shouldn't do. We were educating them as what would happen if they violated federal rules and then they were allowed to make a choice. [00:19:00] If this law is more important to you, then moving forward within the context of the BEAD program, and by all means go forward.

Christopher Mitchell (19:07):
One of the things that I heard from some of the state offices was that at times they felt like when you were pushing back that they were getting pushed back both in terms of big picture and small picture and they were more annoyed at small picture where, and I can assume, I can only imagine if I was the person I have word choice and verbiage that I like, and I had heard some folks [00:19:30] from some folks they felt like there were times when NTIA was too focused on small word choice and small things overall and then not as focused on the bigger picture. I was just curious how you react to that.

Evan Feinman (19:42):
That wasn't who we wanted to be. That's not to say that in the exchange thousands and thousands of pages between hundreds and hundreds of staff among the various states and folks on our side, we didn't get caught up on that. Also, for what it's worth, if I'm reading a document and I'm going to give you edits [00:20:00] anyway and I find a dramatical error, I'm going to go ahead and mark it. You may as well know it. And I don't know if that was always appreciated, but I certainly appreciate it if somebody says, Hey, I don't know if you edited this sentence a bunch of times or whatever, but you've got a tense or number disagreement between your first and third clause. I'm like, cool, thanks. I would prefer people believe that I speak and compose the English language properly. I do think there were times when to throw them under the bus a little bit.

(20:29):
Our [00:20:30] council's office was more exacting than made sense if one was merely focused on the big picture. There were these multiple layers of review and one of the final ones was pushing things through our office of the chief counsel and lawyers are going to lawyer, right? And so when they found something that they said they were concerned that perhaps something was ambiguously stated in a way that would lead to, not that it's definitely illegal or violative of federal grants rules for example, but it [00:21:00] could be interpreted to be that way. They didn't want us to sign off on agreement that then later we're going to get into a dispute where it's like, well, we want to do this thing and so you're not allowed to do this thing, blah, blah, blah. And so there were a fair number of back and forths, but I would say that the vast majority of the work that was done was around substance and it was around making sure that the program was going to succeed and also by the way, was on firm legal footing.

(21:25):
One of the things that you've got to be really conscious of when you run a grants program like [00:21:30] this is the fact that you're talking about billions of dollars and you're talking about people's business. And so if you're going to, for example, fund a build that is going to of a fiber network that's going to imperil the profitability of someone's unlicensed fixed wireless network, which we didn't think was a reliable option for a bunch of reasons, we can talk about that person is going to then that's their business, it's their livelihood. They're going to go try to find any way they can prevent that from happening. [00:22:00] And that will include yes, trying to figure out if somebody didn't dot an I or cross a T, and that might be sufficient for them to create enough legal headache that they can either slow down or prevent this from happening. And so you don't get a lot of credit for avoiding problems because of course it's a counterfactual. You can't prove that that's what happened. But there have been vanishingly few legal challenges to BEADs conduct, [00:22:30] and I think there's been a great deal of saber rattling, which is not uncommon, but I think the fact that there have been vanishingly few actual lawsuits is in part the result that we were really careful to make sure that nobody had anything to hook their suit onto.

Christopher Mitchell (22:44):
Doug Dawson and I have periodically every month, not every month, but I mean for months I've been saying, well, the lawsuits are starting now or they're going to start soon. Here they come. So I don't don't want to just brush that away, but you're right, we've been wrong. [00:23:00] Cox sued Rhode Island and we were all like, whoa, look at this. This is the beginning. And then they resolved it and no one else had a lawsuit that we were tracking. And so

(23:09):
Yeah, I mean that makes sense in terms of how you came about that with the exclusion, excluding the wireless systems, the unlicensed systems. I was very frustrated by that. And at the same time, I'll say that just before you and I were talking, I was listening to a podcast with the two FTC [00:23:30] commissioners who may or may not be on the commission right now depending on how the courts decide, the modern equivalent of Humphrey's executor as we all know now, and they were talking about the benefits of Bright line rules and I was like, yeah, I love bright line rules except for when I don't. I guess because you could have had some sort of deep, really complicated criteria that everyone would've been fighting over. Instead, you took the move and you said, no unlicensed wireless, we're nervous about it. Just tell me a little bit about [00:24:00] what went into that decision because I'm sure there's good unlicensed fixed wireless networks, but you chose to go down the path of a bright line rule.

Evan Feinman (24:08):
We permitted, of course, licensed fixed wireless and we permitted licensed by rule fixed wireless, and we permitted technologies that utilized both licensed and licensed by rule spectrum as well as unlicensed spectrum. So there's plenty of good technologies that make ease of all the white space available to them and getting better. We don't talk enough about [00:24:30] the quality of the work that terrestrial fixed wireless providers and terrestrial fixed wireless engineers have done. That's a technology that's come a really long way. 10 years ago, people hated it, and now you look at particularly the cutting edge developments there, companies like Tarana are generating and there's a really extraordinary amount of work going on there. The reason we ultimately came down on all fixed wireless is good reliable broadband except unlicensed is [00:25:00] specifically because of that lack of a license. If we're talking about making an investment that will last decades, which is what you want when you're investing billions in infrastructure, there's real risk that look, hey, you don't have the license to this portion of the broadcast spectrum. Somebody else could come in and get that license and then use it for not your network, and all of a sudden everybody who was getting serviced via that network is out of luck and there would be nothing to prevent it. [00:25:30] And so it created a dramatic risk that we'd invest a ton of money in a network that would then become defunct really quickly, and we just didn't want that.

Christopher Mitchell (25:42):
Right, and not only was there a risk that someone could get a license for some technology there, but also the rules are that no one can permit someone else from using that unlicensed spectrum and doing it in ways that was significantly undercut the ability to efficiently deliver services there.

Evan Feinman (25:59):
Right? Absolutely. Yeah. Somebody [00:26:00] might not go get a license, but they can just decide to do what these people are doing, which is just use it and

Christopher Mitchell (26:04):
Yeah, we've seen that and we've seen places where they can work it out in a reasonable manner and also where they haven't. Exactly.

(26:11):
So permitting has been a forever a major issue when you were designing the rules, I think you had a sense that permitting goes slow, particularly around public lands in the west and things like that. Now we're hearing horror stories about major agencies where [00:26:30] they've lost significant workforce. The people who took the buyouts may have been people who are close to retirement and had decades of experience doing this, and now we have fewer people with less experience responsible. I'm just curious, how did you originally think of it, and then what are you thinking about now in terms of the threats that this poses to this program for any kind of fiber project that moves forward?

Evan Feinman (26:57):
We had an approach to federal permitting that I think [00:27:00] was very strong, which was we went and gathered every single exemption or exclusion that had existed in any federal program, and we said all of these work for us, and we really worked hard to get states ready to deal with federal permitting. We hired a brilliant woman named Jill Springer who worked assiduously to continue to build out a really strong permitting approach. And it's worth noting people think about permitting in this context [00:27:30] primarily federally, that is going to be a big chunk of the permitting work that needs to happen, but state and local permitting as well as this whole suite of permitting adjacent activities, negotiating pole attachments, negotiating access to rights of way and easement, negotiating rail and road crossings, those are all things that will make it challenging to deploy these networks. One of the things that I'm really concerned about is in the context of the federal permitting, there's a reason that [00:28:00] prior federal efficiency efforts were big and slow and deliberative, and it's because if you say, oh, this permitting office is inefficient, let me just cut the staff by 50%, you've made it less efficient.

(28:15):
Having fewer people administering the same rules isn't an efficiency that you've gained. Instead, you've actually created more costs on the businesses that agency serves than you've created savings in terms of the [00:28:30] relatively small amount of salary that you're reducing the outlays for. And so I have tremendous concerns about that, but then I also have concerns about the really talented team that we put together of federal program officers out in the states we're supposed to be turning their attention after the approval of a final proposal to being sort of the chief permitting solutions person in their state, leveraging their connections, their ability to pull the governor's office, to [00:29:00] pull the private sector actor, to pull the local actor and the federal actors together and say, okay, you've got a fiber that you want to bore under an overpass and come up in a state forest and then cross a railroad.

(29:14):
Let's get this together. Let's make sure this baton is handed off in an orderly fashion, and then we project manage it from here to there. The really unpleasant working conditions that my former folks are experiencing now is going to drive a bunch of them out of NTIA and [00:29:30] NTIA is not going to be permitted to backfill their positions, and instead the private sector is going to still have the same ticking clock that they had before, but they're not going to have any support in solving these government problems. And I'm encouraged that the Trump administration is talking about making improvements to the way the National Environmental Protection Act and the National Historic Protection Act are implemented, but nothing that they've done so far gives me confidence that they have the policy rigor necessary [00:30:00] to do this in a way that is both effective and withstands legal challenge.

Christopher Mitchell (30:03):
One of the things I'm hearing about is issues around categorical exclusions, which I barely understand, but as I understand it has to do with the fact that we don't expect most of these investments to require significant environmental review. Most of them should be a matter of saying, oh, we know that this kind of a thing is not a big threat and it can move forward, but that we have more and more people who don't understand that history and that process.

Evan Feinman (30:28):
Yeah, I mean the biggest problem [00:30:30] is that if you'll let me sort of dive into Evans theory of governance, a lot of these rules exist for a reason, and there were significant identified problems that these bureaucracies and rule sets exist to solve. That said, I'm a big believer in particularly where we sit today in an enforcement mechanism as opposed to a procedural safeguard. [00:31:00] And what I mean by that is often we erect these mammoth procedural edifices that are designed to prevent significant environmental damage or damage to historical resources or damage to or misuse of human subjects or misuse of government procurement authority. The problem is these new procedural edifices impose significant new costs that didn't [00:31:30] exist before, and I would argue that they often create more costs than the harm that they were designed to hedge against. And I would much rather in the context particularly of categorical exclusions for environmental, the way it works right now is if you want to teach a digital literacy class under the digital equity program or the BEAD program in a community college classroom in a building that already exists, you still are going to need to hire a NEPA consultant at five figures, and that consultant is going to need because you need somebody who speaks the language [00:32:00] or it's going to get harder.

(32:01):
That consultant is going to need to spend couple of months, three months, four months engaged with the federal permitting folks to prove that you fall into that categorical exclusion so that there's no environmental impact to you teaching class in a building that's already there. That is nuts. The way it should work is there should be in big bold letters, you and your attorney need to review this. Then there should be a thing that says, this is what the categorical exclusion says. If you fall into this [00:32:30] categorical exclusion under penalty of something, bad sign here and you can move forward. But then later we're going to audit a bunch of these and if it turns out you lied, it's your but. Right? And we can figure out what that penalty would be, but that's a much better way to do this, and it's much more sensible and there really shouldn't be. I don't think we've prevented significantly more environmental harm than would've happened anyway as a result of the environmental [00:33:00] assessments and environmental impact statements and dramatic back and forth. We require around categorical exclusions than we were going to do before. One of the best examples of this kind of silly overgrowth that I have found is around the concept of human subjects research. That sounds ominous. We shouldn't do more Tuskegee experiments on the citizens. I mean, you can certainly see where this came from, but

Christopher Mitchell (33:24):
I think I know where you're going because I mean, whenever it comes up to our staff, we're like, oh, we've got these cool partners [00:33:30] with the university and we want to do this thing. And I'm like, are we going to have to deal with what's called the review board if we're going to do chemistry experiments on 'em? Absolutely. But we're going to ask them a few questions and we're going to treat them with respect.

Evan Feinman (33:43):
Yeah, it's gotten overgrown to the point where you've got to, and it takes months to get the review done, and anytime you're doing a survey of people or in our context, it was the challenge processes. I mean people just being allowed to indicate where their broadband service was and there is no human subjects research going on, [00:34:00] and it shouldn't work that way. I mean, another good example of this where it's like we've created this big safeguard to prevent a bad thing from happening in federal hiring, which is deeply, deeply, deeply broken, which we can talk about in another time. The reason all these problems exist in federal hiring is because people used to just hire their buddies,

Christopher Mitchell (34:15):
But

Evan Feinman (34:16):
I would've been much more happy to say, hold me personally liable if I want to hire this person, just let me hire, make me go through months and months and months. The longest one I had was 14 months [00:34:30] to bring somebody onto my team who was very qualified because I have to prove to non-experts that they're qualified. If it turns out I've been hiring a bunch of people who can't get the job done or who are related to me or something, then just prosecute me. But don't make everybody jump through these hoops to prevent a bad outcome, which by the way, you're not very good at preventing, right? Because plenty of unqualified people do wind up getting hired into some of these positions not on my team, and people hire their friends all the time. And so it's like you've got the worst [00:35:00] of both worlds in all of these instances where you've created dramatic new costs, you haven't successfully hedged against the bad outcome you're trying to hedge against, and you burn a lot of time as well. This is one of those things where I really wish the DOGE team had one been multidisciplinary. I think there are a lot of things that smart computer programmers are good at. I don't think it's all the things DOGE is doing, but then two, that they'd, if they'd really spent some time to learn from people what the big [00:35:30] problems are in the federal government, I've got a list a mile long having spent three years in there of things that the federal government does that are bananas.

Christopher Mitchell (35:38):
Yes, I always recommend the book with the woman who created the Digital Service Recoding America by Jennifer Polka. I probably suggested it 10 times. She goes through a lot of that. I'm in full agreement with you. I think this is one of the great challenges that we now face, particularly when so many people have been convinced that government cannot work, that they're excited [00:36:00] to burn it down blindly.

Evan Feinman (36:01):
Yeah, I look forward to working with, I think there's a version, and they've walked back some of their criticism of BEAD, which I appreciate, but there's a version of Make government work. Well, that pairs up well with kind of the abundance Klein Thompson approach you're seeing, but that's not new in our political discourse. I mean, we had tons of different presidents and governors across the country talk about running government [00:36:30] well and efficiently and who've made great strides in that regard. I do think there's a lot of room for more improvement there, and if you believe that government can solve problems, you should be really in favor of government doing that efficiently.

Christopher Mitchell (36:43):
By the same token, I feel like we think that our businesses knowing how communities have failed to build networks, because that's really important if you want communities to build strong networks, and I think people who want government to work need to understand how government's failing. But I want to come back and I want to talk about the challenge process [00:37:00] and about how you think about it in retrospect. We've already talked about how the Federal Communications Commission, I think wasn't ready to be a good partner. I've heard more horror stories than I think you could admit if you want to ever work with some of these people again, in terms of some of the frustrations. So we're not going to dig too deeply into all the mapping stuff, but the challenge process is one that I don't know, I came into it, lemme tell you how I came through it and I want to see how you react, which is that I was like, we need to work with people and nonprofits and local [00:37:30] governments to make sure they are ready to engage with this and fix it. And then as I watched it and I saw the rules coming out, I said, we need to encourage people not to waste their time because this is not going to go well. And it is going to be so frustrating that I was afraid people would lose faith in us if we told them to engage in that process. Not in every state, but in some of the states we were focused on.

Evan Feinman (37:52):
Well, so the challenge process is hard, but I think we struck a pretty good balance of permitting, for example, [00:38:00] area challenges. If somebody says, we're not actually in this community, unlike the fcc, we didn't make you go do a speed test or get an affirmation from every single location that was reported to be served. We let you do a small sample, I don't remember exactly. I think it was like six per census block,

Christopher Mitchell (38:16):
Which pretty, yeah, seven might, but

Evan Feinman (38:17):
Yeah, right

Christopher Mitchell (38:17):
In that

Evan Feinman (38:18):
Neighborhood, bad, right? A community that cared about it or a nonprofit that cared about it or an ISP that cared about it could go find six people out of thousands of homes to go test that connection. [00:38:30] And a lot of those tests, if you're doing an availability challenge, you didn't even need to go to a speed test. You literally just get on the company website.

Christopher Mitchell (38:35):
I think it literally does sound reasonable until we tried to do it. I think your question then is, okay, what is a reasonable number? And I think ultimately it's just not reasonable for us to expect random people to correct the failure of a different agency to create a good map. Now, you were handed a bad map, and so you had to try and figure out how to make it better. My [00:39:00] conclusion ultimately was that we just need experts to create a good map rather than trying to convince people to use their own time to engage with this process, to try to correct errors. Well,

Evan Feinman (39:11):
There were ways to get around it. So I think the speed test is the hardest thing to do because you've got to do it properly, which is hard to educate the public on, but availability challenges are easier. You get on a company's what you get on an ISPs website and you just plug in the address. Can you go to 1 2 3 Elm Street? No. Can you go to 1 2 5 Elm Street? No. Can you go to 1 2 7 ELM Street? No. Can you go [00:39:30] to 1 2 9 down the street? No, and so on, and you've built your area challenges and you're good. Now, what we saw pretty fast was a bunch of ISPs took those tools down off their website, which I thought was like direct malicious activity on their part, and a few of them, we got Secretary Armando to place calls their CEOs, and lo and behold, those tools popped right back up and we were able to get 'em on there.

(39:50):
Others weren't as responsive to that. We also were able to, a set of states made use of crowdsource data, things like ULA data [00:40:00] and the team at ULA was really remarkable in making that happen, and Brian and the rest of his team are really strong there, and that was another way to substitute it in. At the end of the day, all sampling methods are proxies for literally counting every single one, which is itself a sampling method. And so you're only ever going to talk when you're talking about millions and millions and millions of data points, you want to drive your error as low as possible. It's impossible to get a perfect [00:40:30] map, and the maps we're building to now are pretty darn big, certainly the best we've ever had.

Christopher Mitchell (40:35):
I would want to dig into that more with my colleagues who dig into this a little bit more. I just feel like we're still seeing a lot of frustrations and I think there are areas in which I think we would all agree, yes, these areas are doing really well. There's other areas. I mean, I'll say that on tribal lands, it's not at all clear that we have a very workable solution there, but I appreciate that background. I mean, I do feel like [00:41:00] this is one of those things which I'm just so frustrated that I just feel like it's the federal communications commission's job to have an accurate map. You got stuck trying to figure out what's the best you can do without that. And I feel like as long as there is an incentive for people to put in incorrect data and not be punished for it, it's a hard problem to figure out how to correct If we're trying to correct 10,000 errors, it's different than trying to correct 7 million errors. And so some of the states, I think with the [00:41:30] rules you put up, some of 'em worked well, others did not

Evan Feinman (41:33):
Well, and that was for those states to design. I mean, that's the other thing to remember is that people get the state government they elect, and there are some states that design challenge processes that I didn't think were particularly great, but that was a choice that those states made.

Christopher Mitchell (41:46):
So last thing I wanted to bug you on I think is LEO again and

(41:51):
From a different direction, which is that you, and I think we covered this before, and sometimes critics, or [00:42:00] not even, critics isn't even the right word. People who are trying to figure out how to correctly forecast the future are accused of wanting Starlink to fail or wanting low earth orbit satellites to fail. That is not the case. I think these are important. We think we talked about this before. They're really important. It's an important technology. At the same time, I feel like from what I've learned of financial games, we don't actually know right now if it is in the longterm economical to launch these satellites into orbit [00:42:30] and maintain them and to be able to provide a decent connection. There is a sense that if Starship works, that there will be a price of launching these satellites into orbit, and that will be low enough that it'll all work out and there'll be enough subscribers. And this is not the sort of thing where you want to pin the future on it, as you've said before. And so I'm curious, how were you all thinking about LEO in terms of the worst case scenario, which is that in five years, we might all just be laughing about how we [00:43:00] thought that that was going to be a sustainable model because it may turn out that the launch vehicles aren't as good as we thought or any number of other problems that could crop up.

Evan Feinman (43:07):
Well, Sterling's up there. Now you don't know that it's financially sustainable, but you do know that it is technologically feasible to launch a low earth constellation and deliver pretty good service to a bunch of people in a bunch of different places. And so given that the other options were worse either on risk or performance, what we were willing to do was [00:43:30] say, look, exhaust every other reasonable option, right? Figure out if there is a terrestrial reliable broadband solution, whether it's fiber fixed wireless, blah, blah, blah, hybrid fiber coax, what have you. But if that's not an option, let's use Starlink or LEO Networks as a connection option of last resort. And for those most remote locations, what we know is that they'll work pretty well and that by pairing down the number of [00:44:00] locations that we're going to add to those constellations, were not overly burdening them.

(44:04):
And it was reasonable to believe that at least in the short to medium term, that those connections would continue to be available for folks and continue to operate at a pretty good speed. But that was always a calculated risk. It is very costly to operate these networks. There's a reason that the only two customer facing networks that sort of exist are Starlink and KY net, which is a little bit in the air and plans to [00:44:30] get in the air over the next of years. Those two LEO networks also, by the way, have sister companies that are rocket companies, right? Because even with that vertical integration, you've got to launch thousands and thousands of satellites a year forever because the orbits degrade and the satellites orbit and burn up in the atmosphere, and you got to keep launching literally thousands of them every single year to maintain coverage. And I think there is significant risk that that business model is not going to be viable longterm [00:45:00] or that prices will need to escalate really dramatically. And both of those are bad outcomes for low income rural communities. And it's yet another reason not to put too many chips on that be

Christopher Mitchell (45:10):
Right to make sure that we use it only where it's absolutely necessary, and we're not just arbitrarily deciding to put an extra a hundred thousand people on it or 500,000 depending on where the numbers come down

Evan Feinman (45:22):
Or 3 million, we truly don't know what they're going to do. That's not to say that they could actually take it, but there's real [00:45:30] risk that given the runway that we've got here without calling anybody out, there were bad actors in RDOF that made promises that they were never going to come through on knowing that they were never going to come through on them because they didn't think they were going to face significant penalty. The same thing could be true here. LEO network actors could say, oh, yeah, yeah, in four or five years we're going to have plenty of capacity to take on all these customers. They could take the money now, they could prevent other technologies from getting there. And then four or five years from now, they [00:46:00] could fail to deliver. They could say, we're sorry, maybe they pay it back, maybe they don't. Maybe they go to court, blah, blah, blah.

Christopher Mitchell (46:05):
They could honestly believe it's going to work out. I mean, there's all kinds of technologies where people read their own press clippings,

Evan Feinman (46:10):
Right? It's just a lot to gamble with. And so what we wanted to do is minimize risk and make sure that we were going to get a good outcome for as many Americans as possible. And the way to do that is the way we designed it to do it, which is roll down your cascade of reliability so that you make sure you're getting the best connection for folks that you can. That by the way, is [00:46:30] going to cost them less out of pocket. It's going to last longer, is cheaper to operate and delivers higher speeds. Pushing more to more risky high cost technology is just not a good policy choice

Christopher Mitchell (46:41):
About the rockets. I'm curious, so you didn't come in a rocket expert and you have people on your staff who, right? So I'm curious. I am just going to go out there and read some blogs. I'm going to do some research and stuff like that. When you're in charge of something at this level, how do you become to a point where [00:47:00] you feel confident about a decision that's that important? I mean, are you able to call NASA? How did you guys staff up on this?

Evan Feinman (47:07):
You read and you talk to experts. When you're in the federal government, you have a lot of experts that exist within government and people pick up the phone when you call in the private sector. And so we talked to a ton of people to put together these, including by the way, the engineering and business folks from the LEO networks, both the two that are offering [00:47:30] customer service as well as other LEO and other satellite networks. There is not a dearth of expertise in these areas. It doesn't exist between my ears. And so you learn a lot and you make the best call on the basis of disinterested expert parties. And that's what we did.

Christopher Mitchell (47:48):
We're about out of time. There's still other questions, and I think I'm hoping that when we get another big piece of news, we're going to have you on connect this to have some fun with the other guests. [00:48:00] But I was checking in with some folks today to be like, Hey, what should I ask Evan about? And we're just reflecting on different things. And all those conversations led back to the thing. And one of the people I talked to was just like, but it was working. All of these criticisms and all of these challenges, and Jake Varn over at Pew States, they're tracking all this and all these states were coming in on budget with mostly fiber building these networks out. We're there. And so we [00:48:30] don't know when this airs. I don't know exactly what's going to happen, but I think we are hoping that the Trump administration will take their victory lap and go forward with it. But I want to say after all of this, it was working.

Evan Feinman (48:41):
Yes, I agree. And the timeline, it's frustrating that the timeline is where it is because I think if we could have pushed everything back, which is say back in time six months, and we'd had 20 states that were done, there wouldn't have been a way for the new administration to do other than say, yeah, [00:49:00] we're going to let this thing keep rolling. Because of the unique moment in time that we were in at the moment of the transition, they were still able to put a halt to everything. But as I said, I think with you and another context saying it took you too long to get to the finish line is not a good reason to then say, now that you're here, we're not going to let you cross it. And the primary criticism of the BEAD program that was at least accurate was it's taken you a long time.

(49:27):
You haven't built anything yet. There were other criticisms that I think were [00:49:30] just sort of made up or unfair. That is a true criticism. I don't think we talked about it before. I don't think it holds a lot of water, but whatever. But the fact of the matter is there'd be fiber on poles and in the ground right now in not just the three states that were approved. They could have approved more states. The system that we built them, we, by the way, we approved those final proposals. People got mad at us for taking a long time to approve initial proposals. It took us an average of 11 days to approve each of those three final proposals. We could have approved another 15 [00:50:00] since January and be approving more of 'em over the course of the summer. And there could be construction trucks rolling. There could have been construction trucks rolling for months.

(50:09):
And it is really disappointing that there doesn't seem to be an ability on the part of this administration to just say, yeah, we bought too much of the criticism of this program that came out in the politically charged campaign environment, and it turns out that there were some things they were doing [00:50:30] that we didn't like, so we took those out, but otherwise, we're just going to let this program move forward and take the win. And they wouldn't even need to say any of that. If they would just do it. If they would just waive the quote woke stuff and then just let the program proceed, it would be a huge win for them. That's the thing that I find so stunning is they're being self-defeating, even as they're defeating this tremendous progress for rural America. As I said before, I'm still hopeful that we'll get a good outcome. I'm just not particularly optimistic.

Christopher Mitchell (51:00):
[00:51:00] Well, time will tell, although more time than we think generally is what it seems like happening. Thank you, Evan, for spending some more time with us.

Evan Feinman (51:10):
Thank you, Chris. Really appreciate the opportunity to be on.

Ry Marcattilio (51:12):
We have transcripts for this and other podcasts available @communitynets.org/broadbandbits. Email us at [email protected] with your ideas for the show. Follow Chris on BlueSky. His handle is @SportShotChris. Follow communitynets.org stories [00:51:30] on Blue Sky, the handles @CommunityNets. Subscribe to this and other podcasts from ILSR, including Building Local Power, Local Energy Rules, and the Composting 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. [00:52:00] Thank you to Arnie Sby for this on Warm Duck Shuffle, licensed through Creative Commons.