Search This Blog

Friday, May 31, 2019


SECA (the State E-rate Coordinators’ Alliance) has submitted a list of proposed changes to the Eligible Services List (ESL) to the FCC.  Let's see what they're proposing:

  1. Make content filtering eligible
  2. Make more data security functions (anti-virus, anti-intrustion) eligible
  3. Allow applicants to choose multiple vendors to provide the same service to the same building
  4. Cabling should be eligible, regardless of what's plugged into it
  5. Make break/fix and time/materials maintenance contracts ineligible
  6. Fund multi-year support contracts in the funding year they're paid, not pro-rated across the length of the contract.
  7. Eliminate sub-categories (I think they mean getting rid of the Basic Maintenance category)
  8. Continue MIBS ("Managed Internet Broadband Service"; paying a company to provide equipment, installation, management, repair of an applicant network)
  9. Two suggestions
    • Bring back the Glossary of Terms
    • Include a CIPA (Children's Internet Protection Act) compliance reminder
  10. Allow Wi-Fi on buses (and other off-campus Internet solutions)
To which I say:
  1. I've been saying this since 2007.  It made it into an NPRM in 2008, but withered on the vine.
  2. See #1.  The most important reason: it's increasingly difficult to find firewalls that aren't called "total security appliances" and don't include all this stuff.  Let's not burn applicants who don't have the tech savvy to match manufacturer jargon with USAC jargon to figure out which features are eligible.
  3. Yes!  Buying connections from 2 different vendors is a cost-effective means to increase network availability.  "Redundant" is not a dirty word in IT; it's a virtue.  The GSA (the US gov't organization responsible for purchasing) has made it a requirement of Internet contracts since 2008 (if not earlier).  The GSA requires that even government agency backup sites for disaster recovery have "redundant access to the GSA data network."  Check out page 24 of the GSA's ordering guide for data services: it explains what to do if "[t]he agency wants to use one solicitation to select multiple contractors for the same requirements: such as for a primary and backup, or for network redundancy/diversity."
  4. I was going to get my hackles up a little, but the $150-in-5 Rule makes me just say "fine."  I don't feel good about security cameras and phones being funded if they're riding on their own network (with their own switches, separate from the data network used by students).  But forcing applicants to cost-allocate their cable plant just seems like a huge waste of time since C2 budgets don't allow applicants to get all the eligible equipment they need, especially since most of those devices are just going to be plugged into the same switch as the WAPs.
  5. Huh?  I'd be OK with dumping maintenance entirely, but why single out break/fix contracts?  I've always thought those contracts fit in better with the FCC's position of "we don't want to pay for warranties, just for services delivered."  And they definitely fit better with the E-Rate calendar, which forces applicants to start the year without maintenance funding approved.
  6. Yes.  Just another area where enforcement saves no money, and creates headaches.
  7. Agreed.  The only reason to have a separate Basic Maintenance was the 2-in-5 Rule.  Let's toss that distinction.
  8. Again, with C2 budgets, go ahead and make it eligible.  My only complaint is that I can get E-Rate funding for someone else to manage my network, but I can't get funding for the products I need to manage my network.  Let's make network management software and appliances eligible.
  9. Yes.  Shortening the ESL didn't simplify the rules, it just hid the rules.
    • The Glossary of Terms was important, because when I read the ESL, I hear Inigo Montoya in my head: "You keep using that word. I do not think it means what you think it means."
    • Let's take it a step further and tell applicants whether CIPA requires: 1) applicants to filter devices which are not applicant-owned if they are using the applicant network, and 2) applicants to filter applicant-owned devices if they are not on the applicant network.
  10. I have heard of some great uses of Wi-Fi on buses.  Back when we were hitting the funding cap for the program, I didn't think it was a good use of funding, but we're awash in funding, so I'm OK with it.
So I agree with 9 out of 10.  As usual, SECA delivers wise suggestions.

And they got me thinking.  Several times they mentioned how the 470 drop-down menus should work, and it occurred to me: the ESL should just be an explanation of how to use the drop-downs on the 470 and 471.  Organize the whole thing that way.  It would serve 2 purpose:
  1. It would increase the chances that 470s and 471s were done right, which is kind of the point of the ESL.
  2. It might make the FCC realize how freakin' hard it is to make reality fit into the drop-downs.

Stop the madness

Now it's official.  The FCC is seeking comment on a Petition for Rulemaking from a trio of small phone companies who want to "discourage overbuilding of existing federally supported fiber networks."

I've already stated my objections to this harebrained attempt to preserve monopolies.  So I'll just say this: get your comments in.  Because I am sure that the companies who have received federal funding to install fiber are going to be vigorous in their support of this proposal to protect their monopolies.  And the fact that it took only 8 days for the FCC to put out this notice has me worried.; it feels like the FCC is rushing to get this out there.  Perhaps because Commissioner O'Rielly is obsessed with overbuilding.

Comments are due by  July 1, reply comments by July 16. 

Filing is not hard.

Just go to the Express Comment form, put in Proceeding RM-11841 (as you type in the text box, you'll see a list of proceedings; after you've typed "RM-1184", you can click on "RM-11841" on the list).  Then put in your name and address, and under "Brief Comments," type in what you think of the idea. 

It seems the FCC likes to hear how the rule would affect specific organizations, so if you can show how this rule would harm your school or library, that's very powerful.  In general, facts are powerful.

If you want to write your comments offline and then upload them, you can use this form.

Comments matter.

Friday, May 24, 2019

Overbuilding again?

Well, now it's moved past one Commissioner's complaints.  A trio of rural phone companies have filed a request for rule changes with the FCC, seeking to preventing overbuilding of existing fiber.

First, I'll say that they do have a point.  But their suggested rule changes are terrible.

Where do we agree?  "Because the regions include hundreds of schools and cover thousands of square miles, only select, large service providers have been able to respond to the RFPs. Smaller providers that are already serving individual schools within the region, via their USF-supported fiber networks, were unable to respond to the RFPs due to the sheer size of the requested WANs. "  Yup, that's a problem, and not unique to fiber; I have provided examples of statewide purchases excluding bidders.

I don't have a great solution to the "too big to get bids" problem.  I think the best solution is to get rid of consortia and use purchasing cooperatives; applicants combine their purchasing power, but contract individually with the selected service provider (which of course means filing their own Form 471).  So in this case, each district in the TX consortium would have to compare the consortium offering to other bids, including bids that only serve that applicant.  If there are reasons that a district would prefer to be connected to the consortium network, they can include that as an evaluation criterion.  But price has to be primary.

The problem I can't solve is that such a system is less attractive for bidders in two ways: 1) the most lucrative clients are more likely to find a better deal and not purchase through the cooperative, leaving the winning bidder to cope with the lower-quality clients; 2) if a client or two in the middle of the area drops out, it could blow a hole in the network design.

But that solution is much better than the change proposed by the trio of phone companies: "Category One services shall not include special construction costs for the construction of fiber where it has been demonstrated that fiber already exists, unless the existing fiber owner is unwilling to negotiate in good faith to lease that fiber at reasonable market-based prices."

How will USAC determine if fiber exists?  Thusly:
  1. Anyone applying for Special Construction will have to have a "FCC Form 471 Special Construction Exhibit" which includes a list of locations and a map of the fiber route.  The Exhibit is posted publicly for 60 days
  2. During that 60 days, anyone with fiber in the area "shall submit information to the Administrator to show that fiber already exists in the applicable locations."
  3. If USAC decides there is existing fiber, the service provider on the 471 will have 120 days "to negotiate in good faith the terms, conditions and reasonable, market based price of a fiber lease agreement."
Reasons I dislike this solution, off the top of my head:
  1. More forms and more review?  The application process and review process for self-provisioned fiber is already incredible onerous.
  2. The process would not conclude until 180 days after the Form 471.  So the FCDL is delayed over 180 days (60 days + time for USAC to decide if fiber exists + 120 days to negotiate).  It's already a rush to get a big network lit by June 30, and this process would remove almost half the time available.
  3. " fiber already exists in the applicable locations" is too vague.  If the map includes one pole that has someone else's fiber, are we going to put the application on hold for 180 days?  What if someone has fiber connecting some of a district's locations, but not others?  What if there is fiber running with a half mile of the locations?  Within a mile? Ten miles? 
  4. Negotiate the "market based price"?  These 3 telecoms are trying to ensure that there is no market, by maintaining their monopoly on fiber in the area.  This whole situation arose because a new service provider can lay new cable more cheaply than the cost of leasing existing fiber.  Even when the existing fiber was installed with USF subsidies (from the High Cost program).  Because the existing fiber is at monopoly prices.
  5. The new service provider won't know it's costs when it submits a bid, since the cost will be dependent on the results of negotiations, half a year after the 471 is submitted.
  6. You think this won't end up in court?  Any time I hear "good faith," I think, "there's a lawsuit waiting to happen."
  7. This solution assumes that only large projects will be affected.  What if the Baby Bell has fiber coming into a building, but the cable company, which has fiber on a nearby pole, will have to pay Special Construction to bring the fiber into the building?  Does the cable company have to negotiate for the existing fiber?
  8. More fiber is better.  The E-Rate is not in the business of running fiber out to remote areas; that's the job of the High Cost program.  So the E-Rate should only pay for overbuilding in cases where it is cheaper.  That is currently the rule.  These new rules seek to increase the cost to the E-Rate program in order to protect a monopoly.
So I hope the FCC will not make the changes suggested.  Instead, maybe they should just stop pushing consortia.

Friday, May 17, 2019

Subsidizing monopolies

All the FCC Commissioners testified before Congress yesterday.  No, I didn't listen to it.  But I am scanning their statements to see if there's anything important.

First, of course, the most important question: Did they capitalize the "R" in "E-Rate"?  Well, it turns out only Commissioner O'Rielly used the word.  He did capitalize!

So what did they say about the E-Rate?  Not much obviously.  Is that a bad thing?  Part of me just wants Congress to forget all about our little program.

Chairman Pai:  While the Chairman did talk about the Universal Service Fund, and said the digital divide was a top priority, he focused on on the Connect America Fund and mentioned the Rural Health Care Program.  The only statement I found that concerns the E-Rate:  "...we developed a reorganization plan to create a Fraud Division within the Enforcement combat USF fraud...."  Good.  Stay focused on fraud.  Because most improper payments are the result on unwitting applicants getting snared by the huge, secret, nebulous and ever-shifting rules of this program, and don't benefit anyone personally.  Better to focus on people intentionally taking improper payments for self-enrichment.

Commissioner O'Rielly: Anybody want to guess?  Yup, overbuilding.  This time, he takes his complaint further: "providers serving hard to reach areas can face serious financial difficulties if a new government-subsidized provider 'competes' to serve existing customers—or worse—takes only the most highly profitable customers.... It recently came to my attention that new E-Rate-subsidized fiber networks were overbuilding local USF-funded Texas broadband providers and stealing their core anchor customers. By manipulating the contracting process to favor the bids of particular providers or self-provisioned service, some local school districts have been actively undermining local USF-supported providers’ existing investments, and as a result, making it even more difficult to serve surrounding communities where some households may lack any Internet access at all."  Wow.  Let's unpack that.

" a new government-subsidized provider": Yes, the new provider benefits from the E-Rate subsidy, but the existing provider benefits from the E-Rate subsidy and the CAF subsidy.  So it's a government-subsidized provider winning a bid over a doubly-government-subsidized provider.

" 'competes' to serve existing customers": Yup, competes.  The Form 470 process does increase prices and reduce flexibility for applicants, but it can create competition.  And clearly this whining is coming from someone who lost a competition.

"takes only the most highly profitable customers": I think of this as Ma Bell's Lament.  My dad worked for Ma Bell, and when they lost their monopoly, he said that since competitors would skim the profitable business (long distance and business local), Ma Bell would have to raise rates on residential local service.  He wasn't wrong, but....  Sucks when your monopoly gets taken away.

Wait a minute, though.  Take a step back.  "most highly profitable"?!  Shouldn't LCP (Lowest Corresponding Price) guarantee that school districts are the least profitable customers?  I know LCP isn't enforced, but should a Commissioner be admitting that to Congress?  And LCP aside, I object to the idea that the cost of bringing service to remote areas should be funded by overcharging school districts.  Why are you trying to make the E-Rate fund that service?  That's the job of the High Cost Fund.  And why aren't Connect America Fund projects required to serve surrounding communities?

"manipulating the contracting process to favor the bids of particular providers": If the Commissioner has evidence that an applicant's procurement has not been fair and open, he should Code 9 that applicant. [Whistleblower calls used to be called "Code 9" calls, because if you called in and said "Code 9," you could be anonymous.  Alas, the Client Service Bureau has no sense of tradition, so if you want to report a program violation today, you press 3.]  Except, you know what?  Self-provisioned fiber FRNs are essentially pre-Code-9ed; the procurement process is scrutinized more closely than it would be under a typical Selective Review or Audit.  If, on the other hand, the Commissioner only has a complaint by a telecom company with no evidence of malfeasance, he should not say such things to Congress.

"favor the bids of ... self-provisioned service": Has the Commissioner seen what applicants have to go through to get approval for self-provisioned service?  The system is set up to favor lit fiber over leased dark fiber, which is favored over self-provisioned.  And why would an applicant want to favor self-provisioned?  E-Rate unpleasantness aside, self-provisioning projects are a tremendous pain in the filament.  The only reason to go through the hassle of laying your own fiber is to save money.

"some local school districts have been actively undermining local USF-supported providers’ existing investments":  Does the Commissioner have evidence that districts are trying to undermine someone's investment?  What possible motivation could they have for actively undermining a telecom company?  I'll grant that the provider with existing fiber may have been counting on charging a lot of money to the school district, but the district is not seeking out lower prices because it wants to undermine anyone.

"making it even more difficult to serve surrounding communities where some households may lack any Internet access at all": So taking away one customer makes it more difficult to serve other customers?  Hmm.

Perhaps the Commissioner is right and an unscrupulous school district (or, based on the questions in his letter to USAC on the subject, more likely a consortium of districts) has colluded with a service provider to fix a procurement and wastefully install more fiber on poles where fiber already exists.  If that's true, the Commissioner should give it to the local Attorney General, not complain to Congress about it.

Or perhaps a service provider got CAF funding to lay some fiber in a remote area, with a business plan based on monopoly pricing, and another company came in and undercut them.  Speaking as a small business owner, I completely agree that competition sucks, and it makes it much harder for me to bring E-Rate to all applicants.  Speaking as a taxpayer, though, I'm glad that no one has a monopoly on E-Rate consulting.  (OK, if I had a monopoly, I would be a beneficent overlord and the program would be much improved, but that's an exceptional case.)

Or perhaps the applicant consortium was larger than the footprint of the already-subsidized existing carrier's monopoly, which again messed with the business plan.  I have certainly pointed out in the past that large consortia can limit competition., and suggested that applicants "take a close look at whether that consortium is really going to save you money."

Whatever the case, it's not an issue for Congress.

And it doesn't strengthen the Commissioner's "humble request ... that Congress consider the FCC’s Universal Service Fund (USF) as a primary means to distribute new funding" in order to avoid "wasteful and duplicative spending and adverse consequences...."  I mean, even if his example were valid, he's basically saying: "We need to keep all the funding in the USF so that we don't see the kind of duplicative and wasteful spending that we're currently seeing in the USF."

But the Commissioner did get the most important thing right: the big "R" in "E-Rate."

Thursday, May 16, 2019

再见, 华为

What does President Trump's Executive Order on Securing the Information and Communications Technology and Services Supply Chain have to do with the E-Rate, you might ask?  It will probably affect which vendors you can buy equipment from.

Not that the FCC needed an Executive Order to get on this.  Back in April, the Commission released a Notice of Proposed Rulemaking "to prohibit, going forward, the use of USF funds to purchase equipment or services from any communications equipment or service providers identified as posing a national security risk...."  The NPRM quickly focused on two vendors: Huawei and ZTE.  The FCC created Docket 18-89 for comments, and they keep coming in.

Meanwhile, the 2019 National Defense Authorization Act (skip to section 889, page 282) declared that no executive agency should purchase from Huawei Technologies Company, ZTE Corporation, Hytera Communications Corporation, Hangzhou Hikvision Digital Technology Company, or Dahua Technology Company.  And the Commission sought comment on whether that applied to purchases through the USF.

So one of these days, the FCC is likely to prohibit the use of equipment from those countries.  Does that matter?  Well, I couldn't find any instances of E-Rate applicants making any C2 purchases directly from Huawei  (SPIN 143036885) or ZTE (SPIN 143044152).  But some of the comments in the docket are from telecom companies using equipment from those vendors who are going to be facing some serious replacement costs.  So perhaps the cost of some C1 services will be creeping up.

Meanwhile, the tariffs on Chinese imports have already affected C2 pricing.  The September 10% tariffs increased prices from many C2 manufacturers, but I haven't read anything about the effect of the new jump to 25%, but it's got to have some effect.

Tuesday, May 14, 2019

Someone's head restin' on my knee

I happened across some suggested changes to the program over at Funds for Learning, and I can't resist running my mouth.  The suggestions:

  1. Bring back voice
  2. Make all network infrastructure eligible (security, monitoring and management are mentioned)
  3. C2 budgets should be per-entity, not per-entity.  By which I mean, per-organization, not per-location.
  4. C2 budgets should be doubled.
  5. The C2 budget floor should be tripled.
To which I say:
  1. Let's expand C2 first, and see if we've got money left.  Back when the program ran out of money every year, I think I was the first to suggest throwing voice out of the program.  Now that we don't hit the cap, I'm OK with letting voice back in.  But first, let's fully fund C2.
  2. I agree that the per-student cap reduces the need to nit-pick nework equipment.  And all sorts of security equipment should be eligible in any case.  Monitoring and management are a little more questionable to me, but if we're going to pay for MIBS (the "M" is for "managed"), then it seems like management should be eligible, whether it's a service or a piece of equipment.  But what FFL said was, "There should be zero ineligible network infrastructure."  That's a bit too far: let's not let servers back in, or network storage (I'm looking at you, video servers).  But yes, let's allow security, management and monitoring.  It's irresponsible to run a large network without them.
  3. Yes, yes, yes!  This should be the FCC's top E-Rate reform priority.  It's a simple change, and would dramatically simplify the C2 process.  And it would give school districts the power to decide where they need equipment, rather than having to live by the fiction that all locations have equal C2 needs.
  4. Yes!  My back-of-the-napkin calculations say we should make the C2 cap at least $500/student, and the program could afford $600/student, but doubling would be a step in the right direction.
  5. Yes.  Small applicants are leaving this program in droves, so the FCC should do whatever it can to bring them back.
What is the FCC doing about C2, anyway?  When the Wireline Competition Bureau released their Report on Category 2 Funding, I wondered if the next step would be an NPRM.  I mean, it seems like they'd have to do one, no?  They did request comment on C2 budgets back in 2017, but that doesn't count as an NPRM, does it?

If no Order is issued, applicants who applied for C2 in FY 2015-2016, as well as applicants who have never applied for C2, would find themselves back under the 2-in-5 Rule for FY 2020-2021.  (Applicants who first filed for C2 in FY 2016-2017 would go back to 2-in-5 in FY 2021-2022, and so on; as their 5-year budgets expired, applicants would go back to 2-in-5.)

If the FCC is going to release an Order, I'd think it would be before the Eligible Service List is prepared.  At this point, I don't think that leaves enough time for an NPRM.  An interim Order would create more chaos, as applicants try to figure out if they should lock into a new 5-year cycle, or see if something better is coming.

Here's a question that I would not have considered back in 2014: would a return to 2-in-5 be a bad thing?  I mean, the worst part of the 2-in-5 Rule was that most applicants never got a sniff of C2 funding.  With the vastly increased cap and everything but broadband tossed out of C1, we wouldn't run out of money unless applicants started requesting more than $600/student.  That might be enough.  But there are other problems with 2-in-5.  So if we're not going to run out of money, how about we go to 5-in-5?  Request whatever you need whenever you need it.

You can say it's stupid, but in the WCB's C2 report, they said that $150/student was enough. (Well, they said the "budget approach appears to be sufficient for most schools and libraries," but then they went on to talk about how few schools had spent all their money, so they were basically saying that $150/student was enough.  So if applicants don't need more than $150/student, then why not just get rid of the restriction, and let them spend what they need?  Even if applicants spend $300/student every 5 years, we'll stay under cap.  The only argument for keeping the cap so low is that applicants don't need more.

So let's try it: everyone can request all the C2 they want.  If it turns out that the WCB was wrong about C2 needs and demand exceeds the program cap, then approve applications on a first-come, first-served basis.  Yes, EPC might buckle under the weight of applicants trying to be the first to submit, but the strain would probably be less than the current strain at the end of the window.

But wait, why have a gold rush of applicants jumping on EPC at the start of the filing window?  Why have a filing window?  If we're approving applications as they arrive, there is no need for a filing window.  

And since C2 caps and 2-in-5 would both be gone, we'd no longer have a need for C1 and C2, would we?

Well, I did not expect to end up there, but wouldn't it be loverly?

Thursday, May 02, 2019

The Bus is Back

Some in Congress are trying to force the FCC to allow WiFi on buses.  When I blogged about WiFi on buses five years ago, I was not sanguine on the idea.  Now, I'm OK with it.

But in order to meet the expectations of long-time readers, I'll start with my complaints about the idea.

First, is the Digital Divide (or Homework Gap or whatever) a big problem?  In a 2017 survey, only 13% of students said they sometimes they cannot do homework because they lack Internet access outside of school.  OK, it was an online survey, but still....

Second, we can't solve the Digital Divide (or Homework Gap or whatever) during the bus ride home. 
  1. Only about half of kids take the bus
  2. Bus rides aren't long for most kids.  How long?  Well...
    1. One vendor says 40 minutes a day.
    2. North Carolina's DOE says 24 minutes/ride, 48 minutes a day.
    3. Arkansas says 49 minutes each way, 98 minutes a day.
    4. In West Virginia, it's 40.7 minutes in the morning, 81.4 minutes a day.
    5. The largest research study I found (covering rural schools in 5 states) showed that even in rural areas, 15% of kids rode less than 30 minutes, 75% less than 60 minutes.
  3. Bus WiFi is only useful if connected to the Internet.  How stable is the Internet connection on those long rural routes?  I predict a new excuse: "The bus WiFi ate my homework."
Third, connectivity is half the problem.  The other half is a device.  Students can only work on the bus if their district allows them to take laptops or netbooks home.  A 2017 survey found that 14% of schools give students a computer to take home [skip to slide 19 of the presentation].

On to the things I like.

First, we have the money.  The program is nowhere near cap, so an additional $250 million per year won't break the bank. (484,000 school buses x $44.97/month x $12 months)

Second, some school districts are using school buses as neighborhood access points.  Just park the bus near the end of its route, leave the WiFi on overnight, and boom! you've lit up that neighborhood.  Assuming, of course, that the neighborhood has good cell phone reception.

Third, it gives the kids something productive to do on the ride home.  Assuming that you can lock down the Internet access to prevent the kids from having any fun.

I said I was OK with it.  I didn't say I loved it.