The view from the top

If you're big enough E-Rate geek to read this blog, you've certainly seen Mel Blackwell's Letter to the Field.  I think that, as usual, he strikes a pretty good balance between being positive and being frank.

While he notes that USAC very nearly hits its goal of processing all the "workable" apps by Sept. 24th (well, there was a big batch of FCDLs that came out Sept. 25th, but that's close enough for government work), he does note that 6,000 applications are "unworkable," and expects that most will be processed within 90 days.  "Most" means at least 3,000, so it's possible that close to 3,000 applications will be languishing 90 days from now, which is half-way through the funding year.  And Mel admits that the remaining applications tend to be larger than average.

What is "workable" anyway?  The E-Rate Modernization Order gives us this definition:

“Workable” means that a funding request is filed timely and is complete, with all necessary information, to enable a reviewer to make the appropriate funding decision, and the applicant, provider, and any consultants are not subject to investigation, audit, or other similar reason for delay in a funding decision. Funding requests from applicants that decline to respond to USAC inquiries over the summer may be considered “unworkable” for purposes of this performance goal, though USAC will process these applications as quickly as possible when school staff return for the year. 

Doesn't "...with all necessary information, to enable a reviewer to make the appropriate funding decision..." exclude most applications?  It would seem that if PIA needs to request information from the applicant, the application is unworkable.  Unless they streamline PIA procedures dramatically, most applicants will have to supply additional information to PIA, in part because even though the Form 471 is more detailed than it used to be, it still does not collect all the information that PIA needs.  It seems to me that if the Form 471 doesn't collect the necessary information to allow 6,000 applications to be processed, the form has some serious shortcomings.

But back to the letter.  We get promises about how much better EPC will make our lives, but also acknowledgement that the roll-out has not been flawless.  I particularly appreciate Mel saying, " the accelerated timing of the deployment did not allow for adequate user testing."  I think that more could have been done to bring the applicant and service provider community into planning meetings, but I guess they can't let anyone look behind that curtain, since the FCC considers the application process to be a law enforcement activity whose workings need to be kept secret from the public (and the people going through the process).

But here's some good news: "We are committed to...user testing for the new EPC-based FCC Form 471 filing process."  Where do I sign up?

Worries about an old friend

Has the ax finally fallen, even before the big outage/upgrade?  Has hraunfoss gone the way of the dodo?  If you know what I'm talking about, you are a true E-Rate geek.  If you don't, hraunfoss.fcc.gov is the name of the FCC server that holds (held?) appeal decisions.

Just today I noticed that my links to documents on hraunfoss give me a 404 error (page not found).  I have feared the disappearance of hraunfoss since 2010.  We went through a "transitions.fcc.gov" phase, and a flirtation with "www.fcc.gov/document/...."  Could the reign of hraunfoss finally be over?

One encouraging sign: the FCC links are all broken, too.  Try going to the list of 2011 WCB appeal decisions; it looks like 800some broken links.  It wouldn't be the fist time that hraunfoss had trouble getting started on a Monday.

In the short term, you can get at the documents by editing the URL: take out "http://hraunfoss.fcc.gov" and put in "https://apps.fcc.gov" at the start of the Web address.  Note that you have to put an "s" after "http."  Perhaps that will be hraunfoss' demise: everybody's moving to https, and as long as you're going to have to change every URL....

FCC going dark

Check out this announcement: from Sept. 2nd to Sept. 8th, "most Commission resources normally accessible through the Commission’s web site,including access to all electronic filing systems and electronic dockets, will be inaccessible.... The Commission’s web site will remain available, but with reduced content and limited search capabilities. ... FCC telephones will work but voicemail will not, and most Commission staff will not have access to e-mail."

As a practical matter, not really a big deal, since they're giving automatic extensions of filings due during that time.  And there's always snail mail....

But I'm dismayed: it's going to be a really uphill slog to explain that redundant connections are cost-efficient to an organization that knowingly crashes itself for a whole week.

Who asked for C2?

Oh goody, more numbers from Funds for Learning!  It's a summary of Category 2 (C2) funding utilization.  [OK, first off, no graph?!  C'mon FFL, you know I like my data in pictures.  And I can't even copy and paste the data into a spreadsheet to do my own analysis, since the table is a screenshot.  Really, the presentation could not be more annoying.]

And the data's not as interesting as I had hoped.  I wanted the data sliced and diced by discount level, school size, etc.  Oh well, let's see if I can tease any information out of this paucity of data.

First, the bottom line: The total possible C2 demand is $5.8 billion.  That's in line with earlier calculations.  I wonder if it might not be a little low, as some small applicants who have been scared off by the huge pile of paperwork (and huger pile of rules) may stop in to get their dollop of C2 gravy.

On the whole, applicants requested only 28% of the total possible C2 demand.  So thanks to rollover, the FCC didn't even have to up collections to the new $3.9 billion cap.  I'd like to speculate on what C2 demand will be like next year, but I don't really have any good indicators.  My gut says slightly less demand than this year, but I'm just guessing.

What can we tell about who's getting the money?  Let's see....  What's the total budget/site for each row?  The C2 budget is $126,243.59/site for those who requested their entire budget.  Those are some large, high-discount schools.  If they were all 85% schools, then the average enrollment would be 990 students/school. If they were all 40% schools, the average enrollment would be 2,104 students/school.  NCES says the average school has 517 students.  Only about 10% of schools have more than 900 kids.  Check out this table of average enrollment size taken from an NCES report (page A-5).

Urbanicity	Elementary	Middle	Secondary
Total	424	569	825
Urban	528	528	1,313
Suburban	492	492	1,197
Rural	354	463	577

[OK, first, I love "urbanicity."  It sounds like a word I'd make up.  It's not in Merriam-Webster or the OED, but I like it.  Can we get USAC to use "urbanicity" instead of "urban/rural status?"  Actually, "ruralocity" might be a better made-up word for the E-Rate.]
So we can safely say that there is no way that many rural schools applied for their full discount.  And no way that many 40% schools used up their full budget.

Once you get below that top row, the numbers aren't so stark.  Sites that requested some funding (but not their full budget) had average per-site C2 budgets in the $56,000-$59,000 range, which means that they tend to be larger and/or higher-discount schools, but there could be a significant number of smaller, lower-discount schools. Schools that requested no funding had an average budget of $36,344.77.

So as you'd expect, the schools that stood to get more funding were more likely to apply.

Sometimes the forbear eats you

The FCC has released the Third Quarter Contribution Factor: 17.1%.  I have the 3Q factor from 2014 handy from a previous post, so let's see how things have changed over the last year.

That's a 9% increase in the contribution factor!  What caused the jump?  Let me dump all the data into a spreadsheet and see what happened.

Well, it's not due to a jump in spending.  Total demand is up less than 1% (an 8% decrease over at Low Income compensating for a 4% increase in E-Rate and a 15% increase from Rural Health Care).
The problem is a 6% drop in the contribution base.  People just aren't making enough long distance calls.  I found the Contribution Factor announcement from early 2007 in a previous post, and the contribution base has dropped 20% since then.  It sure does seem like the FCC is going to have to end its forbearance of USF fees for ISPs.

Be still, my beating heart

Want to make your heart skip a beat?  I've got 2 ways to do it, courtesy of USAC.

The first one applies to everyone.  Go to the Data Retrieval Tool, and look up your funding for FY 2007.  Isn't it exciting to get "No Records Found Matching Your Criteria" where your funding should be?  Don't worry, it's not you.  All the data from 2007 is gone.  Where did it go?  No idea.  When did it vanish?  I can say for certain the data has been missing at least since yesterday morning, but that it was there a week earlier.  Does anyone have a "No Records" sighting for 2007 before yesterday?

The second only applies if you submitted (or certified) your application after 8:00 p.m. Eastern time on the last day of the window.  If you did, go to the 471 Display page, put in your 471 number (if you didn't submit late, try using 471 #1051523), then click the "Display" button.  Look at the "Date Submitted" and "Date Certified."  April 17th?!  Out of window?!  Not to panic; the display tool seems to be showing you the date UTC (Coordinated Universal Time (the acronym is a sort of anagram, but at least it's properly capitalized, unlike "E-rate")) instead of Eastern time.  So don't worry; your application is in window.  Probably.  In the old 471 Display page, you would have had "Certified - In Window" at the top of the form to calm you, but now you have to go to the Application Status tool to bring your pulse back to normal.

E-Rate is always a thrill!

Apples to apples to apples to apples

As promised, today we'll look at comparing the cost of fiber leases (lit and dark) to self-provisioning and bandwidth service agreements.

I tried to build a spreadsheet that broke out every conceivable cost.  Under a normal contract, you would probably be leaving the vast majority of them blank, but I think PIA is going to be instructed to ensure that all costs are being considered, so you should be prepared to defend every number in there, especially blanks.  You can see that in my hypotheticals, I didn't fill in every number for every type of service.

How to use my spreadsheet?  Fill in the yellow blanks from amounts in the bids you're considering.  Feel free to leave blanks for amounts not included in the bid.  (The spreadsheet assumes that you've already selected the most cost-effective solution in each category, so there is only space for one Dark, one Lit and one Self-Provisioned.)  You can fill in grey blanks, too; I greyed them out just because I thought it was unlikely that your contract would have those fees, but they're still included in the calculations.

Notice the tab for the BandwidthNeed worksheet.  It's basically yesterday's worksheet, and here serves 2 purposes.  First, the cumulative costs are copied into the Cumulative Cost area of the CostCalculation worksheet, so you can compare to the costs to your leased solutions.

The second purpose of the BandwidthNeed worksheet is to give you a basis for stating future bandwidth needs.  You'll need that to figure out when you'll need to buy new electronics.  If you look at the sample numbers I put in, a 1 Gbps connection is projected to be enough until Year 11.  I rounded it down to Year 10.

The handling of upgrades is a little clunky.  For lit fiber, I assumed that equipment would be leased (or included in the cost of the bandwidth), so it's just a matter of guessing...er..."determining" how much the monthly fee will change.

For dark fiber leases and self-provisioning, I assumed that a bandwidth upgrade would probably mean an equipment purchase.  So over in cell H8, you'll see that I'm anticipating an equipment purchase in Year 10 (120 months into the contract).  There are two rows, so you can put in a second upgrade if you want.

You'll notice that there are 3 terms available.  In the first row of each term, in column F you'll put the length of that term in months.  (Note that if your term is not divisible by 12, there will be an error in the Annual Cost for the year during which the term ends.) Fiber contracts often have 2 terms; the initial term, when you're paying off upfront costs, and subsequent term(s), when you're just paying maintenance.  I added a 3rd term for flexibility.  In the hypothetical contracts I put in, I have an initial term of 5 years, because the hypothetical dark fiber lease has 5 years of higher fees paying off most of the cost of stringing the fiber, then a much lower fee for the next 15 years.  Then I have the second term end 10 years into the contract, where I'm anticipating the upgrade to 10 Gbps will increase the Monthly Cost of the lit fiber lease.  The Monthly Cost of the dark lease and self-provisioned don't change, but you'll notice that the equipment purchase from H8 bumps up the Annual Cost in Year 10.

Some shortcomings of the spreadsheet:

• Only allows 2 equipment upgrades
• Only allows 3 terms
• Doesn't do a present-value calculation 
• Only one contract of each type can be inserted
• It's way more intimidating than I'd like it to be, since most of it will be blank
• Contracts with terms that are not whole years will cause a miscalculation; for example, if your first term is 42 months (3.5 years), the spreadsheet will act as if it were a 48-month (4-year) contract.

Any feedback on the spreadsheets would be welcome.