Search This Blog

Friday, December 19, 2014

We have the dates, but not the rules

At last, the filing window has been announced.  I updated my table to reflect the new dates.

FYESL releaseWindow
Days to
 60 days?Window

What jumps out at me?  Nothing really.  A few mildly interesting things:

  1. This is the latest that the window has been announced at least since 2005.  Not way later than other years, though, and USAC had a good excuse, with the overhaul of the 471.
  2. The same window close date as last year.  I like it.  Imagine if they made it the same every year....
  3. The latest window open date since 2005, which probably means the latest start date ever (I'm too lazy to check).
  4. Only 23 days between the announcement of the window dates and the opening of the window.  Not bad, but it just feels like there should be at least 29 days between the announcement and the window opening, so it would be possible to file a Form 470 after the announcement and still have time to file a 471 on the first day of the window.

So nothing to complain about on this timing, really.  Now if we just had the Second Report and Order and Order on Reconsideration, so we could know what the rules are.

Where oh where has my E-Rate 2.1 gone?

Where's the dang Second Report and Order and Order on Reconsideration (FCC 14-189), a.k.a. E-Rate Modernization Order Part 2 or, as most applicants will call it, the Manna from Heaven Order?  It was approved on December 11th, but the only information we have so far is a press release and a recording of the meeting where it was approved.  I was going to be all responsible and read the Order before commenting, but I can't wait any longer.

The big news, of course, is the $1.5 billion dollar increase in the cap.  So now instead of  $2.4-billion-a-year program, it's a $3.9-billion-a-year program.  That is unquestionably good news, but we've been expecting that since November.  The big questions left for me:
  1. Is the cap raised for 2015-2016, or do we have to wait for 2016-2017?
  2. What about the $1-billion-a-year-for-the-next-two-years bump that the Chairman promised?  Will that be added on top of the $3.9 billion cap?  That would mean $4.9 billion for the next two years.
I'm sure we'll know the answer to question #1 when the Order and Order comes out.  I just hope the Order and Order also makes clear the fate of the $2 billion that the Chairman found.

But I'm really curious to see what the Order and Order has to say about other changes.  According to the press release, the Commission approved the following changes:
  1. Large up front construction costs don't have to be amortized, and applicants can pay up-front construction costs over multiple years.
  2. Dark fiber and lit fiber will be treated the same. 
  3. Schools and libraries can self-provision fiber.
  4. Providing an incentive for state support of last-mile broadband facilities through a match from E-rate of up to 10% of the cost of construction, with special consideration for Tribal schools and libraries
  5. Carriers that receive subsidies from the High Cost program have to offer high-speed broadband to schools and libraries at rates reasonably comparable to similar services in urban areas.
  6. $150-in-5 is no longer a 2-year pilot; it's a 5-year pilot.
My reactions:
  1. Buy that WAN!  The contract might say "lease," but if you're paying the cost of building the network in year 1, then paying a small maintenance fee, it looks like a purchase to me.
  2. Go over to the Dark side!  If you're leasing fiber, dark almost always makes more sense.  In the past, you couldn't get funding for the buildout of dark fiber, so it only made sense if there was some unused dark fiber on nearby poles, but now you can pay someone to build a dark fiber network to suit your needs.  And you know what?  This actually simplifies the program.
  3. Rollout your own!  This is going to be a good choice for a lot of applicants, especially combined with #1 above.  I'll be very interested to see if the Order gives any thought to existing self-provisioned networks.  Will pole attachment fees on existing applicant-owned fiber be eligible?  What about maintenance fees?
  4. What?  Does that mean if an 80% applicant builds a fiber network, and the state opts to kick in 10% of the cost, the E-Rate will kick in another 10%, and the applicant gets a free WAN?  Based on Commissioner Pai's comments, it sounds like it.  Oh, crap!  Look for 80% applicants to self-provision fiber rings, whacking the E-Rate for big bucks to cover the cost of installation and have the non-discounted 10% covered by state and E-Rate funding.  Even 70% schools can spend millions, get 90% of it covered, and then spread the 10% cost over several years.  The staff comments at the end of the meeting make it clear that only applicants wanting to self-provision also have to consider fiber leases, which should restrain the gold rush.
  5. Yeah, yeah, just like they have to give applicants the Lowest Corresponding Price.  I'll believe it when I see it.
  6. This makes sense.  I had been advising 90% clients that if they didn't get any C2 funding for the next 2 years, unless the FCC took action, they'd be back to sky's-the-limit 2-in-5 funding in year 3.  I never thought it was likely to happen, but I had to advise them of the possibility, and it was a small incentive for them to wait.  Now that perverse incentive is gone. 
So instead of calling it the Second Report and Order and Order on Reconsideration, maybe we should call it the Cap-Raising Dark Fiber Order.

How did the telecom carriers let this happen?  Think about what's happened since July.  Voice is being phased out.  Mobile data got chucked.  Now digital transmission is going to transition to dark fiber (leased or self-provisioned).  So the people paying into the fund are being squeezed out of payments from the fund.

O'Rielly mentioned that the definition of rural will change.  I'm guessing that's a change to the rules surrounding Urban Clusters.  The original E-Rate Modernization Order said only buildings located in Urbanized Areas would be considered "Urban," which would have moved a lot of formerly Urban applicants to Rural.  But in an Erratum, they added Urban Clusters with a population of at least 2,500.  That would have moved a lot of formerly Rural applicants to Urban.  Commenters asked to have only really large Urban Clusters be considered Urban, and it looks like they won.

Anyway, it all looks like good news for applicants.  But I want to read the Order before I decide whether it's really a good thing.

Wednesday, December 17, 2014

I'll reserve my applause

Well, I think we finally know where Chairman Wheeler found that extra $2 billion.  In a letter to USAC on reserve funding, the FCC changed the rules for how much USAC should hold in reserve for: 1) FRNs not yet approved, 2) invoices not yet paid, and 3) appeals not yet decided.

I don't know what the old rules were, but the new rules are refreshingly simple: hold enough in reserve to cover not-yet-approved applications for the current funding year and the previous two years.  Don't hold anything more for uninvoiced FRNs or FRNs under appeal.

That, combined with the new tighter rules on invoice deadline extensions, will allow USAC to cut the size of their reserve funds.  Apparently the Chairman believes it will free up $2 billion over the next 2 years.  Well, it actually only has to free up maybe $400 million per year, since the normal rollover is apparently included in the $1-billion-a-year, and that's been running $600 million or so every year.

You know I have to complain.  I have four complaints:
  1. The fund will be more highly leveraged.  Not a problem unless the program loses its ADA exemption.  And that would never happen, right?  I mean, we only had a complete shutdown of new commitments that one time like 10 years ago; since then we've gotten temporary extensions every year or two.  And the current extension runs through Dec. 31, 2015, so Republicans in Congress will have a whole year to cool down after the FCC this month ignored their objections and approved the President's plan to fund a new education initiative by increasing the USF contribution factor, circumventing the need to get Congress' approval.  Surely they won't still be feeling vengeful a whole year from now....
  2. In 2 years, rollovers will be much smaller.  The funds that will be made available in the next 2 years are largely funds that would have become available in later years once invoice deadline extensions ran out and decisions were made on appeals.  So the $2 billion is mostly just future rollbacks clawed back into the present.
  3. You want to cut the reserve for unpaid invoices?  Stop treating the invoice payment process like another opportunity to review FRNs.  If USAC said the services were OK a year and a half ago, don't let them change their minds after the bills have been paid.  Even better, on Sept. 30 every year, send out notices to applicants of any FRNs that have not been invoiced.
  4. You want to cut the reserve for undecided appeals?  Decide the appeals.  The FCC has gotten much better about deciding appeals that come in, but there is a huge backlog still sitting waiting for decisions.  Just implement my "any appeal not decided in 90 days or the appeal is automatically granted" suggestion.
But at least we got to see the rules.

Monday, November 24, 2014

Hypothecational argument

I guess no one will be surprised by the Gray Lady's position in their op-ed on increasing the size of the E-Rate funding.  That's right, our country's paper of record has once again come down solidly on the side of capitalizing the "R" in "E-Rate."

The rest of the opinion piece?  Not surprisingly, the Times came out strongly in favor of the increase.  They did have the unfortunate honesty to point out that the 16-cent increase that the Chairman talked about is actually, on average, a 48-cent increase per household, since the average American household has 3 phone lines.  But at least they had the discretion not to calculate the cost for businesses.  Not to worry: Commissioner O'Rielly already exaggerated the impact on business for the Chamber of Commerce (see my comment on this post).  Otherwise, I don't have much to say on the article.

I also stumbled across an opposing op-ed in Forbes, written by Tim Worstall.  Actually, the editorial is not about the E-Rate (it's mentioned only in the title and a quote from the Times), it's about "hypothecation."

But first, I'll write a paragraph correcting a misconception described by Mr. Worstall.  Before getting to his thesis, he takes a paragraph to launch a dig at the Times by pointing out that rural areas are underserved by sewer and phone service, so why are we focusing on Internet?  OK, first sewer.  You can live just fine without connecting to a sewage network if you handle your sewage with a septic system.  However, you can't get information from the Internet by installing a digital septic system; you have to be connected to the network.  Phones?  Well, if Mr. Worstall had done any research beyond reading the Times op-ed, he would have found that the idea of universal service originated in 1934 for telephones.  Universal service didn't expand to cover the connection of schools and libraries until 1996. Connecting rural areas to phone service has been the job of the High Cost Program, the largest component of the Universal Service Fund.

The opening of the next paragraphs says, "this isn’t really about phones or the internet, broadband or not."  I should have stopped reading there, but he sucked me in with a new word: "hypothecation."  I think I'd heard it in reference to a mortgage I signed, where it means pledging collateral, but when it comes to taxation, it's the ear-marking of revenue to a particular purpose.

Mr. Worstall makes sound arguments against hypothecation, but I have two problems with his conclusion:
  1. No recommendation concerning politics and taxation should be based on logic, since there are few subjects in which logic and reality are further apart than in politics.
  2. Moving the E-Rate into the Treasury was part of Joe "Bleed It Dry" Barton's plan to scuttle the E-Rate back in 2005.
So, rather than argue logic, I will respond like the Jersey boy I am: "Hey, pally, I got your hypothecation right here!"

Thursday, November 20, 2014

Where's that window?

So now that we know that the new Form 470 will go live tomorrow, the next question is, when will the new Form 471 go live?  On the first day of the filing window, I suppose.  When will that be?  The ESL is out, so USAC can pick a date whenever they like.

Which got me wondering: what is normally the delay between ESL release and announcement of the filing window?  The answer:
FY ESL release Window announced Days to announce Window open  60 days? Window close Window days
2005 10/14/2004 11/5/2004 22 12/14/2004 61 2/17/2005 65
2006 11/22/2005 11/23/2005 1 12/6/2005 14 2/16/2006 72
2007 10/19/2006 10/20/2006 1 11/14/2006 26 2/7/2007 85
2008 10/19/2007 10/28/2007 9 11/7/2007 19 2/7/2008 92
2009 11/21/2008 11/24/2008 3 12/2/2008 11 2/12/2009 72
2010 12/2/2009 12/3/2009 1 12/3/2009 1 2/11/2010 70
2011 12/6/2010 12/10/2010 4 1/11/2011 36 3/24/2011 72
2012 9/28/2011 11/22/2011 55 1/9/2012 103 3/20/2012 71
2013 9/27/2012 11/13/2012 47 12/12/2012 76 3/14/2013 92
2014 10/22/2013 11/20/2013 29 1/9/2014 79 3/26/2014 77
2015 10/28/2014 11/20/2014 23 12/27/2014 60

[I started with 2005 because that's when the Third Report and Order created the requirement that the ESL be released at least 60 days before the start of the filing window.]

The information in green italics is "what-if" data.  If USAC announced the window dates today, that would be 23 days after the ESL was released.  If the window opened exactly 60 days after the ESL release, that would be December 27th.

This year is different, because USAC has to deal with a major systems overhaul before opening the window.  The Form 470 changes were very minor, but the Form 471 is a whole new beast.

So my uneducated guess on the window dates is:
Window opens: January 15
Window closes: March 25

That would give us a 69-day window, which is the shortest since 2005, but not a lot shorter than recent years.

Wednesday, November 19, 2014

Now who'll get C2?

Back in July, I looked at who might get Category Two (C2) funding.  My analysis said that it would take $2.3 billion to cover 80% schools.  If the Commission gets the $1.5 billion cap increase in place for 2015, and we get the $1 billion that the Chairman plans to leverage out of the existing fund, we're looking at something like $4.9 billion.  I'm guessing that C1 funding demand will be maybe $2.4 billion (based on an earlier estimate of $1.8 billion in broadband, with growth of 10%, plus $400 million to cover phasing-out voice).  So that leaves $2.5 billion for C2.

That's a little more than enough to cover every 80% applicant.  They won't all use their entire $150/student budget, so it looks like the 70%ers will be able to wet their beaks, too.  By FY 2016-2017, we could reach 50% applicants.  By FY 2017-2018, all applicants could receive C2 funding.

Of course that assumes that spending on broadband doesn't spike.  My prediction there: increasing demand will outrace falling prices, and we will see a net increase in demand.  But I'm betting that the demand increase for broadband will be 10-20%, unless applicants start installing unnecessary 1-Gbps circuits just because the FCC says they need it (based on President Obama saying it was a good goal, apparently based on SETDA's recommendation,  which was based on...well...if we need a maximum 100 Mbps/1,000 students today (a number we dreamed up on p. 23), then we'll need...well...more in 2017-2018).  My growth estimate is based on the same kind of analysis used to arrive at the 1-Gbps/1,000-students goal; it just feels right.

Monday, November 17, 2014

The $1.5 billion is official-ish

A little more information on the $1.5 billion increase in the E-Rate.  The FCC notice says that a proposed draft is being circulated to the Commissioners.

There is a staff report full of numbers based on the CoSN and SHLB reports, but I don't have time to crawl into those numbers just now.  The bottom line number seems to be increasing the USF charge by $2/line/year to fund it.

The press release seems to be just rehashing the info from earlier reports and anecdotes, so I'm only going to scan the graphs, because I like my data in pictures.

Well, the graphs at the top of page 4 caught my eye.

  • I would love to see the data behind the graph on the left, because it absolutely does not agree with my experience.  I'd like to see how they define "Affluent" and "Low income."
  • The graph on the right shows how fast bandwidth demand is increase, which is what it's meant to show.  What is also shows, which the FCC did not intend it to, is that the Avery County School District, with an enrollment of 2,266, had, as of May 2013, a peak bandwidth need of less than 120 Mbps.  Even with the same fast growth on the chart, they wouldn't be up to 150 Mbps yet.  But the FCC says they need at least 200 Mbps to serve those 2,000 students, and they should have over 2 Gbps.  The artificial goal set by the President based on some numbers made up by SETDA will result in capacity being over 13 times greater than needed.
I look at the graph on page 5, and it strengthens my feeling that we're actually going to see a $1 billion increase in the fund.

But take a look at that graph on page 5.  It says we're going to $3.5 billion in FY 2015.  Really?  If they're going to completely change the funding landscape, I hope they do it at least 28 days before the window opens.

And faster than a speeding bullet, Commissioner Pai has voiced his displeasure.

Christmas a month early

Well, it looks like Commissioner Pai was wrong.  He said that Chairman Wheeler was going to increase the size of the fund in December.  It looks like the Chairman isn't waiting that long.  Less than 2 weeks after the mid-term elections, the N.Y. Times says the Chairman will today announce a $1.5 billion annual increase in the E-Rate.

That's good news for the program, I think, although it might paint a bit of a bull's-eye on the E-Rate program.  Congressional Republicans already proposed an amendment that would have blocked any increase in the E-Rate program.

An additional $1.5 billion per year is probably enough to cover Category 1 demand for the foreseeable future, and it may be enough to get Category 2 funding to all applicants by 2020.  If they tacked the extra $1.5 billion starting in FY 2015-2016, then we might actually reach President Obama's goal of broadband and Wi-Fi to 99% of students by 2018.

But am I the only one who thinks maybe $1.5 billion is the Chairman's opening position, and that we'll see a "compromise" of $1 billion?

Most importantly, the N.Y. Times is sticking with the big "R" in "E-Rate."

Thursday, November 13, 2014

$150-in-5 is forever

As I was poking around in the text of the actual rules at the end of the E-Rate Modernization Order, I noticed something else that will be significant if the FCC fails to take any further action: if you apply for C2 funding in FY 2015 or 2016, you're stuck with $150-in-5 forever.

 The rule that sunsets the $150-in-5 Rule, §54.502(c), says that in FY 2017 and later, the 2-in-5 Rule kicks back in for applicants "which did not receive funding for category two services in funding years 2015 and/or 2016...." So for applicants which receive funding in FY 2015 or 2016, the 2-in-5 Rule never kicks back in.

So as you're planning for FY 2020 and beyond, keep that in mind.

BMIC only temporary

Every time I read the E-Rate Modernization Order (EMO), I see something else I hadn't noticed, and that I haven't seen anyone else point out.  Today I noticed that the last sentence of paragraph 131 says: "Absent Commission action, in funding year 2017 and in subsequent funding years, support for basic maintenance, managed internal broadband services, and caching functionality, as an internal connection, will be available only to those applicants that received support in funding years 2015 and 2016 and are operating under a five-year applicant budget."

OK, sure, we all expect that FCC to take action.  And it's not shocking that caching and MIBS are temporary.  But I had not noticed that Basic Maintenance (BMIC) was on the chopping block.  And really, I think it's likely the FCC will not take action to preserve the eligibility of BMIC.  That's fine with me.  As I've said before: "Basic Maintenance is just a headache and a great place to run small scams.  No one understands the rules.  Approvals always come too late.  Wash it all down the drain."

Did you notice the possible incentive for schools to get some C2 funding in FY 2015 or 2016?  If you get C2 funding in FY 2015 or 2016, then BMIC, MIBS and caching remain eligible for you, even if they become ineligible for everyone else.  Of course, if the FCC takes no action, there won't be any funding available for C2 after FY 2016, but still.

Tuesday, November 04, 2014

Looking ahead to PIA, but not looking forward to it

Now that the new Eligible Services List has been out for a week, I've had time to ponder, "What interpretation might USAC make that the FCC didn't intend?" What has me worried today?  The items listed under categories that have been made ineligible.  I'm afraid USAC is going to interpret what the FCC said literally, and that will result in some decisions that probably don't match the FCC's intent.  Let's go category by category, using the handy list in Appendix B.

Circuit Cards/Components
Taking the broad view, all Category 2 equipment is a concatenation of circuit cards.  I think we can be pretty sure that if you buy a stock router, USAC won't consider everything inside to be ineligible.  But what if you want to upgrade the RAM on that router?  Sorry, that's a circuit card.  What if you want to connect it to fiber?  You'll need to insert a fiber network interface card, which is a circuit card.  Want to get really worried?  Read the definition of "Circuit Cards" on page 30 of the old Eligible Services list, and think about chassis switches.  Your Cisco 6509 is a $6,000 chassis with (ineligible?) circuit cards worth $60,000 or so stuck in it.  I guess since the fan tray and power supply aren't circuit cards, you can get funding on them.  Should we be planning to stack fixed-port switches?  Even then, I guess we won't get funding for the uplink GBICs, since they're circuit cards.

"C'mon," you're saying, "those are components of eligible equipment; of course they're eligible."  Well circuit cards were only ever eligible if part of eligible equipment, so the change is that circuit cards inserted into eligible equipment are no longer eligible.  I don't think that's what the FCC meant, but that's what they said.

Data Protection
Since firewalls and UPS units are exempted, we lost:

  • Tape backup: Since servers are gone, this is no loss.
  • Proxy server: Oh dear, are we going to get stuck on semantics?  Caching devices generally proxy web requests, so caching servers are proxy servers.  So I'm afraid some caching vendor is going to use the words "proxy" and "server" in proximity, and PIA will deny funding.
  • Virtual Private Networks: The FCC wants VPNs excluded, so this should be OK.  VPN functionality is included in a lot of eligible routers, firewalls, and caching devices, so PIA may bring up the issue, but it should be relatively easy to claim ancillary use.
Interfaces, Gateways, Antennas
Since antennas were exempted, the only problem I can see is an unstated rule.  If you read the rules on On-Premise Priority Category One Equipment (O-PPCOE), there is an unwritten assumption that O-PPCOE is equipment that is normally Priority Category Two, but can be made Priority Category One under certain conditions.  If it's ineligible equipment, then it's bundled ineligible equipment, not O-PPCOE, right?  If a CSU/DSU is no longer eligible under Category 2, can it still be O-PPCOE?  I don't see how.  Uh oh.  Every circuit except dark fiber is terminated with some kind of O-PPCOE, and now in many cases, some of that is going to be bundled ineligible equipment.  Oh, the pain.

I actually think this is pretty clear.  Good riddance.

New rule: Only software supporting eligible equipment is eligible.  
Old rule: Only software supporting eligible equipment is eligible.
That's not an eligibility change.

Storage Devices:
First, my unreasonably precise complaint: every piece of equipment on the eligible services list includes some storage device, even if it's just ROM.  My real complaint: "caching" is storage; you can't cache without a storage device.

Telephone / Video / Voice/video Components
Fine, all gone.

I hope that USAC will be reasonable, but I'm expecting at least one of the worries listed above to be realized during PIA this year.

Thursday, October 30, 2014

Want to know the rules? Guess.

The most cynical statement in the new Eligible Services List?  "documents on USAC’s website are not incorporated by reference into the ESL and do not bind the Commission. Thus, they will not be used to determine whether a service or product is eligible. Applicants and service providers are free to refer to those documents, but just for informal guidance."

I shouldn't really complain, since they are just being explicit about something that's always been true: USAC's website is full of their best guesses as to what the rules are.  The FCC has chosen to be crystal clear: information on the USAC website are not rules.  USAC's decisions on eligibility are educated guesses.

And that's the real problem I have with the new ESL.  See, the ESL is binding.  It's a chunk of FCC-approved rules.  By shortening the ESL and moving information like the glossary to the USAC website, they have changed all those words from rule to guidance.  (Which the FCC can later treat as a rule if they want to, like they did with an old PowerPoint slide.)

So what is meant by, for example, "Interconnected voice over Internet protocol (VoIP)"?  No one knows.  There will be "guidance" on the USAC website, but the FCC can make up whatever definition it wants, at any point in the future.

Back at my first USAC training (back when it was a closed "train-the-trainer" affair), when USAC asked how the program could be improved, my first suggestion was "publish a rule book."  And SECA requested a Comprehensive Requirements Manual back in 2011.

Now we're moving in the opposite direction.  The FCC is taking the small part of eligibility rules that were written, and unwriting most of them.  I've said before that the ESL should be much bigger, including rules on on-premise priority one equipment, leased WANs, cost allocation, etc.  It would be an intimidating book, but at least we'd know what the rules are, instead of wandering in the fog.

So where are the rules?  Spread across FCC orders.  So you're thinking, "C'mon, Dan, don't be lazy.  Just read the orders.  I mean, how many can there be?"  To which I answer: "There are over 500."  My quick count found 558, but I think maybe I counted some things that weren't really orders.  On the other hand, there may be more secret appeal decisions out there, so maybe my count is low.

If we're all expected to know the content of 500 orders, could we at least have an index?

And don't think you can skim the orders: the good stuff is often in the footnotes.  For example, the unkillable "two-signature/two-date" rule was finally killed by a footnote.

And unfortunately, even if you read all 500 orders, there's a good chance that there is no rule to cover your eligibility question.  So you're back to guessing.

I feel like the ESL was like a little rock we could cling to in the swirling sea of unwritten E-Rate requirements, and now the FCC has dynamited that rock, leaving us almost nothing to cling to.

Wednesday, October 29, 2014

Obfuscated Eligibility List

The final Eligible Services List for Funding Year 2015-2016 is out.  Since it's basically like the draft ESL, most of my comments on the draft apply to the final (except they added MPLS and made WAP antennas eligible).  Still, I can't let some of the things in the order pass uncriticized.

To address "uncertainty among some applicants regarding services that are no longer eligible for E-rate support..., we include as Appendix B to this order a chart identifying the equipment and services that were eligible for E-rate support in funding year 2014, but are no longer eligible as of funding year 2015."  So that 1% of applicants who actually read FCC orders will be aware of these changes.  The other 99% who rely on the ESL will just have to be alert.

"The ESL is not intended to serve as the central collection for all E-rate program rules and policies. Even the current ESL is not dispositive of all of the precedent related to E-rate program eligibility. It would be inappropriate to leave applicants and service providers with the impression that the ESL is a comprehensive source of all information pertaining to E-rate eligibility. As we have emphasized, applicants and service providers are required to know the relevant rules of the E-rate program and are ultimately responsible for compliance with them."
What I'm hearing is:
"It would be too much work for us to search out all the eligibility rules in this program and put them in one document, so we're requiring each of the 20,000 applicants to search out all of those rules.  And we hope that by making the ESL so obviously inadequate, it will be clearer to applicants that they have a lot more searching to do."

What is the correct spelling for the sound of teeth grinding?

"there may be other high-speed broadband services that fall within the rubric of eligible digital transmission and Internet access services."  First of all, no need to say "may be";  there "are" other digital transmission services.  Second of all, what happened to the "if it's not on the list, it's not eligible" principle?  Now they seem to be saying, "if it's not on the list, it 'may be' eligible."  At least they put MPLS on the list.

Hey, they defined caching!  That's good news.  I'll be interested to see how many ISPs start offering a caching service which uses On-Premise Priority Category One Equipment.

My concern about antennas has been addressed.  That's a relief.

So my quixotic crusades against cell phones, voice service in general and video have succeeded (and 2-in-5 is also gone, kind of, maybe temporarily).  So I think I'll focus my wrath on On-Premise Priority Category One Equipment.  In addition to being a crock, if has proven over and over to be a loophole too tempting for service providers to resist (remember the Sprint case?).  And with the new C2 budgets, even the 90% applicants will have an incentive to push their equipment into C1.  Let's get it out of the program.

Friday, October 03, 2014

Stupid 470 trick

Probably no one but me cares about this stuff, but here's something pointless you can do with the Form 470.

You know how Item 9 (Internet Access) has vanished off the form, and is now "[Reserved]"?  Well, if you pine for the full Item 9, you can still see it on new 470s.

So you fill out a new 470, then go back to view it, you will get a link something like this:,
where xxxxxxx is the last 7 digits of the Form 470. (Don't know why the last 7 digits?  Here's why.)
If you delete that "5/" out of the middle, poof! Item 9 is back!  By the way, it will be a duplicate of Item 8.

Don't have a new 470 to play with?  Use the links below to open the newest 470 at the time of this blog posting.

And yes, if you are tired of looking at Item 9 from old 470s, you can insert the "5/" and Item 9 will vanish.  Don't worry, it's just the display; nothing gets deleted or overwritten.

Monday, September 29, 2014

Freshening up the 471

Oooh, we're getting a peek at the new online Form 471 at the USAC training.

Overall, it looks pretty slick.

First nicety: Application number and security code gets emailed to you.

First minus: It looks like they've split Block 1 into little steps like the old "Form 471 Interview."  It looks like about 6 pages just to get out of Block 1.  That's great for newbies filling out a form, but it will get tedious if you have multiple forms to enter, and as the system gets bogged down close to the filing window, so that it takes 2 minutes to get each new page, the tediosity will be infuriating.

First big plus:  It looks like there is a single page where I can review the whole application, so the second set of eyes looking at an application doesn't have to plod through the application page by page.  And that page flags any problems.  Plus you can get a Print Preview of the whole form from any screen.

State LEA and State School ID has been added to Block 4.  And apparently doesn't autofill.  But it sounds like you can get away with leaving it blank (though who knows whether PIA will follow up).

You can upload the Item 21 attachments from a spreadsheet!  Yippee!  But it's not clear if you have to upload a separate spreadsheet for each FRN.

There is a cool new allocation tool, which allows you to choose which entities in your Block 4 will use the services in each FRN.  I didn't see an "Add All" button, so it may get tedious to repeatedly have to click on each item on the list for each FRN.

The system for choosing how much Category Two funding to allocate to each entity looks pretty slick.

All in all, it looks pretty good.  Not perfect, but probably a big improvement.

Good news this morning

I'm at the USAC training this morning.  Lots of clarifications and confirmations on how the E-Rate Modernization Order is being implemented.

So far, only one change: I had complained that a recent USAC News Brief said:  "[I]f your services will be non-contracted services provided under tariff or on a month-to-month basis, you MUST wait until the FY2015 option is available online before you file your FCC Form 470."

That is no longer true.  You can now use the existing 470 for month-to-month bids, too.

Tuesday, September 23, 2014

Going Rural

[Note from the future: in an erratum to the E-Rate Modernization Order, the FCC reversed course and now urban clusters are considered Urban.  So my conclusions below are reversed.]

[And from still further in the future: the FCC reversed their reversal in the Second E-Rate Modernization Order by saying that only large urban clusters were Urban, so the conclusions below are correct again.]

The Chairman mentioned that a reform goal was to see that Rural applicants don't get "cut out" of funding, and Commissioner Pai was way into ensuring Rural applicants got lots of funding, even suggesting that Rural applicants get twice the funding of the non-rural.

Well, I think I have some good news for them: funding for Rural applicants will go up.  But not the way they're thinking.  What will drive up funding for Rural applicants?  The new system for determining Rural/Urban status will create more Rural applicants in a couple of ways.  First, by looking directly at MSAs, instead of looking at counties, some Rural schools which were in Urban counties will now be classified as Rural.  Second, some schools in Urban areas will become Rural because the majority of schools in their district are Rural, which makes the district Rural, which means all the schools in the district are treated as Rural.

So I think we'll see the percentage of C1 funding going to Rural applicants increase, because the percentage of applicants that are Rural will increase.  I'm not sure how large the effect will be, though.  [Alas, schools and libraries are often built in towns instead of the countryside, which means a lot of schools in rural areas will find themselves in an urban cluster.  With the change in rules, I think the shift is reversed, so there will be more Urban applicants, meaning Rural applicants will get less funding.]  [Not so after the change to only large urban clusters being counted as Urban. While I think a lot of schools serving rural areas are located in small urban clusters, I don't think there are very many located in large urban clusters.]

It won't matter in C2 in the short run, because only applicants with a high concentration of poverty will get any C2 in the next 2 years, and there won't be any C2 funding after that. Rural applicants tend not to have the concentration of poverty necessary to get into the top discount levels, so C2 funding will continue to go to the big urban applicants.  [The Second E-Rate Modernization sure changed that equation; everyone should be able to get C2 funding some time in the next 5 years.]

Tuesday, September 16, 2014

The soul of wit, but maybe not fairness

The FCC has unveiled a new procedure for handling some appeals.  Instead of releasing an Order granting or denying appeals, for "pending matters that do not involve complicated and/or
controversial issues," the FCC will release a Public Notice with a list of winners and losers.  You can see the first such list at the end of the notice.

Is this a good thing?  Well, it's a lot better than secret appeal decisions.  And really, it's not much a change from the brusque orders (like this one) that they've been issuing since at least 2010.  It seems fine for the granting of fairly routine requests for waiver, and even some appeals.

But it doesn't seem quite right in the case of denials.  Applicants presumably thought they had good reason for their waiver or appeal to be granted.  To deny them with no explanation seems a bit harsh.  But I guess it isn't any better if the FCC just says: "We also deny 32 requests because we find that the petitioners failed to present special circumstances to justify a waiver of the Commission’s rules."  It's too bad we can't get the FCC to say, "Departure of key personnel is not a reason for a waiver" (that was the basis for the waiver request from one applicant).

Most confounding, one of the applicants (Anthony ISD) had their waiver request granted in a recent order, but denied in this order.  That case seems complex enough to warrant a little explanation.

So it's not great, but if it helps clear the backlog of appeals, I'm OK with it.

If, however, they're going to try to use this as a way to quietly dispose of the 6-year pile-up of Cost-Effectiveness Review appeals, I'll be squawking loudly.  I don't like the FCC practice of embedding rules in appeal rulings, but it's better than rulings that don't tell you what rules were applied.

Monday, September 15, 2014

The latest from $150-in-5 Rule

I recently blogged about some complications of the new $150-in-5-but-you-better-spend-it-in-the-first-2 Rule, but I missed one of the complications.

The E-Rate Modernization Order says: "...schools in districts that seek category two funding during funding years 2015 or 2016 will be eligible to request E-rate discounts on purchases of up to $150 (pre-discount) per student for category two services over a five-year period."

So while the $150/student budget is (unfortunately) school-specific, the budget rule applies to the district.  Example: if you're a district thinking that you'll wait until the $150-in-5 sunsets in 2 years (probably a foolish decision, I know), but you're building a new school next year.  If you get Category 2 funding for that new school, you force all the district's schools into the $150-in-2 straightjacket.

On the other hand, it appears in my original rant, I was wrong: you don't need to spend a little in every school in the district to start the 5-year window earlier.  Apparently, you just need to spend $1 in one school in 2015-2016, and the whole district's $150-in-5 window starts.

It's too bad that the $150-in-5 rule is applied district-wide, because USAC's Two-in-Five Tool would have become a delightful explosion of color if they'd had to keep track of whether each location was under the 2-in-5 or $150-in-5 Rule, then show status under whichever rule applied.  Maybe I would have gotten my mauve background.

So now I guess we should call it the $150-in-5-but-you-better-spend-it-in-the-first-2-and-all-your-schools-should-apply Rule.

Is the Wi-Fi-Rate chimera growing new heads?

Sunday, September 14, 2014

470 roulette

Say, here's a little information on when we'll see the Form 470.  It appears the FCC submitted it to OMB for approval Friday.  When will we see it?  Well, apparently they requested emergency processing, hoping to shorten the timeline I gave earlier, and get approval by November 3.   That will get us a temporary (six-month) approval to give the FCC time to file a regular request and get permanent approval.

Will there be any comment period for the new form?  Since it's an "emergency" request, the OMB can waive or modify the comment periods.  My exhaustive research (10 minutes on Google and an OMB FAQ) has not determined whether it even needs to be published in the Federal Register.  They have time to do a 30-day comment period and still get the approval by November 3, but I don't think we'll see one.  Let's hope there isn't anything major that the FCC overlooked in putting together the form, because the OMB isn't likely to catch it.

Once the new forms are approved, how long will it take USAC to go live?  For the 471, they'll have a couple of months before the window opens, but people are already looking for the new 470.  I hope that the FCC has shared the new 470 with USAC already.  I will make a bold prediction: the new 470 will be available on November 12th.  Anyone else want to make a guess?

Friday, September 05, 2014

More simplification

So I was listening to an E-Rate presentation yesterday, and the presenter mentioned that the new $150-in-5 Rule sets up a rolling 5-year window in which to spend your $150, and something occurred to me: no matter what, you want to spend some money next year.  You don't see why?  Check it out.

Let's take the hypothetical case of a 90% district (because only 90% applicants guaranteed to get C2 funding over the next two years) that is planning a complete upgrade of their network in FY 2016-2017 (year 2 of the Chairman's $1-billion-a-year funding goose).  If they apply in FY 2016, then the 5-year window for the $150 cap starts in FY 2016, and expires FY 2020.  But if they apply for funding to put a $1 patch cable in each building in FY 2015, then spend the remaining funds in FY 2016, the 5-year window for the cap starts in FY 2015, and expires in FY 2019.  The budget limit ends a year earlier.  So every single applicant in the program should apply for funding in FY 2015, hoping to get the 5-year window started.

Crazy, right?  But of course it's not that simple.  The FCC perversely chose to make the $150-in-5 cap sunset after 2 years.  So unless the FCC takes action over the next 2 years, in FY 2017 those applicants who were approved for C2 funding in either FY 2015 or FY 2016 will be saddled with the $150-in-5 cap, while the rest of the E-Rate community will be back to the 2-in-5 Rule, where you spend whatever you please every 3 years.

So let's say you're thinking of doing a Wi-Fi refresh in FY 2017.  If you spend $1/building in FY 2015, and the $150-in-5 stays in effect, you win: by pre-spending that $1, you moved the end of your cap up from FY 2021 to FY 2019.  But if the $150-in-5 Rule expires, pre-spending the $1 means you're grandfathered into the $150/student cap, whereas if you'd just waited, you would have no cap on your FY 2017 spending.

Crazier, right?  But it's still not that simple.  Once the Chairman's temporary $billion/year leveraging runs out, the amount of funds available for C2 will be completely inadequate, and possibly non-existent.  The $150-in-5 Rule is really the $150-in-5-but-you-better-spend-it-all-in-the-first-2 Rule.

Priority 2 funding was always a crap shoot.  But now Category 2 funding is a crap shoot with new rules, which may or may not change back to the old rules on your third roll of the dice.  Oh, and by your third roll, the casino probably won't have enough money to make a full payout on your bet, and may even be broke.

Place your bets!

When will we see the 470?

Normally, USAC allows us to start posting Forms 470 for the following year sometime around July 1st.  This year, we're waiting for the new form.  OK, how long will we be waiting?  Call CSB, and they'll say it will be available "soon."  Well, according to Office of Management and Budget (OMB):
"The following process is used to obtain OMB approval for an information collection.

  1. The agency develops an information collection that it wishes to implement.
  2. The agency publishes a Federal Register notice about the proposed information collection and provides the public with 60 days to provide comment on the proposed collection.
  3. The agency considers the public’s comments and makes changes as appropriate to address concerns raised by the public.
  4. The agency submits the ICR to OMB for review and publishes a second Federal Register notice announcing the start of OMB’s review. The second notice provides the public with an additional 30 days to provide comments.
  5. After reviewing the ICR and considering public comments, OIRA concludes its review by approving the collection or taking one of the other actions noted above."

The latest rumor I'm hearing is that the draft form will be available by the end of the month.  So let's make a little timeline:
Sept. 26: FCC releases the draft for public comment.
Oct. 2: The notification hits the Federal Register.  The 60-day comment period starts.
Dec. 1: The 60-day comment period closes.
Dec. 8: After spending a week ignoring the public comments (just kidding), the FCC sends the forms to the OMB.
Dec. 17: Notice is published in the Federal Register that the forms have been sent to OMB.  The 30-day comment period starts.
Jan. 16: The 30-day comment period closes.  OMB begins reviewing the forms and public comments.
Feb 11: OMB approves the form. (I can't find any guidelines for how long the OMB takes, but this seems to be about how long it took to approve the 470 changes in 2013.)

That is not an outlandish or pessimistic timeline.  So if USAC is johnny-on-the-spot and releases the Form 470 the next business day, it will be available on Feb. 12.

So an applicant who isn't busy that day can file on Feb. 12, giving them an Allowable Contract Date of March 12.  The latest rumors I've heard is that the window will close mid-March, so those applicants have maybe a week to select a vendor, negotiate and sign a contract, and file their Form 471.

Not to worry: a recent USAC News Brief said that we can use the old form, as long as we note in Item 13 that the form is for FY 2015-2015.  But what's this notice in bold at the bottom of the News Brief?  "[I]f your services will be non-contracted services provided under tariff or on a month-to-month basis, you MUST wait until the FY2015 option is available online before you file your FCC Form 470."

True, most services are going to be under contract (with telecom rates dropping so fast, service providers are desperate to lock in today's rates), but what district doesn't have a BANA circuit or POTS line for alarms or environmental controls or fax lines or whatever?  So what now, I file a 470 now for contracted service, and file a separate 470 just for some stupid POTS line?

This program just keeps getting simpler and simpler.

Monday, August 25, 2014

Prices not yet lower than expectations

Enough whining: it's time to praise the FCC for getting it right on some things.  And by "getting it right," I mean, of course, "agreeing with me."

Back in March, I posted a response to a speech by Chairman Wheeler.  In the speech, the Chairman asked: "What can we do to help all schools pay the lowest price for the best service?"  I made six suggestions.  Let's see how the FCC did on implementing them.
1. Free the Item 21 Attachments! Yes!
2. Enforce your Lowest Corresponding Price rules.  Make service providers tell applicants their LCP. Maybe we'll see enforcement.  I still don't see any punishments mentioned, but the new rules are clearer, so that’s some progress.  And freeing the Item 21 Attachments is a step towards informing applicants of LCP.
3. Toss the Form 470.  Allow applicants to negotiate contracts as they see fit. A little.  The Form 470 is tossed if you use a Preferred Master Contract or buy Internet access for less than $3,600, but you can’t negotiate in either of those cases, and the rest of your purchases still have to use the Form 470.
4. Allow applicants to make operational SPIN changes any time they want. No.
5. Don't force applicants to sign illegal contracts locking in prices 3-4 months before service starts. No.  Well, there is some weird new guidance about not needing a written contract in some cases, but it’s useless and confusing.
6. Lower the top discount levels. Yes.  But they only lowered the 90% discount to 85% for Category 2, which is the absolute least they could do.

So only one unqualified "No."  At least we saw some movement on all but one suggestion.  That's pretty good for the FCC.  And freeing the Item 21 Attachments is really important.  If LCP enforcement happens, that will be huge.

Overall, I give it a C-.  But I'm a pretty tough grader.  If I were grading FCC decisions on a curve, I would give their performance on this list a B, maybe even a B+.

The dark side of a good reform

Will the new district-wide discount level mean more or less funding for districts?  The answer, of course, depends on the district.  But Funds for Learning has given us a big picture.  You know I like pictures, and the graph is a nice summary, but in this case, it boils down to two numbers:
32.4% of districts will get less funding because of the new calculation
14.9% of districts will get more funding because of the new calculation

OK, Block 4 got simpler for some applicants.  Those with more than one location which aren't applying for Category 2 and have no other reason to separate out locations (like different services to different locations) will no longer have to supply location-by-location NSLP data.  That's a good thing, and a real simplification of the process for applicants.  It is also a significant streamlining of the PIA process.

But it's a bitter pill that the overall effect of this simplification will be a cut in funding.  Almost a third of applicants will see their funding cut.

Is it just me, or does it seem like all the "reforms" mean less funding for applicants?  At least this one is a real simplification, unlike so many other changes, which are mythic simplifications.

Friday, August 15, 2014

Kafka Lives!

I just stumbled across a blog post from 2008 concerning the pile-up of appeals and complaints about Cost-Effectiveness Reviews.  It ended with: "So when will we see some FCC action on this? ... Let's hope we don't have to wait five years."

It's been almost 6 years, and no decision has been issued on the appeal I linked in that blog post.  (Unless it was decided secretly.)

The FCC has spoiled another attempt at humor by exceeding my hyperbole.

Secrecy 2.0

We've got secret rules, secret forms, secret funding sources, secret evidence, secret reviews, secret reforms, secret brightlines, secret selection algorithmssecret de minimis standards.  And now you can add to the list: secret appeal decisions.

Seems that back in 2007, the FCC decided several appeals by remanding the applications to USAC.  By email.  As they should, they notified the public of their decision.  Seven years later.

So what if they took their time informing the public about their decision?  No big deal, right?  It's true that the applicants were given a second chance at approval, and at least some of the denials were overturned way back when.  But the applicants might have liked to know that the FCC granted their appeal.

More importantly, we could have used this decision as a precedent.  The denials were part of a USAC practice of requiring that applicants have all the documents listed in the document retention requirements of the Fifth Report & Order (the source of the unkillable "two-signature/two-date" rule).  USAC denied many applications because an applicant didn't create one of the documents in the list.  Several years later, USAC realized that the FCC never meant that all those documents had to be created, only that, if created, they had to be retained.  If this appeal had been published, it would have helped us convince USAC earlier.

What's next?  Secret Commissioners?  Secret disbursements?

Wednesday, August 13, 2014

Hey, Chairman Quint...

Final rant on about the staff report on the E-Rate Modernization Order: projected savings from kicking services out of the program.

I don't know where the FCC got their numbers.  I took USAC's numbers for FY 2013 and dumped them into a spreadsheet.  There is some room for argument, but what I come up with is:

  • $605 million for voice ($700 million if you toss in PRIs, which seems fair)
  • $42 million in other services tossed out of the program

So 5 years from now, when voice funding gets to zero, there will be maybe $750 million in savings.  Not the $968 million the FCC claims.

And here's the real error: that assumes flat demand for broadband.  Look at the increase in P1 demand over the last five years: 29%.  If we take the $1.4 billion in FY 2013 requests that would still be eligible under the new C1 rules, a 29% increase would be $408 million.

So if the FCC's attempt to accelerate broadband deployment is a total failure, and C1 spending keeps increasing at the same pace that it has been increasing, we'll see a savings of $342 million ($750 million in savings less $408 billion in spending increase).  If the FCC is successful in accelerating broadband deployment or in dragging more libraries into the program, that $342 million could evaporate.

We're going to need a bigger fund.

Just some dumb broadband

Part two on my ranting about the staff report on the E-Rate Modernization Order: External Connections.

First, I like using "External connections" instead of "Category 1."  Could we make that official jargon?

The definitions in paragraph 15 don't really describe the situation in most districts.  The definitions make it appear that the purpose of a WAN is to deliver data to the Internet.  Even with so many services moving to the cloud,  most WAN traffic is not bound for the Internet, especially in districts with VoIP internally.  Also, many applicants will have at least two aggregation points.

The data about fiber starting in paragraph 17 is confusing need with ability.  The numbers don't seem right to me, but maybe they're close.  Even if they're accurate, they don't indicate a problem.  Only 15% of libraries have fiber?  Maybe.  But do they need fiber?  I'll use my local library as an example.  For $159/month, they could get 100 Mbps over fiber to their building, since Verizon has FiOS fiber running right in front of the library.  Instead, they get 100 Mbps from the cable company for $139/month.  Would it really improve the program to have them pay an extra $20/month for the same bandwidth, just so they could say they have fiber?  Don't talk to me about scaling: FiOS is not part of Verizon Business Global (which does connections of 1 Gbps and greater), so if the library decides to go to 1 Gbps, it's going to mean pulling a new fiber, no matter which company they're using for their current 100 Mbps connection.

"Based on this analysis, we estimate that only 40 percent of schools will purchase at least 1
Mbps per student of last-mile bandwidth in Funding Year 2014, and only 10 percent of schools will purchase the 10 Mbps per student of last-mile bandwidth needed to achieve the SETDA last-mile goal....  EducationSuperHighway has found ... that over 90 percent of schools are currently below the 1 Mbps per student five-year goal."  And based on my analysis, only 1% of schools will actually use 1 Mbps/student in 2014, and 0% will use 10 Mbps/student.  SETDA just pulled some numbers out of the air.  I can't argue with their guess on the need five years from now, but I can say that at present, using 1 Mbps/student as a standard is just promoting waste.

Those numbers in paragraph 32 look pretty conclusive for the power of consortium purchasing.  Network Nebraska dropped member costs from $87 to $1.28 per megabit per second per month.  A savings of 99.985%.  Oh wait, if you read the actual submission, that's over 8 years.  

For comparison, I'll pick a local district here with about 1,000 students.  In 2006, they were spending $568.65/Mbps/month (breathtaking when you look back on those T-1 prices, isn't it?).  In 2014, they're paying $24/Mbps/month.  A savings of 99.958%.  Or they could go to a 1 Gbps connection for $3.90, for a savings of 99.993%; that 1 Gbps connection would make everyone happy, from President Obama on down, but 90% of the bandwidth would be unused, and it would be a waste of $16,200/year in E-Rate funding.

By the way, that district has access to a purchasing cooperative.  If they joined that consortium and contracted with the same service provider, their costs would go up from $24/Mbps/month to over $34/Mbps/month.  Consortium purchasing would drive their cost up 43%.  (The price for 1 Gbps is 173% higher through the consortium.)  Same vendor, same service.

When bandwidth prices have dropped 99% over the last 8 years, showing that your consortium has cut prices by 99% is not demonstrating success.  And Mississippi definitely shouldn't be bragging about lowering costs only 90% in 9 years.

[Side note: Network Nebraska joins Utah Education Network on my list of consortia that are no longer eligible for E-Rate funding under the new definition of "consortium"; they have private universities and museums in their consortium.]

So my advice to E-Rate applicants: ignore the 1 Mbps/student benchmark.  And take a close look at whether that consortium is really going to save you money.

C2 blues

The FCC has released a staff report on the E-Rate Modernization Order (I guess I have to give up on calling it 7R&O and go with EMO) "to help stakeholders and the public navigate the large and data-intensive record in the E-rate Modernization proceeding...."

You know I'm going to do an "analysis," which for this blog means, "picking out a few things here and there and making snarky comments and rants about them."  In this post, let's look at the new Category 2 (C2) info.

[But first, a quick side rant: " the schools and libraries universal service support program, commonly known as E-rate...."  I would submit that it is more commonly known as "E-Rate" (with a capital "R"), among Congresspeople, the press, and even FCC Commissioners.  Was this report edited by our Russian mole?]

I like pictures, so let's jump right to Figures 1 and 2.  The staff says it shows "provided internal connections support for only four to 11 percent of ... schools participating in the program each year...[and] no more than three percent each year of public library locations."  I say it shows that spending roughly $1 billion/year on Priority 2 met the needs of only 4-11% of schools and 1-3% of libraries per year.  Now we're going to try to meet the needs of 20% of schools and libraries with $1 billion.  Oh that's right, we're not going to try to meet anyone's needs, unless that need is less than $150/student or $2.30/sq.ft.

"In February 2014, USAC determined that, for the first time, there was insufficient E-rate funding to fully support any internal connections requests...."  Even I'm tired of this rant.

On to Figure 4.  It looks like a good justification for the $150/student number, since it looks like $150/student is the median request.  Some problems:
  1. I think this includes libraries.  Library applications have student counts in Block 4, so you can do a per-student analysis, but it really throws things off, since library requests are much smaller than school requests.
  2. More than half of applicants are over the $150/student limit.  That means $150/student would not have met the needs of most applicants.
  3. The idea of the ConnectED is to bring Wi-Fi to 99% of students.  That means better (more expensive) networks.  The amount of money that 90% applicants found sufficient to maintain networks in 2012 is not sufficient to upgrade networks at schools that have been making do for years.
  4. Those applicants could get P2 funding twice every 5 years.  So if $150 was the median need in one year, the median needed over 5 years would be $300.
What the data actually show is that the per-student cap for five years needs to be well over $300.

Figure 5: Wait, what?  How were the deciles divided?  The most profligate decile is about a seventeenth the size of the most miserly decile.  Even the two most spendthrift deciles have less than half the number of students in the stingiest decile. The top "decile" is just a few outliers.  Calling these divisions "deciles" is extremely misleading.  It's like doing an analysis based on household size, and dividing it into "deciles" starting with 0 kids and ending with 9 kids, obscuring the fact that over 30 million households would be in each of the 2 lowest "deciles," while less than 2 million would be in the 3 highest "deciles" combined.

"Under the new approach, 'schools in districts that seek category two funding during funding years 2015 or 2016 will be eligible to request E-rate discounts on purchases of up to $150 (pre-discount) per student for category two services over a five-year period.'"  Yeah, not exactly.  Applicants who apply in those 2 years will be limited to $150/student over 5 years, but under current rules, there will be no C2 funding past FY2016.  So while I earlier christened this the "$150-in-5" Rule, it would be more accurate to call it the "$150-in-5-but-you-better-spend-it-all-in-the-first-2" Rule.  By my analysis, the funding currently committed will not be enough to fund applicants with discounts of 81% or less.

Figure 5 shows that by giving applicants less than half what they need (see my comments on Figure 4 above), we can serve more than twice as many students.  Algebraically, that would be:

Hey, make it $25/student, and with the $2 billion we already have, we'll be able to cover all the school districts and those few libraries willing to put up with CIPA and the paperwork.

"funding category two requests for all schools and libraries nationwide will require approximately $1 billion per year over each of the next five years."  Not quite.  If certain assumptions turn out to be correct, funding C2 requests at $150/student may require about $5 billion.  Spreading it over 5 years is not required.  And spreading it over 5 years when you only have 2 years of funding is irresponsible.

I guess modernization ≠ improvement.

Tuesday, August 12, 2014

Let's do the time warp

Here's a sentence from the 7R&O (paragraphs 203-204) that has me thinking about slingshotting around the sun:
"...after a commitment of funding, an applicant’s receipt of services consistent with the offer and with the applicant’s request for E-rate support will also constitute evidence of the existence of a sufficient offer and acceptance."

So in April when PIA requests the contract, all I have to do is give them proof that I have received service in the future (after commitment).

Why is a contract required?  I mean, isn't the inclusion of the terms of an offer on a federal form at least as good as an email saying we accept your offer?

But really, why is a contract required?  What negative consequences would there be from allowing applicants to file without a contract?  I'm pretty good at imagining potential negative consequences, and I can't think of any.

And let's not forget, those contracts are illegal.  Can an illegal contract be legally binding?  And to be clear, an email accepting an offer would also be illegal.

I'm not a lawyer, so I may just may not be mentally limber enough to follow the contortions involved, but what do these two paragraphs change?  I no longer need a "written contract" as long as I have a "legally binding agreement."  Wait, what is a contract?  I guess the exact definition varies state by state, but let's look at some general ones.  Cornell Law School's online legal dictionary defines it as "An agreement creating obligations enforceable by law."  The FAR (Federal Acquisition Regulations) defines it as " a mutually binding legal relationship."  "Contract" and "legally binding agreement" are the same thing.

The difference must be that word "written."  But the order says, "A verbal offer and/or acceptance will not be considered evidence of the existence of a legally binding agreement."  The FCC is saying you need a written offer and a written acceptance.  That is both written and a contract.

So the FCC is saying you don't need a written contract, as long as you can provide a contract that is written.

Or prove that you have received service in the future.

Thursday, August 07, 2014

Perversity is always ready to consort with human nature

I wonder if we can get some clarification on a misconception that the 7R&O "cleared up" in Paragraph 179.  The FCC "reminded" applicants that consortia can choose a range of service providers to meet applicant needs.

How does that work?  Let's say I've got a consortium with 100 members, all looking for 100 Mbps.  I post a 470, and 3 vendors respond.  The first says, "I can supply all 100 members for $1200/month each."  The second says, "I can supply Internet to 50 members for $1100/month each."  The third says, "I can supply 90 members for $1300/month."  (Let's assume that they are all of equal quality, so price is all that matters.)  [There were a few other potential bidders, but they could only serve 20 or so members, so they figured it wasn't worth bidding.]

Now what?  Well, I guess I give the second vendor a contract for the 50 members it can cover.  But now the first vendor says, "Wait, you gave away the 50 easiest sites, the remaining 50 members are further from the existing cable plant, so I have to charge $1,400/month each."  Now vendor 3 looks good, but can only serve 40 of the remaining members, so for the last 10, I have to go with vendor 1.  Total cost? $121,000.  That's $1,000 more than if we'd just awarded the whole shebang to vendor 1.  What should the applicant do?  Am I even allowed to accept the new pricing from vendor 1, since the bid period was over when I changed the size of the project?  Once I award the 50-member contract to vendor 2, should I post a new 470 for the remaining 50 sites?

If I award it all to vendor 1, how are those 50 members going to feel when they find out they would have paid less if you'd split the award.  But if the award is split, those last 10 members will be paying $200 more per month.

Let's say my consortium was wise enough to know that only one vendor could supply all members, and divided up the bid to allow more vendors to compete.  Remind me again, how are consortia supposed to decrease pricing?  By aggregating demand.  My consortium is disaggregating demand.

I'll restate my position on bulk purchasing: sometimes purchasing as a consortium lowers prices, but sometimes it increases prices.  The FCC should do nothing to either encourage or discourage consortia.

Wednesday, August 06, 2014

Toiling in Obscurity

The draft of the new Eligible Services List is out!  Let's see what's changed.

Everything.  It's nothing like the old ESL.  It's more like a summary of the ESL.  Oh dear, is this another case where the FCC claims to be "simplifying" the rules, but is in fact obscuring them?  Let me meander through before I answer that.

There's a 7-page intro to the 5-page ESL.  That doesn't bode well.  All the information we need should be in the ESL, not in some press release that most applicants will never see.

"Streamlining the list of supported services" is delightfully orwellian.  It makes it sound like they're simplifying the program, when in fact they are cutting everyone's funding.  Later they do cop to eliminating "outdated, legacy, and other services that do not provide broadband," but even that's spun pretty hard; they could have eliminated the words "outdated, legacy and other" from that sentence.  Outdatitude and legacity have nothing to do with it: they're eliminated all services that do not provide broadband.  Direct inward dialing is not outdated.  Text messaging is not a legacy service.  The E-Rate is now only funding broadband.

Looks like they kept maintenance.  Damn.  That should have been the first thing to go.

Category 2 is only for LAN/WLAN components.  Fine; servers were a cost-allocation nightmare, and I'm sure phone systems were a thorn in the side of the telecom carriers.  Wait, gateways and antennas are gone?  While I can't think of anyone selling a separate device called a "gateway" any more, the functionality (conversion of one protocol and/or physical medium to another in order to provide continuity of data flow) is certainly eligible.  But without antennas, there is no Wi-Fi.  Granted, many access points have bundled antennas, but not all: here's an example of some antennas that had better be eligible if applicants want to use the Cisco Aironet 1260 Series, 1600e Series, 2600e, 3500e Series, 3600e Series, or 1550 Series, since "These access points require the use of external antennas to make them fully functioning units."

It's good that installation can be purchased separately from equipment.  Wait, did I just fully support a change?  Rats!

They merged digital transmission and Internet access.  Good!  Oh, wait: "Applicants are still required to identify the category one service type on the FCC Form 471."  So with no guidance from the ESL, applicants have to figure out which of their services are telecom and which are Internet.  A great example of thinking that a change simplifies the program, when in fact it merely obscures the complexity of the program.

And here's another example.  The FCC has removed the "helpful" list of ineligible services because the list doesn't look "simple."  "Also, rather than examining long lists of ineligible services, it will be more efficient for applicants to assume that any service or component not listed in the ESL is ineligible for E-rate support." If a list of ineligible services is inefficient, why is it efficient to provide a mashup of some of the eligible data transport protocols, devices, signaling, handoffs and physical media?   Let's see how I can match up my Comcast Business Class bill with the ESL.  I guess "Internet service" on my bill is the same as "Internet access" on the ESL, but there's no mention on my bill of any of the protocols or physical media on the ESL.  Is my Internet service eligible?  Oh wait, over here under "Additional Digital Voice Services" there is a $5 charge for "Cable Modem Lease."  Since "Cable modem" is on the list, I guess that's eligible, but does that mean the whole service is eligible?  And I have my computer connected to the service by an Ethernet cable; does that make the service eligible?

Wait, "any service or component not listed in the ESL is ineligible for E-rate support"?  Does that mean because the FCC didn't list MPLS, it isn't eligible?  Or is it eligible if it has an Ethernet handoff?  What about free-space optical?  Does that come under wireless?

"The proposed ESL removes descriptions of E-rate program requirements that may be related to eligibility but do not directly name or describe the services that are eligible, and descriptions that provide extra information pertaining to certain services."  More simplification by obfuscation.  Yeah, all those rules were overwhelming and confusing.  But removing them doesn't make them less overwhelming and confusing.  It just means that in addition to being overwhelming and confusing, they're also hard to find.

"The proposed ESL also removes the 'Special Eligibility Conditions' section of the ESL because the requirements therein are already explained in the Commission’s rules or in Commission or Bureau orders and USAC provides information about these requirements on its website."  I'm having trouble expressing my opinion without vulgarity.  Last time I looked, there were 781 pages on the USAC Web site and 200 FCC orders, and that was 2008.  And don't forget the rules to be found in USAC PowerPoint slides.  It is incredibly irresponsible to believe that just because a rule exists somewhere, applicants will be able to find it.

Moving the glossary out of the ESL again obscures rather than simplifies.  Only people who worked as WAN engineers in the 1990s can be expected to know that Switched Multimegabit Data Service refers to a particular protocol, not just a concept (MPLS and Frame Relay, for example, are both switched, multimegabit data services, but neither is SMDS).  And even those aging engineers might not recognize "Switched Multimedia Data Service," since they always just called it SMDS.

OK, on to the list.

The DTS/Internet list is, as I mentioned, a strange mix of protocols, handoffs, physical media, etc.
  • ATM, frame relay and Broadband over Power Lines could probably all come off the list. Back in FY 2011, they had a combined total of 220 FRNs for $5.3 million.  Since then, those numbers have certainly gone down.  And while ATM can top 100 Mbps, I don't think you can get there with the other two, so they don't serve the goal of the program.
  • "Integrated Services Digital Network"?  Really?  In my experience, ISDN is like ISTE or AFS: at some point in the past, the letters stood for something, but not any more.  Also, ISDN comes in two flavors: BRI and PRI.  PRI is only used for voice, and BRI is 0.128 Mbps, so it's not broadband.
  • You want to get rid of an outdated service?  Take "Switched Multimegabit Data Service" off the list.  So 1990s.
  • You know T-1 and DS-1 circuits are 1.5 Mbps, right?  They do not provide "broadband" unless bonded.  And there is no way you can use them to get to 100 Mbps, which is the purpose of this program now, right?  
  • Fractional T-1 circuits are not broadband.  They do not get us closer to ubiquitous 100 Mbps.
  • Telephone dial-up?  This service is more outdated and legacy than any of the services they tossed, and it's not broadband. While I would love to see someone try to bond 1800 dial-up connections to create a 100 Mbps connection, it's hard to see how paying for 1800 POTS lines is cost-effective.  Also, putting "telephone" in there is going to make applicants stretch to fit POTS lines.
  • Say, when my fax machines send a fax, they are using a POTS line the same way a dial-up modem uses it, so can I keep my fax lines eligible?
  • Hey!  The wireless Internet access entry includes info about eligibility limitations.  I thought lists of that kind of information were inefficient, and should be hidden in the name of simplicity.  
  • Cost-allocating cell phone use made the IRS throw up their hands.  Up until now, I've considered it hyperbole when people said that PIA was worse than an IRS audit, but now the FCC really is going for nitpicking preeminence.
  • Where is MPLS? FSO? Hot air balloons?
Voice services
  • "Circuit capacity dedicated to providing voice service" leads to cost allocation hell.  I have to figure out what portion of my broadband connection is used for voice?  I've advocated tossing voice, including VoIP, but this goes too far.  Taking the broader view, why is it that the E-Rate will pay for broadband to allow students to watch Nyan Cat videos, but not to dial 911?  OK, it wouldn't be technology-neutral to fund the last mile for VoIP but not analog, but since even CLECs turn their noses up at POTS lines these days, no one cares about being fair to analog voice.  "Circuits dedicated to providing voice service" would be an OK way to toss PRIs and T-1 lines that VoIP providers install.
  • "Excluding ... text messaging."  Why?  The savings from removing the charges is tiny compared to applicants' administrative cost to remove those charges and USAC's administrative cost to verify the removal.
  • ISDN should be on this list.  I would also clearly state "PRI" here.
C1 limitations
  • If an IRU is considered a dark fiber lease, what kind of upfront charges could there be?  It can't be for build-out, right?  Because construction costs are not allowed on dark fiber leases.  What else could large upfront charges be for?
  • An IRU is not a purchase agreement.  You can add a buyout clause to a contract that contains an IRU, but an IRU is by definition not a purchase.  Think about it: why would I need to pay for a Right to Use (regardless of degree of defatigability) if I own the fiber?
  • You know what we should do for fiber instead of getting into the weeds on which pieces of fiber are eligible and which aren't depending on whether its lit, dark or IRU?  Say this: "Any applicant seeking a dark fiber lease must also request and consider lit fiber proposals."  Because, really, if it's cheaper to build a new dark fiber WAN than to lease a lit one, shouldn't we fund self-provisioning of dark fiber?  Oh wait, then we'd be pulling dark fiber all over Alaska.  That needs more thinking.
  • The Internet Access portion has some more ineligible components listed.  I guess having a full list of ineligibles is inefficient, but having several partial lists is efficient.
  • Firewall service is only eligible if it meets the criteria for Ancillary Use.  We have a term for services that are only allowed if they're ancillary; we call those services "ineligible."  Firewall service is an ineligible service.  Add firewall, DNS and DHCP to that (inefficient) list of ineligible items, and then let them come in if ancillary.
  • Managed internal broadband: So if a cell phone provider sets up a microcell in a school building, is that ineligible?  Where is the line between C1 wireless broadband and and C2 wireless broadband?
Category Two
  • If "equipment ... necessary to bring broadband into ... schools and libraries" is eligible, why isn't modulating equipment for dark fiber WANs eligible?
  • Client Access Licenses are not software.  In the case of LANs, CALs are generally just a way of throttling how many devices can connect, and don't involve any software.
  • If voice traffic goes through my router or switch or access point or firewall, do I have to cost allocate?  If I don't have voice on my data network, but later move to VoIP and my voice starts going through them, do I have to retroactively cost-allocate?
Managed Internet Broadband Services:
  • Clearly, the Commission liked the idea of Wi-Fi as a Service (WaaS).  And it's not a bad option for applicants.  But I wouldn't sign more than a 2-year contract, since we only have 2 years of C2 funding.
  • Interesting that an applicant can get funding for a service provider to monitor its LAN if it's "Managed Internal Broadband," but if exactly the same service is called "Basic Maintenance," it's not eligible for funding.  With the managed service, can get back to paying service providers for maintenance whether they do any maintenance or not.  "Managed Internet Broadband Services" (MIBS) will completely replace "Basic maintenance" (BMIC).
  • Generally, WaaS contracts include hardware replacement coverage.  The BMIC section still excludes "unbundled warranties."  That used to mean hardware replacement support contracts, but maybe that's changed; we'll have to search the USAC website for the meaning of "unbundled warranties," and hope that the FCC agrees with whatever we find there.  Another advantage of MIBS over BMIC.
  • Why aren't universal service administration fees eligible, anyway?  It's not a service that you can opt out of.  I think the separate charge is an unsavory practice, but rather than punish the victims of the fee, the Commission should contact whichever branch of government is in charge of regulating what can go on a phone bill. (Hint: they'll be able to use 4-digit dialing.)
Have the eligibility rules gotten simpler?  No. The removal of everything but LAN electronics simplified C2 rules a little, but this "Managed Internet Broadband Service" is a complexity explosion.  The C1 rules are more complicated because of the addition of a third category service: applicants now have to separate out Voice from the old Telecom Services and Internet Access categories.  The eligibility rules are more complicated.

As I feared, the brevity of the new ESL is just the Commission trying to hide the complexity of this program.  If they want to release this 5-page thing as an ESL Summary, fine.  But don't try to "simplify" the rules by burying them in some hidden corner of the USAC site, where the FCC can consider them rules when it suits them, or say that they're not rules it it's inconvenient.

The ESL should be at least 100 pages long.  Because that's how complex the rules are.