Search This Blog

Wednesday, March 26, 2014

A whole lot of crashing

The last day of the filing window is a day of error messages.

The Form 471 tool got sluggish for a little while this afternoon, but has been pretty solid.  But once the West Coast got into their offices, the sluggish Online Item 21 Attachment tool completely bogged down.  Not only was it taking 45 seconds to get to the next screen, the MCBC (Mean Clicks Between Crash) was about 3, and since it takes at least 9 clicks to create and submit an attachment, not much has been getting done.  It seems worse than usual.  Could it be a denial of service form the folks at the Item 21 Portal?  Just kidding.

Anyway, USAC has committed to a grace period for submitting Item 21 Attachments online.  There has always been a grace period, but I think this is the first year that the period has been confirmed beforehand.

My favorite error message of the day:  When you Create a Form 471, the first screen is a search for your BEN.  If you put anything but a number in the search box, you get: "The search value must be a positive whole number."  I feel like I'm back in 7th grade math.

Best experience of this filing window: There is some combination of clicks that actually puts you in Block 2 (or Block 3 if it's a library application).  Really, I did not imagine it, and I'm not the only one it's happened to.  I'm enough of a geek that it felt cool, like finding your way into a hidden crypt.  But the funniest part is that once you're in there, you can't get out without filling in at least one box.  I have a mental image of some database geek at USAC poking around and discovering Block 2 information associated a 2014 application and wondering how it got there.  I wish I could have typed in a message like "the cake is a lie," but all the boxes in Block 2 only want positive whole numbers.

Sunday, March 23, 2014

Fun with discounts

I don't think highly of consortium purchasing, but I thought of one more way the FCC could make consortiums more widespread: get rid of the discount matrix.  If all applicants got a 65% discount, it would remove one impediment to consortium membership.  Imagine you are a 90% applicant, and you're offered membership in a consortium for bandwidth.  Joining the consortium would cut the cost of your bandwidth in half.  You have to go for it, right?  Not so fast.  What is the consortium's discount?  If the consortium will get less than an 80% discount, your net (post-discount) cost will actually go up.  Think it over: if your $1,000 pre-discount cost was cut to $500, but your discount dropped from 90% to 70%, your net cost would go from $100 (10% of $1,000) to $150 (30% of $500).

It's rarely a good idea to join a consortium with applicants that have a lower discount, unless the consortium charges high-discount districts less than low-discount districts (and I haven't seen that).  Otherwise, the other consortium members are taking money out of your pocket.  The FCC could change that by tossing the discount matrix; if everyone has a 65% discount, then no one has to worry if they're losing funding.


The fun you can have with consortium discounts first occurred to me when I worked for a small district with a 90% discount, surrounded by districts with discounts of 20-40%.  I did an Internet pairing grant with a large 40% district, so conceptually, it made sense to file a consortium application.  However, I looked at the numbers, and my district would have lost money.

But check it out: the consortium as a whole would have gotten more E-Rate funding than we did as separate entities.  Because consortia do a straight average of members' discounts.  So it didn't matter that our 40% partner was 4 times as large; we would both have gotten a 65% discount.  (If we'd had to do a weighted average, the discount would have been 50%.)  Since their costs were much higher than ours, the extra 25% they got in funding would have been much larger than the 25% we lost.  But I couldn't figure out an ethical way for that big district to compensate my little district for the increase in their E-Rate funds, so we filed separately.

And, of course, I wondered if you could take it to an extreme.  Does a consortium have to serve all the schools in a district?  Typically, consortium members include all their schools in the Block 4 sub-worksheet for the district, but could I include only some of the schools in a district in the sub-worksheet, if only those schools are being served?  I can see a lot of ways to play with that.  Now take it a step further: a lot of districts should file as a consortium.  Because NSLP participation tends to fall off as kids get older (free lunches are not cool), and schools tend to get larger.  A district with 300 kids per grade level might look like this:
Grades
# of
schools
Enrollment
Per school
NSLP %
K-5
6
300
80
6-8
2
450
60%
9-12
1
1200
40%
If that district does the normal district weighted average, it gets a 63% discount.  But if it files as a consortium, with each school as a separate consortium member, it gets a 72% discount.

I've got to think PIA would quash that, but I wonder if it's been tried.  

Monday, March 17, 2014

Wheeler? More like Circular Sawyer

Commissioner Wheeler spoke at the Council of Chief State School Officers Legislative Conference today.  The transcript is online.  Let's see if there is any information beyond what's been in his recent speeches and statements.

First, the analysis you're dying for: did the Chairman capitalize the "R" in "E-Rate"?  Yes!  I counted 46 uses: 43 were capitalized, 2 were not, and once he put E-RATE in all caps for emphasis.

As usual, I will eschew a reasoned critique of the speech, and just respond to what jumps out at me.

"At the end of this process, the way in which we pay for certain things may need to change. And the things that have historically been paid for by E-Rate may no longer qualify."  We've heard that before, but then the Chairman gets specific.  Here's the list, and my estimate of the savings to the program (based on this table and guesswork). Of course, I'll give my opinion.
  • "Narrowband pagers." <$1 million.  Yup, cut them loose.  "Paging" always created confusion among schools, which still do intercom paging.  What's with the "narrowband"?  Aren't all pagers narrowband now?  If there are any wideband pagers in the wild, will they still be eligible?  It is amusing that "narrowband" and "broadband" sound like opposites, even though they have nothing to do with each other.
  • "Legacy PBX systems are an historic relic." $0 (no PBX funding in 2013-2014).  Do you mean legacy as in "outmoded" or as in "not Internet-dependent"?  If all voice is getting chucked, then voice hardware should go, too.
  • "Mobile phones."  $176 million.  Chalk one up for Don Quixote!
  • Email. $10 million.  Why should the E-Rate pay for email when Google is offering free email in exchange for being able to mine student emails and send them ads?
  • Text messaging: $0.02 million.  Absolutely.
  • Slow speed connectivity.  I can't estimate this without knowing the "slow speed" cutoff; chopping T-1s would save $142 million.  Now this idea will create some heat.  The two reasons that schools don't have high-speed connections are: 1) can' t afford it, or 2) don't need it.  Is the FCC really going to say, "Those bonded T-1s are too poky: if you want E-Rate funding, you're going to have to spend $200,000 to get fiber pulled to your school?" or "Your usage never exceeds 20 Mbps, but the President says you need 100 Mbps.  Upgrade or lose your funding."  This is the E-Rate equivalent of "The floggings will continue until morale improves."  The poorest, smallest, most remote schools would be exclude from funding for connectivity, because even their discounted share for 100 Mbps would be out of reach.  Is that what we're aiming for?
  • Domain name registration. $0.001 million.  Please, yes.  I hate filing FRNs with a pre-discount amount of $24.95, but I can't tell my clients that their $9.98 in funding isn't worth going after.
  • Web hosting.  $27 million.  Batten down the hatches!  A wave of spamments is on the way.
  • 800 number services. $0.085 million. How did this get on the list in the first place? It's basically paying the toll charges for calls from ineligible locations.
  • Analog voice.  $377 million.  Don Quixote strikes again!  The Chairman dances around, talking about "evolve" and schools deciding to move to digital voice, but it's clear that POTS lines are in the crosshairs.  I would toss out all voice, digital and analog.  Why leave VoIP in there?  It hardly seems technology-neutral to make determinations on the eligibility of voice based on transmission protocol.  And where does that leave PRIs?  They are digital circuits, but the only work with "legacy" PBXes, and whacking them would save us another $51 million.  We'd better tighten up the jargon, too: a lot of people hearing "digital voice" will think of digital handsets, which are phones that interface with a traditional PBX.  If you want to separate VoIP from other voice protocols, call it VoIP, not digital voice.
  • What should be on this list, but is not: video.  $6 million for Distance Learning/Video Conf.  Only transport of video is eligible, and video is transported over IP, so let's toss video out of the program.
Removing cell phones and analog voice would save us $553 to $604 million (depending on whether PRIs get lumped in with analog voice).  Removing all the other services listed except "slow speed" would save us $38 million.

"I will not hesitate to recommend [raising the contribution rate] should it be warranted."  When are you going to make up your mind about whether it's warranted?  Because the rest of us in the E-Rate community have made up our minds.  Does he want to uncap the E-Rate, and raise the contribution level to cover the April 2015 Demand Estimate (which will show what it will cost to cover the cost of the reformed E-Rate)?  My very conservative estimate was that the program will need more than $3 billion per year.  FFL estimated $6.8 billion.  So even if we cut services worth $600 million, the fund will need to double, if not triple.

"Within the FCC I will soon be announcing a special strike force...to make certain there is adherence to the rules...."  "Strike force"?!  Suddenly, I feel like a Crimean Tatar.  Those of us in the applicant community feel battered enough without being hit by a strike force.  I'm sure venture capitalists often use military metaphors, but librarians and teachers do not.

"First, why is it that the prices paid by like schools in like circumstances vary so widely?"
  1. Prices drop every year, and applicants generally lock in for 3 years or more.
  2. The Form 470 process prevents schools from finding the best price.
  3. Service providers ignore the Lowest Corresponding Price rule.
  4. When you have a 90% discount, money is no object.
  5. Services that look the same can be different under the hood.  Around here, 100 Mbps over FiOS or a cable modem costs about $200/month, while 100 Mbps over metro Ethernet costs over $1800/month.  Same bandwidth number, same service provider, 9 times the price.
  6. Circumstances may not be as like as they appear.  A few miles can make a huge difference in price, depending on things like installed infrastructure, cable franchise, line of sight.
"What can we do to help all schools pay the lowest price for the best service?"
  1. Free the Item 21 Attachments!
  2. Enforce your Lowest Corresponding Price rules.  Make service providers tell applicants their LCP.
  3. Toss the Form 470.  Allow applicants to negotiate contracts as they see fit.
  4. Allow applicants to make operational SPIN changes any time they want.
  5. Don't force applicants to sign illegal contracts locking in prices 3-4 months before service starts.
  6. Lower the top discount levels.
"Second, how do we have a program that has a growth path in terms of the bandwidth it enables?"  Uncap the program.  Pare down the Eligible Services List, then fund everything on the list for every applicant.

"... only a united front will move the E-Rate program forward."  You heard the man!  Everybody unite behind my suggestions!

All in all, not a bad speech.  I think Chairman Wheeler was pretty upfront about the pain that's coming to applicants who count on the E-Rate to cover 90% of their phone bill.  I just wish he would acknowledge that the fund needs to increase, and lay off on the "strike force" rhetoric.

Sunday, March 16, 2014

A little fact-checking

Over on The Hill's blog, a couple of economists have blasted the E-Rate.  I think that Evan Marwell of EducationSuperHighway did a pretty good rebuttal on the question of whether schools need more bandwidth, so I'm going to stick to fact-checking the first post, because it seemed full of misinformation.  Follow along as I read the screed.

What's this about funding being moved from the Connect America Fund (née High Cost Program)?  Check it out: there is over $1 billion sitting in reserve over there (see page 10 of this report).  Is the FCC going to transfer that to the E-Rate?  Let's take a look at paragraph 561 of the USF/ICC Transformation Order: "Excess contributions shall instead be credited to a new Connect America Fund reserve account, to be used as described below."  What's the description below? "we direct USAC to use the balances accrued in the CAF reserve account to reduce high-cost demand to $1.125 billion in any quarter that would otherwise exceed $1.125 billion."  Nothing about E-Rate.  Not that the Commission doesn't waive its own rules when it wants to, but I don't think they've ever moved funds between programs.  I'm betting that money stays in the CAF.

"...the money comes largely from low-income telephone users...."  "Largely" makes it sound like the majority of the fund comes from the poor.  I find that hard to believe.  I don't know of any figures describing how much of the fund comes from low-income phone users, but it's hard to see how the poorest 16% of our population (that's the percentage below the poverty line) has higher telephone costs than the rest of the population plus all the businesses in the country.  Even if you expand "low-income" to 25% of the population, and say "largely" means "25%," I'm not buying that the poorest 25% of our population spends as much on interstate telephone services as the other 75% of the population plus all the businesses.

"As of 2008 (the year for which the most recent government data exist), 98 percent of public schools were broadband connected...."  Yes, but "broadband" is an imprecise and moving target.  In 2008, "broadband" meant maybe 1.5 Mbps or 3 Mbps.  Now I expect 5 times that speed to my cell phone.

So we just need enough bandwidth to allow access to Khan Academy?  OK, let's see what bandwidth Khan Academy recommends: " it's best to plan for >1 Mbps per student or device."  Hey, what do you know, that means that the higher ConnectED goal, the aspirational goal of 1 Gbps per 1,000 students, is just a little bit too small.  So let's skip right over the 100 Mbps/1,000 student goal.

"The Government Accountability Office has repeatedly criticized E-Rate programs...."  And by "repeatedly," you mean "thrice." In 20052009 and 2010.  Interesting how the GAO stopped generating reports once Rep. Joe "Bleed the E-Rate Dry" Barton stopped requesting them.  Nevertheless, the GAO had some valid criticisms.

"...school districts are using the money to buy fancy 'Promethean Boards' to replace white boards."  Not exactly.  The Montgomery County Superintendent wants to use the reimbursements they receive from E-Rate to make lease payments on Promethean Boards.  Montgomery County does what many applicants do: they pay the full amount of their phone and Internet bills, then at the end of the year they file reimbursement requests.  When those reimbursements comes in, they are free to use it however they want.  I agree it sounds fishy to say they're using E-Rate funds to purchase the boards, but there's nothing improper going on.  The E-Rate is subsidizing telecommunications and Internet costs, but the district is choosing to apply the subsidy to purchasing the boards rather than applying it to their telecom budget.

"...a 2013 expose found that over $3 million had flowed to an orthodox Jewish high school where students are forbidden to use computers...."  Yes, that does look like program abuse.  And it's not limited to NYC.

“'E-Rate' is now called 'ConnectEd.'”  Nope.  Two different things.  The E-Rate will help fund ConnectED, along with some contributions from the private sector to cover those parts of the ConnectED program that are not eligible for E-Rate funding.

"...the tax that funds it falls disproportionately on low-income consumers."  First, it's not a tax.  If the authors had read the 2005 GAO report, they would know that one of the GAO's criticisms of the program was that the USF contribution is not a tax.  "Disproportionately"?  I don't know.  USF contributions are a percentage of interstate telecommunications costs.  So for the fee to fall "disproportionately" on the low-income, their interstate telecom expenditures would have to be disproportionately high.  I guess it depends on the proportion we're talking about.  I would not be surprised if the USF contribution was a higher proportion of family income for low-income families, but I haven't seen any figures on that.  But I would bet that the percentage of contributions paid by low-income families is disproportionately low relative to their percentage of the population.  Again, I don't have any actual data, though, so I can't say.

"...no justification in law or policy."  The Telecommunications Act of 1996 specifies setting up a program to bring advanced telecommunications services to schools.

For an article by a couple of college professors, it seems to me there was a paucity of facts and an excess of unsupported claims.  And I'm disappointed that sources weren't cited.  I give it a C-.

E-Rate getting hipper?

How big an E-Rate geek am I?  The only think I know about SXSW 2014, a super-cool annual arts/culture/technology extravaganza, is that Commissioner Rosenworcel talked about the E-Rate there.  Let's see if it fleshes out her E-Rate 2.0 ideas.  She broke it down into 3 sections, so let's do the same.

Speed
  • "In the near term, we want to have 100 Megabits per 1000 students to all of our schools. By the end of the decade, we want to have 1 Gigabit per 1000 students to all of our schools."  Nothing new there.  It's a little slower climb to 1 Gbps than the SETDA suggestion, but faster to 100 Mbps than ConnectED.
  • "...simply bringing these kinds of speeds to schools makes it incrementally less expensive to deploy higher-speed broadband to the homes and businesses nearby."  Well, at least she didn't say "anchor institution."  I'll grant that in some cases, if a school pays a carrier to lay new fiber, it could lower prices a little.  But where fiber exists, the amount of bandwidth a school purchases does nothing to lower prices for anyone else.  It could increase prices by increasing demand on a finite supply of fiber, but I doubt that's the case very often.
Simplify
  • "I want us to reduce the bureaucracy associated with E-Rate."  Hear! Hear!
  • "I would like to see multiyear applications."  That sounds like a fine idea: if the terms of a 3-year contract are the same, why do I have to keep submitting 471s?  Don't look for a huge benefit, though: most schools do not have all their services on multi-year contracts starting and ending at the same time, so now they'll have to keep track of which contracts have to go on a 471 which year.  Look for a lot of "oops, I forgot that the multi-year contract which will auto-extend 6 months from now had to go on this year's 471" appeals to the FCC.
  • "...more incentives for consortia...."  Why?  "When schools work together they can navigate the process together and benefit from more cost-effective bulk purchasing."  E-Rate purchasing already layers FCC rules on top of state law, and now you want to add in a bunch of independent local government agencies trying to coordinate efforts?  And there is no evidence that consortium purchasing is more cost-effective, except a few anecdotal opinions. 
  • "...greater transparency during the review process....  Critics have charged that our existing process is a bit opaque...."  A bit opaque?!  700 pages of secret rules is more than a bit opaque.  Keeping secret the records from application review is more than a bit opaque.  The FCC consistently denies FOIA requests, in part because it says that the routine processing of funding applications by a for-profit subcontractor to a non-profit contractor to the FCC is a law enforcement action.
Spending Smart
  • "...better accounting practices that the FCC has already identified will free up for more E-Rate broadband support...."  The practices have been "identified," but not "made public," so I can't say for sure what the practices are.  I'm guessing that they're going to take advantage of the ADA exemption to oversubscribe the fund, knowing that applicants won't actually spend all the funding they request.  That is not adding to the fund.  That is taking future rollover funds and using them this year.  So we'll get a bump in available funds for a couple of years, until the oversubcription causes the rollover funds to dry up.
  • "...we need to make sure that all E-Rate support is focused on high-speed broadband.  To that end, the time has come to phase down the estimated $600 million this program now spends annually on outdated services like paging."  What's in that $600 million?  I'm betting the amount is calculated using the amounts in this table.  I don't see how to get to $600 million without throwing out telephone service, including cell phones.  
  • "At a minimum, we need to restore the purchasing power of this program by bringing back what inflation has taken away....that is nearly $1 billion. But we should go beyond this...."  That's good news!  The early rumors swirling around E-Rate 2.0 and ConnectED seemed to be about increasing the size of the program, but Chairman Wheeler seemed to putting that on the back burner.  That's bad news; if all the other reform gets done before the funding is increased, I worry that funding won't get increased.  Keep up the pressure on this one, Commissioner Rosenworcel!
I didn't really see anything new, but I don't really disagree with anything the Commissioner said, except maybe encouraging consortia.  I'm glad she's supposed to be leading the reform agenda.


Debunking the bulk

Many people in the E-Rate community seem to believe that consortia or other bulk purchasing schemes will lead to lower prices. Where's the evidence to support that hypothesis?

As far as I've seen, the only attempt at providing proof came in last summer's NPRM: “the Commission’s Rural Health Care Pilot Program showed that bulk buying through consortia coupled with competitive bidding can reduce the prices that recipients pay for services and infrastructure.” (paragraph 88)

Let's take a closer look.

The NPRM's footnote refers to paragraph 93 of the Healthcare Connect Fund Order,  which in turn refers to paragraph 83 of the Wireline Competition Bureau Interim Evaluation of Rural Health Care Pilot Program Staff Report. That report cites 4 examples:
  • a letter from a consortium (Colorado Telehealth Network Feb. 28 Ex Parte Letter) which states that the members got better pricing, 
  • a letter from a consortium stating that members negotiated individual agreements
  • two comments on phone calls that sharing reduced costs, not that prices were reduced (Pilot Conference Call Mar. 13 Ex Parte Letter (PMHA et al.) and Pilot Conference Call Mar. 26 Ex Parte Letter (WNYRAHEC et al.)). 
 The report also relies on USAC Data Reporting letters like this one, which says things like:
  • “Through conversations with Pilot projects, USAC observed that projects were able to obtain lower rates for services and to realize other purchasing efficiencies because the services were competitively bid and the projects purchased services for a consortium of HCPs. In situations where a project chose to bid as a whole network, service providers appeared more willing to provide large discounts because the Pilot projects had a large number of HCP sites.”
  • “Based on discussions with Pilot projects and observations during the course of the Pilot program, USAC believes that these consortium arrangements provided the individual HCPs with lower rates.” 
The assertion that consortia and bulk buying lower prices seems to be completely based on the anecdotal evidence that program participants told the program administrator that they had gotten good prices. Would anyone expect someone running a consortium to say that the consortium was increasing prices?

The only "evidence" that prices are lower are the self-serving comments from program participants.  There are no actual numbers.

So let me give some examples from here in NJ.

First, a couple of years back, the NJ State Library offered all the libraries in the state a deal: if you connect to one of our three nodes, we'll give you free Internet access.  At that time, they put out a state-wide bid for those connections, and they got incredible pricing on circuits from Verizon, which was the only company that had the infrastructure to supply bandwidth to every library in NJ.  So with great circuit pricing and a free Internet port, that had to be the best deal for libraries throughout the state, right?  For that product, yes.  But both Verizon and the cable companies have other Internet access products that are not enterprise-class (think FiOS or a cable modem), but are good enough, and the price is a small fraction of the cost of a circuit to the state library.  So for many libraries, the connection to the state library was a wonderful opportunity.  For others, there were better options.  And the state-wide nature of the bid precluded, for example, the cable companies from bidding, because cable coverage in this state is a patchwork of companies.

Second, we have had for years a statewide cooperative purchasing program for long distance.  The state wanted to encourage schools to use the program, so they passed a law saying that schools had to join the cooperative.  So I always advised clients to join it.  Was it the lowest price they could get?  No.  I could always find lower prices.  In fact,  I could get lower prices for individual schools from the same vendor that won the co-op bid.  Being a member of that co-op was never the cheapest way to go.  But a small district's long distance bill is a few hundred dollars a year, so the savings from lower prices would have been less than $100, so why fight the state?  Like many purchasing co-ops, this was more about administrative convenience than about price.

Third, we currently have a state-wide purchasing co-op for Internet bandwidth and VoIP.  When they went out to bid, they got good pricing, but not substantially better than members could have gotten on their own.  And at this moment, the price for an individual school is 24%-55% lower on bandwidth by contracting individually over joining the co-op.  For a 1 Gbps circuit, the co-op can be twice as expensive.  And to be clear, I am talking about the pricing from the two vendors that are on the co-op bid.  I'm not even talking about looking at alternative vendors like the cable companies, who couldn't bid on a state-wide contract.

Does consortium or bulk purchasing lower costs?  Sometimes.  Does consortium or bulk purchasing increase costs?  Sometimes.  Should the FCC be encouraging consortia or bulk purchasing?  No.  Because if the bulk purchase is the better deal, applicants will go that direction anyway.  If the FCC puts its thumb on the scale, it will only have an effect in those cases where the consortium would otherwise not be the most cost-effective, but are made so by E-Rate rules.  Do we want to encourage applicants to join consortia that are not otherwise cost-effective?

The FCC should take steps to remove barriers to bulk purchasing.  Here are some ideas:

  1. Put consortium applications at the head of the line for PIA.  Apparently, Chairman Wheeler did this.
  2. Get rid of LOAs. The consortium agreement should be enough.
  3. Reduce the need for the Form 479.  Take CIPA certification off the Form 486, and make a separate form.  Then allow us to look up online any applicant's CIPA certification status for a particular funding year.  Because right now, consortium leads have to chase down Forms 479 from members which have already certified CIPA compliance on a Form 486.
  4. Allow consortia and co-op members grow after a Form 470 is posted.  Currently, the Form 470 has to list the BENs covered, so if applicants later want to join the consortium, they aren't covered by the Form 470.  The FCC should allow them to be retroactively covered.
  5. Don't require consortia to file Forms 470.  Even better, scrap the Form 470 all together.

Tuesday, March 11, 2014

Tempest in a portal

One of the duties of this blog, I feel, is to be a sort of People magazine for E-Rate, posting information that you might call news, or you might call gossip.  This post has a kind of gossipy goal.  But of course, I won't gloss over details, so you'll have to dig to get to the dirt. [I can't decide if that mixed metaphor is ingenious or stupid.]

If you spend much time around ed tech, you may have heard of EducationSuperHighway (ESH).  It's an organization dedicated to "removing the roadblocks to high-speed, high-capacity Internet for America’s schools and paving the way for digital learning in every classroom."  The have a lot of high-profile donors and partners, and a really nice website.

ESH runs the SchoolSpeedTest, which Chairman Genachowski lauded.  Basically, schools were asked to identify their school, then run a simple speed test.  The test results were provided to the school, but also collected by EducationSuperHighway, which gave a summary of data to the FCC.  I think they have the best collection of data about school's actual Internet speeds.  I can't say how good their data is, though, because I don't know how many schools participated in the test.  And I can't say if I would agree with their analysis of the data, since only ESH has the data.

ESH's new initiative is the Item 21 Entry Portal (I21P).  Basically, they have provided a tool that allows you to enter your Item 21 Attachment (I2A) data on their site, then get an I2A suitable for emailing.  I took a look at the tool, and the forms are a little better than USAC's.  The I21P also has the ability to do bulk uploads of Item 21 data.  I won't be using the portal, but it could be convenient for some people.

The idea behind I21P, though, is to allow ESH to collect all the information in everybody's I2As, to build a big database of information on what schools are using and how much they're paying for it.  I think that's a good goal; I've already said the FCC should free the Item 21 Attachments.

So ESH has been sending emails to schools, encouraging them to use the I21P.  I will be kicking the decision to my clients, because I won't release their information without their OK.  I've told clients that I can't give them legal advice on the release of that info to a third party.  Certainly, some service providers won't like having their pricing info shared, but I'm not sure if their claim to privacy for proprietary info stands up against public school and library information being public.

Apparently, I was not alone in my concern.  The AASA (the Superintendents Association) and the Association of Education Service Agencies (AESA) posted a joint Q&A about the portal.  They were mostly pretty neutral, but they did have a couple of discouraging things to say.  T.H.E. Journal published some responses from ESH.

Oh, it's on!

Let's break it down, showing the two sides, and then my assessment:
Issue AASA/AESA ESH On-Tech
Slow applications? Using the I21P "delays the issuance of Funding Commitment Decision Letters." "USAC does not have a policy that people who file using their online tool get their applications reviewed before people who file with any other manner" Neither side can prove what they say.  I don't think anyone has evidence on whether apps with online I2As get funded faster than paper ones.  PIA procedures are secret, so we don't know the policy on processing online I2As first. 
Use of data "Participation in the ESH Item 21 Portal represents an E-Rate applicant submitting information to a private group, and that gives us and our members concern." "will use the information to publish aggregate reports at the national, state and regional level that sort of summarize the key learnings from the information. That we will provide government agencies, including the FCC the United States Department of Education, states and intermediate units, with the specific data that we gather from districts and libraries within their jurisdiction…and that we will give districts and libraries access to both that aggregate information, but also to anonymized detailed information in their region." If you give the data to the FCC and DOE, how are they going to keep it secret? ESH can't protect the data by claiming it's proprietary to ESH, so FCC and DOE would have to share, I think. The Privacy Policy does not delineate who gets what. The Privacy Policy seems to forbid ESH from giving specific data to anyone, including FCC and DOE. If everyone else gets aggregated data, ESH will have an opportunity to spin the data any way they want, and no one else will have the data to contradict them.  "Knowledge is power."1
Who benefits? "the main beneficiary of the ESH portal is ESH" "there will be many beneficiaries from participating in this process": FCC, state DOEs, districts and libraries Both are right: ESH is the main beneficiary, but others will benefit.

I think ESH has a problem with their Privacy Policy.  It severely limits how they can use Personal and/or Applicant Information (PAI), which is good.  PAI is defined as "information that alone or in combination with other information may be used to readily identify, contact, or locate you or the E-rate applicant you represent," which is also good.  The problem is that price is PAI.  Price is on the 471, so if you tell me an applicant is paying $1,253.53/month for Internet access, I'll tell you in less than 5 minutes the FRN and associated 471 and 470.  (If you wanted me to tell you for every applicant in the country, it would take me a while to cook up the right query, but I could do it.)  So price transparency would violate the Privacy Policy.  They could provide price translucency by rounding pricing off to the nearest $50.  Then it would take some detective work to get to PAI, and in some cases I wouldn't be able to say for sure, but in most cases I would be able to say definitively (especially if you give some geographic info), so translucent pricing would still be PAIish.

Price transparency is a laudable goal.  But I don't think I21P is the way to get there.  At least not the way it's set up now.

1 Sir Francis Bacon, Religious Meditations, Of Heresies, 1597.

For those of you who are thinking it's about time for me to do some grammar woolgathering, I won't disappoint.  How does one acronym (or is it "acronate") "Item 21"?  I chose to just take the first digit of "21" to form "I2A"; ESH kept the whole number in their "I21P" acronym.

Sunday, March 09, 2014

What a difference a decade makes

How had I forgotten that the per-student funding idea proposed by Funds for Learning and supported by Commissioner Pai was part of the big 2005 NPRM (the subject of the very first post on this blog), where it was called "formulaic funding"?  Back then, it seemed like the proposal might get somewhere, because the Chairman was in favor, but like most of the ideas in that NPRM, it never went anywhere.

One big shift since 2005: the position of the Council of Great City Schools.  The current per-student funding idea was their initiative.  But back in 2005, they were opposed to it.  What happened in the meantime to change their minds?  The Priority Two gravy train ran out of steam.  Once the Great City Schools couldn't count on Great Piles of P2 Funding, they realized that per-student was to their advantage.  City schools tend to have fiber running by them already, and mileage charges are not an issue, so P1 costs are lower.  Also, larger districts can take advantage of economies of scale, further lowering relative P1 costs.  So suddenly, if funding is based on cost, instead of getting the lion's share of funding, the big city schools will get a smaller share (on a per-student basis) than other applicants.  Now that per-student funding is in their interest, the arguments they made in 2005 about complexity and inequity don't seem so important, I guess.

Another serving of Pai

I like Commissioner Pai's style.  His statements always seem to be more straightforward than most of the commissioners I've seen over the years.  And he's willing to stake out his own position.  His response to the recent request for comment is a good example of both.  He comes right out and says that the notice asks the wrong questions, and asks them the wrong way.

As usual, however, I disagree with him on policy.  Towit:
  1. "Reform should mean eliminating the priority system...."  No, I think we should triple down on priorities.  My comments on this summer's NPRM included 6 priority levels. I think priority levels are the correct way to ensure the program accomplishes the Commission's priorities.
  2. "Reform should mean abolishing the discount matrix...."  I don't think all districts deserve a 75% discount, as Commissioner Pai has suggested.  I think a top discount level of 65% will be more effective in reducing waste, fraud and abuse, and provide a stronger incentive to make cost-effective choices.  Also, I like the idea of giving a larger discount to districts with more needy children.  I'm for simplifying the matrix, but not abolishing it.
I also disagree with him on procedural matters, though I don't agree with the Commission's approach, either.  Basically, he's saying that the recent request for comment should have been a Notice of Proposed Rulemaking with concrete proposals.  I take the other approach.  This summer's NPRM should have been a request for comment.  I'm no lawyer, but it seems to me that a Notice of Proposed Rulemaking should be full of proposed rules, not reform concepts.  Instead of saying, "What do you think about changing the discount level?" an NPRM should say, "We plan to make the following changes to the discount matrix...."  This summer's NPRM didn't do that, so it should have been a request for comment.  Once the FCC boils down all the comments in Docket 13-184, they should do an NPRM to let us know what rules they plan on making.
Still, I just can't get my fill of Pai.

Ignorance

I was just sitting here thinking on a sunny Sunday, and it occurred to me: why doesn't the FCC scrap the prorating rule?  I checked the big NPRM from the summer, and I couldn't find any mention of prorating funding within bands.

They really should scrap the rule, because they violate it every year.  47 C.F.R. 54.507(g)(iv) says that when the money runs out, USAC should take whatever's left and distribute it on a pro-rata basis.  Every year, USAC has not provided any funding to a discount level it couldn't fully fund.  Applicants at the 80-89% discount levels have grown accustomed to it.

But this year even 90% applicants got nothing, and they are complaining.  And if you look at the rules, they have a right to complain; according to the rules, USAC should have taken the $200 million currently unused in the 2013-2014 pot and doled it out to the 90% applicants on a pro-rata basis.  Of course, they would have only gotten something like 12% of their request, but rules are rules.

So if you're going to ignore the rules every year, maybe you should change the rules.

Saturday, March 08, 2014

A wave on the horizon

When will the first FCDL go out?  Yesterday the FCC approved the Schools and Libraries Funding Year 2014 Program Integrity Assurance FCC Form 471 Review Procedures.  That means PIA can press the "Approved" button any time.  Is this the first time the procedures have been approved inside the filing window?  Let me take a quick look back:
Window close Days passed PIA approved Days passed First FCDL Total days
3/26/2014 -19 3/7/2014
3/14/2013 57 5/10/2013 19 5/29/2013 76
3/20/2012 35 4/24/2012 77 7/10/2012 112
3/24/2011 75 6/7/2011 19 6/26/2011 94
2/11/2010 91 5/13/2010 13 5/26/2010 104
2/12/2009 49 4/2/2009 26 4/28/2009 75
2/7/2008 63 4/10/2008 21 5/1/2008 84
(I stopped at 2008 because I couldn't quickly find earlier procedure approvals.  Maybe the FCC didn't approve procedures before that?  Or maybe the approvals were as secret as the procedures?)
It is the first time the procedures have been filed inside the window.  The FCC has never before come within a month of the close of the window.

So now that the PIA procedures are approved, there is theoretically no reason that we couldn't see some FCDLs on Tuesday.  But that seems unlikely; while some applications have gone into PIA review, there doesn't seem to have been enough PIA activity to create a wave.

When will we see the first wave?  The optimist in me says March 25th.  Because it would look nice to have the first approval take place inside the window, and historically, USAC has been able to turn around the first wave in 19 days from PIA procedures approval.  My inner pessimist says May 20th, because the first wave has only come earlier than May one time.

Normally, I'd say that USAC would have to wait until they release the Demand Estimate, so that they can be sure they have enough funding to meet Priority One demand.  The Demand Estimate seems to come out about 40 days after the close of the window.  But this year, the recent NPRM said that the Commission has found an extra $2 billion, and in a recent speech, Chairman Wheeler committed to funding Priority One for FY 2014-2015.  So maybe USAC doesn't have to wait for the Demand Estimate.

Still, betting on some time in May seems sensible.

Friday, March 07, 2014

Bad ideas repeated

The Public Notice seeking comment, promised by Chairman Wheeler and leaked last week, has been published.  Let's see what's in it.  Just a warning, I'm feeling a bit cranky today.

In paragraph 7, we see the claim that "an initial review by Commission staff has found that the Commission can free up an additional $2 billion over the next two years...."  Can we get a peek at that review?  Where are they freeing it from?  The Low Income savings aren't going to be that big.  The Chairman mentioned freeing up money from appeals; are they going to implement the asinine suggestion from the recent NPRM that all appeals more than 180 days old be denied?

Paragraph 8: "57 percent of district leaders do not believe that their schools’ wireless networks have the capacity to handle a one-to-one student-to-device deployment."  My reaction: most of the other 43% of district leaders are in for a rude surprise.  802.11 wasn't conceived to handle 30 devices in a single room, with 30 more in the adjacent rooms.  For coffee shops, it's fine.  For a school building, buckle up, it could get bumpy.

Paragraph 12 raises a real conundrum.  It's absolutely right that if you suddenly pump 1 Gbps to most applicants, their firewalls, routers and content filters will choke.  But if we're trying to figure out how to find enough money to fund things, it doesn't seem like a good idea to expand eligibility.

Paragraph 14:  The 1-in-5 Rule!  Ack!  Didn't we just read in paragraph 9 that the 2-in-5 Rule was a failure?  This rule will change applicant thinking from "What do I need next year?" to "What might I need in the next 5 years?"  Applicants buy crap they don't need because they might need it years from now, and they won't be able to get E-Rate funding then.  Did someone say gigabit by 2017?  Better buy gigabit-capable router, firewall and content filter this year, even though by 2017 those products will probably be services in the cloud, so I'd never actually need the equipment, and in any case would be much cheaper in a few years.

Paragraph 16: I like this approach to pro-rating: fund the applicants with the most low-income first, then work your way down until the money runs out.

Paragraph 17:  As bad as the 1-in-5 Rule is, the Dinner Table Rule is worse.  Put yourself in the position of a tech director who can get E-Rate funding on whatever equipment you want this year, but there is no way of knowing the next time you'll get funded.  Of course, you want to buy a ton of gear, but then you have to wonder how you're going to replace it all if your number doesn't come up for another 10 years.  This rule takes the unpredictability of P2 funding to the next level.

Paragraph 18: "If the Commission were to use available funds to front-load support for eligible internal connections in funding years 2015 and 2016...."  What planet are these people on?  For 2013-2014, even if the Commission had "freed up" an extra $1 billion from whatever mysterious source they have, it would not have covered P2 even at just the 90% level.  If they want to fund Internal Connections for 40%ers, they're going to need $billions without front-loading.

Paragraph 20: Now here's a compromise that might work.  Apply the per-student funding model to Priority Two only.  I like it more than the straight per-student model.  But it's not a good way to focus spending on FCC priorities, since it allows each applicant to focus on its own priorities.

Paragraph 22: How would per-student p2 give funding certainty?  If we're going to fund all P1, then do per-student for P2, then they won't be able to set the per-student amount until after the funding window.  Or are they going to set aside, say, $1 billion for P2, and then pro-rate P1?  But even if they set aside $1 billion, then the amount per student would be affected by the number of applicants applying for P2.

Paragraph 26: It is a shame that schools and libraries can't afford to pay to get fiber laid.  The E-Rate program isn't really set up to deliver large, one-time infusions of capital, so why don't we set up a separate program for that?  Let's give it plenty of money, maybe 3 times the annual E-Rate budget.  Oh yeah, we did that in 2009: $7.2 of the American Recovery and Reinvestment Act was used for bringing broadband to places where it wasn't.  BTOP was part of that.  Take a look at the tiny dent $7.2 billion made.  Let's not pretend we can do better with $2.3 billion.

Paragraph 27: Why minimize recurring costs?  Both school/library budgets and the E-Rate program are much better able to handle recurring costs than one-time expenditures.

Paragraph 28: Why would you want to give more support to up-front costs?  The idea of incentives is to change the cost-effectiveness calculation.  Why would the FCC want to encourage applicants to pay a large up-front fee if that arrangement would not otherwise be as cost-effective as the alternative?  If the applicant decides an up-front fee is more cost-effective, fund that.  If spreading the cost out is more cost-effective, fund that.  I don't see any need to put a thumb on the scales.

Paragraph 30: Support to help schools to meet bandwidth targets.  What on Earth?  This paragraph seems to be saying that if you're being a responsible tech director, and you've realized that there is no real reason for you to install a 100 Mbps connection to a school that isn't close to saturating its current 50 Mbps pipe, we're going to move you to the back of the line, behind those applicants who want to install some wasteful dark fiber.  Or are they saying that if you already have a 100 Mbps connection, they're going to cut you off the following year?  Again, there is no reason to put a thumb on the scales.  Let each applicant purchase the bandwidth that makes sense for that applicant.  If that doesn't match some bandwidth targets that SETDA pulled out of the air, which the President repeated in a speech, well, I just can't feel bad about that.

Paragraph 31: Ways to prioritize deployment costs.  Wait, isn't broadband going to be Priority One?  So now you're preparing us for the fact that these whacky new rules might mean Priority One demand is going to exceed available funding?  OK, I know that the Chairman wants to try his "business-like management practices" before raising the contribution level, but after we've applied those practices and we're still out of money, isn't it time to increase contributions?

Paragraph 33: Requiring amortization doesn't hurt any specific size of applicants.  Most service providers will finance the installation cost over whatever period the customer wants.  The amortization does cause an increase in cost, though, since the service provider is going to have to pay commercial interest rates, and if I were a bank, I'd want a high interest rate to loan money to a company laying fiber; they go belly up all the time (see some examples at the end of this post).

Paragraph 35: Consortium purchasing.  Again, why put a thumb on the scales?  If applicants can get good pricing through a consortium, fund that.  If purchasing individually is cheaper, fund that.  Why spend money trying to get applicants to move to a consortium when it's less cost-effective?

Paragraph 36:  Require tech plans for all services.  Hey, if applicants have a 3-year tech plan, how about not changing the tech planning rules every 2 years?  I think tech plans can be useful, but not mandatory tech plans.  Applicants who are only making a tech plan just to get E-Rate funding will do the minimum, then put it on a shelf for 3 years.  And what evidence is there that tech planning cuts spending?  If I want a gold-plated network, I'll make a gold-plated tech plan.

Paragraph 37:  Get the Item 21 Attachment data into a standard format, then publish the data.  The FCC is already collecting the data, they just can't use it in the current format.

Paragraph 40: It would appear from this section that the question is no longer whether to toss voice, but when to toss voice.

Paragraph 42: "We expect that the diminished availability of E-rate funding for voice services will be ameliorated by the fact that many applicants have transitioned or will transition to VoIP, which is generally considered to be more cost-efficient than traditional voice services."  Who considers it more cost-effective?  Let's check the footnote.  Ah, VoIP providers consider VoIP more cost effective.  But let's say VoIP were cheaper: if I cut your salary by 15% per year until it was zero, and cut your living costs by 10% one time, would you feel "ameliorated"?  I wouldn't.

Paragraph 52: Allow POTS for remote schools.  The goal is to get broadband to 99% of students.  Paying for POTS, no matter how necessary, doesn't get us there.  And hey, those applicants are already taking more than their share by paying for Internet from a satellite.

Paragraph 53: We pay for POTS until after the applicant has 100 Mbps?  First of all, that provides an incentive to keep bandwidth below 100 Mbps.  Also, the reason that we're talking about getting rid of voice is because we can't afford to pay for voice and broadband, so don't we have to toss voice before we can afford to give everyone fat pipes?

Paragraph 55: No need for demonstration projects.  Just let everyone look at the Item 21 Attachments.  The data you want is already there.

Paragraph 62:  Demonstration projects to streamline the application process?  Let me file for 100 applicants without submitting a Form 470, and file funding requests based on quotes, not illegal contracts.  Then ask me if I thought it went well.  The evidence for the effectiveness of consortia is anecdotal statements from a few Rural Health Care applicants.  I'll give you anecdotes from 100 applicants.

Hey, where is Lowest Corresponding Price?  That's the best way to lower costs.

Thursday, March 06, 2014

Tradition!

Who, day and night, must scramble for a living, and then deal with a emailbox clogged with "Notification of FCC Form 470 Posted, But No Associated FCC Form 471" letters?  That's right, it's time for my annual whine about the these useless updates.

But this year I have company.  SECA's comments on the process reform NPRM include a paragraph criticizing the notifications, concluding with "SECA believes that this notification process, as implemented by USAC, is unduly confusing and is a misinterpretation of the FCC's instructions and intent."  Normally, I'd be nettled that they put it in a footnote, but the rest of the document was asking for more notifications, so the request to scrap this notification did kind of stick out.  Besides, the E-Rate program has a proud tradition of putting important items in footnotes.

Wednesday, March 05, 2014

I'll wait for the service pack

Why is Microsoft weighing in on E-Rate 2.0?  Maybe they want to change the name to E-Rate Vista?  Let's see....

OK, it starts with a nice bit of self-promotion, using colored boxes that evoke the Windows® logo.  Except that's the old Windows logo.  Then a section on the importance of increased connectivity.  Then another section on how much Microsoft is doing for our schools and libraries.

The next section has some meat, encouraging the FCC to look at local caching, multicasting and leveraging regional networks.  Does Microsoft still have a caching product?  They no longer sell Forefront (f.k.a. ISA Server f.k.a. MS Proxy Server).  And while I'm sure Media Server can multicast, no one's decision to buy one is going to be influenced by E-Rate.  Is this whole section completely lacking in self-interest?

I have to take issue with Microsoft's reasoning in support of regional education networks (RENs).  I think their diagram correctly identifies the bottlenecks in access to content: the last mile to the school and, if a school relies on WiFi internally, the WLAN.  MS claims that RENs will be able to get substantially cheaper Layer 1 or 2 connections, rather than Layer 3.  Apples and oranges.  They're comparing a Layer 1 or 2 WAN connection to a Layer 3 Internet connection.  The connection to the REN is cheaper, but somewhere there's going to be an additional charge to connect the REN to the Internet.  Latency on the REN would be lower than over the Internet, so if the REN also provides hosted VoIP, the quality would be better, assuming that the REN has the facilities and expertise of an actual telco.  But that's a big assumption.  Also, I'm not buying the claim that "individual schools needn’t maintain individual IT expertise."  The need for local IT staff might decrease, but as long as there are Microsoft products in a school, someone's going to have to touch them from time to time.

The next section is about rethinking the prioritization levels.  There might be a bit of self-interest when an internal connections manufacturer suggests giving more funding to internal connections, but as I pointed out above, most of Microsoft's products are unlikely to benefit.

Protecting student privacy?  Where did this come from?  Oh, yeah, from here.  Google has the ability to mine data to increase ad revenues, and uses it in its free Google Apps for Education (GAFE).  Microsoft does not have the capability in Office 365.  So MS wants to make Google's capability forbidden.  I find the mining of student data distasteful, but I don't think the FCC should be too swayed by a competitor that would benefit from a change in the rules.

And then we're off to white-space wireless networking (which was apparently christened "super Wi-Fi" by former Chairman Genachowski, to the displeasure of the Wi-Fi Alliance, since it isn't Wi-Fi (it isn't super, either, but that's a different argument)).  The idea is to put wireless networking on the unused portions of the TV frequency spectrum.  It's not a bad idea, but it doesn't meet E-Rate needs.  First, it doesn't help remote schools.  No one is talking about pushing the range of SuperWiFi past 6 miles, and the SuperWiFi access point has to be connected to a wire or fiber; if you're running the cable anyway, why stop 6 miles away?  Second, no one is talking about speeds near 100 Mbps.  So while it's a nice technology, it doesn't meet the needs of ConnectED.

Am I being too cynical in thinking that this filing is a sandwich, and the meat is the attempt to use privacy concerns to remove GAFE's unsavory technological advantage in the online app suite market?

Tuesday, March 04, 2014

Step away from the bus

Another reason to kick cell phone carriers out of the program.

I recently came across an RFP from a district in GA is going out to bid for WiFi on its 168 buses.  Of course, access points are to be included in the service.  Clearly one of the cell phone carriers has been telling the district they could get E-Rate funding for this, but I see some problems:

1) Once the bus drives off campus, it's not an eligible location.  It's an interesting idea to give each bus a BEN and say that it's a mobile eligible location (like libraries do with bookmobiles), but I don't think that's going to fly.  While cell phones have an off-campus loophole, mobile data does not.  So the district would have to turn off the Internet access as soon as the bus leaves the parking lot.

2) As On-Premise Priority One Equipment, the access point fails because there is no single point of demarcation.  Are they going to try to get a free one as part of a bundle?  I don't think any carrier has a standard offering that includes an 802.11ac hotspot.

3) The RFP calls for 4G GSM service.  Sorry, GSM topped out at 3G.  All U.S. carriers use LTE for 4G.  However, this does help us narrow down which service provider came up with this terrible idea: only AT&T and T-Mobile have GSM networks.

4) Why does the RFP require 802.11ac, specifying 1.5 Gbps now, 7 Gbps later?  The LTE uplink is going to give you maybe 20 Mbps, so unless you're going to put printers or servers on the bus, that 1.5 Gbps WLAN is going to be 98.7% unused.

5) Now that recess is gone and lunch is cut to 20 minutes (with a 10-minute-long line), bus was the last opportunity for students to interact face-to-face.  Do we really need to take that away by having kids stare at their screens during the bus ride home? It's more of the "extend the school day" nonsense.  And my fellow parents know how often the bus driver will hear: "I can't get off the bus now; I'm in the middle of a boss battle!"

So we've got 1) ineligible location, 2) ineligible bundle, 3) technological impossibility, 4) wasteful over-engineering, and 5) less socialization.  But otherwise it's a cute idea.