[FRIAM] The Last Mile, again

Tue Apr 24 00:24:06 EDT 2018

Steve, 

Thanks So much.  Can you explain the following  passage in greater detail?
What do you mean by "over-subscription".  

The bad news is not too bad.  First,  They can't be promising 0%
oversubscription of bandwidth... that would be prohibitively expensive and
excessive for the type of utilization  a collection of (part-time?)
homeowners would need.   Their level of oversubscription (in theory and in
reality) is what will make or break your real bandwidth.   If I read you
correctly they are offering 25Mbps down and 3Mbps up?    I'm not sure what
the range of oversubscription is, but I've heard on the order of 4x to 32x
with the former being very rare and usually only during a startup phase
before a service actually connects enough customers to begin to consume the
bandwidth.   

That seems to be what they are promising.  See my answer to Gary, just sent.
The said that they had contracted for a 1 gig pipe and that that would take
care of anything subscribers could throw at them.  I didn't make any sense
to me, but perhaps I just misunderstood.  If I understood the units, 40
users could exhaust I G if they all got on at the same time.  I assume we
are talking about Mega/Giga per second here.  So to carry out the promise as
I heard it, they would have to have a 1 gig pipe for every 40 users and they
are talking about 100-200 users to begin with.  So, is that what you mean by
"over-subscription":  the number of paid subscribers who would be left out
if everybody tried to get on simultaneously at full speed?  

What questions should I be asking them? 

Nick 

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

 <http://home.earthlink.net/~nickthompson/naturaldesigns/>
http://home.earthlink.net/~nickthompson/naturaldesigns/

From: Friam [mailto:friam-bounces at redfish.com] On Behalf Of Steven A Smith
Sent: Monday, April 23, 2018 11:32 PM
To: friam at redfish.com
Subject: Re: [FRIAM] The Last Mile, again

Nick -

I'd say there is good news and bad news... but most likely mostly good news.
(TLDR and TMI for most here):

Richard Lowenberg formed a group here in SFe some 10 years ago (nearly?)
called "the FIRST Mile" to turn the language around.   "Last Mile" leaves it
to the industry to treat us like an afterthought... and while there is solid
logic to building a network from the center out... it doesn't fit the logic
that went into things like Rural Electrification and Telephony and even
Cable Franchises,   requiring corporate interests to service the more
difficult and remote long before (if ever) they would have based entirely on
market forces.   

I don't know "Radwin"  but what you describe is roughly what the "La Canada"
cooperative did for the South Santa Fe region (including El Dorado) some 15
or more years ago.    I attended one of their board meetings about 7 years
ago when my best (only?!) option coming from the San Ildefonso Pueblo
(established with a Grant) shut down abruptly (not long after the Grant was
over).   I was trying to gather enough information and support to possibly
form a similar non-profit for the Pojoaque Valley area which has notoriously
bad coverage.    I am looking into it again (with a little less naivete)
because the Federal Grant service known as RediNet with phat fiber to our
valley seems to finally be ready to receive a coop if we can get our act
together.

The good news is that what they are describing is generally quite doable,
the only question is whether they can get enough customers, raise the
capital for the primary infrastructure, and keep it operating long enough to
get into the black.  The same challenges I am seeing with a not-for-profit
effort here... just without any need to make more $$ than it takes to
operate viably.

The bad news is not too bad.  First,  They can't be promising 0%
oversubscription of bandwidth... that would be prohibitively expensive and
excessive for the type of utilization  a collection of (part-time?)
homeowners would need.   Their level of oversubscription (in theory and in
reality) is what will make or break your real bandwidth.   If I read you
correctly they are offering 25Mbps down and 3Mbps up?    I'm not sure what
the range of oversubscription is, but I"ve heard on the order of 4x to 32x
with the former being very rare and usually only during a startup phase
before a service actually connects enough customers to begin to consume the
bandwidth.   

My own service is commercial, using similar hardware (in our case, Ubiquiti)
and I get mine beamed a solid 23 miles from Big Tesuque.   I pay $70/month
for 10/5Mbps (which I rarely see), including a $10/month lease of my
receiving equipment (vs one-time of about $300)  I have variable service and
gaps in service based on (apparently?) wind and icing on the mountain (much
more than your towers will likely ever see) and landline cable cuts between
SFe&ABQ.   If your guys are *commercial* then ultimately they may fall into
the trap of trying to maximize profits which will ultimately marginalize
services for you.  Meanwhile I suspect they will outperform DSL (over old
copper with oxidizing connection blocks on poles or underground) and cable
(similar but also often not as pervasive as twisted pair phone lines).

I also took a whirl at prototyping a "mesh" system designed to support third
world ruralities with Voice over IP and Internet.  I was trying to set up a
network in La Puebla to serve a large handful of small farmers/horse-people
who either had no service (no phone or cable wires to their property) or
were unhappy with the quality of service they were getting and were
(superficially?) willing to participate in a cooperative experiment with
their neighbors.   The basic units are called "Mesh Potatoes" (I forget the
origin of Potato in this) and cost about $30 each.  They are fundamentally
802.11 routers with two radios for "bridging" packaged in a weatherproof
case, with a VOIP chip (so neighbors can pick up a simple phone handset and
ring one another), and can therefore also receive external longer range
(beam shaping) antennae to reach further in a given direction.

Each house would have one of these and possibly one or more identical units
acting as "repeaters" in between them and one or more other houses in the
"mesh".   They were designed to be easy to set up and cheap enough to swap
out if there were a hardware failure.  They run off of a wide range of
low-voltage, making them ideal for solar powering, but the $30 units only
come with a 110/220 wallwart requiring mains.   Their draw is low enough
though that you could probably hook one up in a vehicle and park it in a
convenient spot without drawing the battery too badly.  Most of the regions
they are aimed at don't have mains nor vehicles, so batteries
(intermittently charged or swapped) or solar are the most common.  

I DID have some technical difficulties, mostly in placing routers close
enough to both power sources (mains) and amongst the participants, but the
big thing was that in spite of being underserved, rural self-made types, I
could tell very quickly that they all "just wanted a service that works" and
weren't very cooperative when I started asking them for permission to
install a directional antenna on the side of a barn or casita to reach a
neighbor, or instructing them on how to reset the system if their neighbors
were having trouble...   They also wanted a minimum of Netflix Streaming
without any buffering 99% of the time....   In a few cases I saw what they
were wanting to replace (DSL providing Netflix Streaming with no buffering
90% of the time) and how unhappy they were and realized I was setting up
another thankless role/task for myself.   On the other end, the best way to
get an uplink (or several redundant) was from the existing providers who
were not willing to adjust their End User License Agreements to allow the
kind of sharing implied...   As awkward as it was for me trying to set all
of this up, I realized it confronted *their* service model of one
contract/bill/POC per household/family, even if half of the people I was
working with *couldn't* access their current service (lack of infrastructure
or even line of site to towers).   

My final goal was to help a friend from Kenya light up the series of
villages along the western edge of the Rift Valley where he is from...   he
is an Olympic runner who came to Northern NM to train for the 04s and
stayed... they feel lucky there if there is an elementary school within a 30
minute *run* from home... that is how he (and many Kenyans) apparently get
started running... "running to school", forget "up hill both ways in the
snow!"

I doubt that  the incumbents in your area will try to compete these guys out
of business directly... they may be happy to have them picking up the
"difficult to serve" clients.   I also think that  despite the downfalls of
a commercial service that ultimately you will likely get better service...
coops tend to depend heavily on dedicated volunteer help...  members who are
highly motivated and technical... who carry most of the weight (until they
burn out or are replaced by fresh folks).

In any case, congratulations and carry on!   All that extra EM radiation may
slow the mosquitos down for you?   Don't forget to wear your tinfoil hat!

- Steve

There is a movement afoot to bring broad band to us here in the mosquito
infested bog.  A group of locals is forming a for=profit company to bring
internet (25/3) to hundreds of subscribers in our hilly, rural town.  They
will put 4 "Radwin" transmitters atop 150 foot towers on two local hill tops
with smaller repeaters as necessary.  The transmitters look for all the
world like Mac Powerbooks.   Each house will have a waffle sized receiver.
The plan for 200 dollar initial buy-in cost and a one hundred dollar per
month subscription cost for UNLIMITED service at the advertised rate.  (No
"up to".)  I now pay about a hundred dollars a month for a Verizon jetpack
which pays for only ten gigs of data.  To stay within that limit I have to
turn off anything that moves on the internet, and go to the local library to
get podcasts, movies, or to update software, or do a cloud backup.  

In short, I am enthusiastic about the idea.  What's wrong with it?  And if
nothing is wrong with it, why haven't  all you Eldorado folks done it
already.  Go ahead.  Rain on my parade.   I asked them if they were afraid
that Verizon would get religion and put in DSL at the last moment just to
put them out of business.   Their response was that  local DSL service is so
crappy that it probably wouldn't make any difference.  They say their real
competitor is Elon Musk who is planning a vast satellite service that will
light up everyone in the universe

I gather you have all been suffering gale force winds and duststorms.  Ugh.
We, for our part, have had seven snowfalls since we got here. (All minor,
but still, relentlessly gray and chilly) The weather broke this weekend and
the garden is beginning to be populated.   I hope the equivalent break is
happening for you.  

Miss you lots, 

Nick  

Nicholas S. Thompson

Emeritus Professor of Psychology and Biology

Clark University

http://home.earthlink.net/~nickthompson/naturaldesigns/
<http://home.earthlink.net/%7Enickthompson/naturaldesigns/> 

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20180424/5e622e5e/attachment.html>