Device-centric

I've been thinking about Chairman Wheeler's goal of bringing Wi-Fi to 10 million students.  My first thought was, "I guess that leaves libraries out."  Then I got thinking about how you count that.  Let's take my little town, because it's a nice small example.  Our local school district has about 1,000 students in K-8.  To provide enough Wi-Fi in their buildings, they're going to need something like 90 access points.  Those 1,000 students are served, so giving funding to the local library would be redundant.

Hang on, though: in theory, the local library serves all those kids and more.  But it has at present 4 or 5 computers in the kids' area, which I've never seen all used.  I've never seen a kid using the Wi-Fi (except after Hurricane Sandy, when a lot of us were getting our Internet there).  A 1-to-1 initiative at the school may drive up BYOD traffic as kids bring their Chromebooks in to get Internet access after school, but I can't imagine the library needing more than a single 802.11n access point to connect all the students that come in.

So the E-Rate has two choices to cover the kids: the school needs 90 APs, the library needs 3.  If you're looking for most bang for the buck, the most students per dollar, covering the library is the way to go.

If the Chairman wants to do a "per-something" budget for applicants, instead of going per-student for schools and per-square-foot for libraries, we should go per-device.  Instead of getting $100/student to implement Wi-Fi, schools would get maybe $250/computer.  It makes no sense to allocate funding based on human bodies or building size.  Bodies and buildings don't use Wi-Fi; devices do.  It should be obvious that a 1-to-1 school with 500 kids needs more Wi-Fi than a school with 1,000 kids and 200 computers.  And much more than a library that serves all 1,500 of those kids with an access point or two.

Would some applicants lie about how many devices they have?  Probably.  But we accept NSLP numbers, and look how unreliable they are.  The number of devices could at least be audited without violating anyone's privacy.

What do we do about BYOD?  Nothing.  Because if everyone has about the same amount of BYOD traffic, then the division remains equitable.  Remember, with per-whatever funding, we are no longer concerned with actual need.  We're just looking for a fair way to allocate insufficient funding.  I think that the allocation would be slightly skewed, but in the right direction.  In my experience, the fewer low-income families in a community, the more BYOD demand there is.  So ignoring BYOD would tend to put more funding in needier communities.  Nice.

And check out this unintended consequence: The more devices a school buys, the more funding it receives.  Without paying a dime for end-user equipment, the per-device allocation encourages applicants to put more devices in students'/clients' hands.  With an incentive program like the E-Rate, we should always look at what behavior we want to encourage.  Broadband and Wi-Fi are useless without end-user devices, so let's encourage applicants to buy more devices.