[173246] in North American Network Operators' Group
Re: Muni Fiber and Politics
daemon@ATHENA.MIT.EDU (Hugo Slabbert)
Mon Jul 21 15:45:55 2014
X-Original-To: nanog@nanog.org
Date: Mon, 21 Jul 2014 12:45:43 -0700
From: Hugo Slabbert <hugo@slabnet.com>
To: Mikael Abrahamsson <swmike@swm.pp.se>
In-Reply-To: <alpine.DEB.2.02.1407212056000.7929@uplift.swm.pp.se>
Cc: NANOG <nanog@nanog.org>
Errors-To: nanog-bounces@nanog.org
--Qxx1br4bt0+wmkIi
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
+1
A municipality nearby adopted this, and I personally like the model.
They built out their own fiber, largely for their own purposes to connect=
=20
municipal buildings and (I would assume) consolidate their internet access =
as=20
well as opposed to a bunch of discrete retail-type connections. Since thei=
r=20
laying conduit and fiber anyway, they just lay down a bigger bundle while=
=20
they're down there; bonus points for piggy-backing on existing infrastructu=
re=20
projects that already dig up the road anyway. The fiber is terminated in o=
ne=20
of two city-run DCs based on geography, and any provider can get space ther=
e=20
and pick up a pair or more to an on-net building. Pricing is very reasonab=
le=20
($400/month per pair) and the colo and power are actually free provided you=
're=20
actually paying for a pair. There's a ring between the two facilities, so =
you=20
basically just have to work out your transport to one or both facilities, d=
rop=20
in a switch or two and you're off.
New multi-tenant construction gets built out by default. If a building is =
not=20
yet on-net, submit it to the department running the dark net; if it's a=20
feasible build, the city actually foots the bill for the build-out and you=
=20
still just pay your $400/month/pair.
They intentionally structured it to only do L1; they don't want to get into=
the=20
business of running L2 or L3 services and explicitly do not want to compete=
=20
with private providers. Infrastructure and utilities are their game, and t=
he=20
city is doing it as a play to encourage competition and draw in more=20
connectivity options for residents and businesses. The figures I heard was=
=20
their their break-even is/was at the 3-year mark. Even if they don't bring=
in=20
massive revenue from providers participating, their still saving money comp=
ared=20
to their previous connectivity solutions.
So:
- level playing field & greater competition: L1 is available to anyone at a=
=20
reasonable cost, so small players can participate and differentiate on=20
anything > L1
- providers are welcome to participate or not: you want to run your own fib=
er? =20
Sure, no problem: business as usual in that department
- city doesn't compete with private business
=46rom what I gather it's targeted more at active Ethernet to multi-tenant=
=20
residential or business locations rather than being a "pass every house to=
=20
enable PON" setup, but what's not to love about this?
--
Hugo
On Mon 2014-Jul-21 20:58:48 +0200, Mikael Abrahamsson <swmike@swm.pp.se> wr=
ote:
>On Mon, 21 Jul 2014, William Herrin wrote:
>
>>The only exception I see to this would be if localities were=20
>>constrained to providing point to point and point to multipoint=20
>>communications infrastructure within the locality on a reasonable=20
>>and non-discriminatory basis. The competition that would foster on=20
>>the services side might outweigh the damage on the infrastructure=20
>>side. Like public roads facilitate efficient transportation and=20
>>freight despite the cost and potholes, though that's an imperfect=20
>>simile.
>
>While I might not agree with the parts of your email you cut out, I=20
>would definitely like to chime in on this part. Muni fiber should be=20
>exactly that, muni *fiber*. Point to point fiber optic single mode=20
>fiber cabling, aggregating thousands of households per location,=20
>preferrably tens of thousands.
>
>It's hard to go wrong in this area, it either works or it doesn't, and=20
>in these aggregation nodes people can compete with several different=20
>technologies, they can use PON, they can use active ethernet, they can=20
>provide corporate 10GE connections if they need to, they can run=20
>hybrid/fiber coax, they can run point-to-point 1GE for residential.=20
>Anything is possible and the infrastructure is likely to be as viable=20
>in 30 years as it is day 1 after installation.
>
>--=20
>Mikael Abrahamsson email: swmike@swm.pp.se
--Qxx1br4bt0+wmkIi
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQIcBAEBCgAGBQJTzW3nAAoJEJqxD/2xeDE+EesP/RF4e+8UiFPDconL7Qm4dN4c
ml/3RRbHzrPPD1V9RHOWGe1RqIgMb/zTDYX+FCqtNZUsa+a4d2qrhp2zDvLnFTT6
Jms4YWAE1sy8uRiWJ1o7xYem8on8uBD1QBwawgNVV74zuwMkBWdFsIG37FS78CxQ
wcmtnE12wHhLP28v58WH3u8nbkfRey6oy3WO6ieQKb0ijVOflDX2oDN+0UZg88gS
CvPdSv53jOf7Hlc2ILslWhbKGyEIN1RbYqea63PFVEmttcJYLPISLlGRkA62WLqc
6XI4kCEcM/mYcRIxmKVlxlMWT4CHdjIQdsxnwIvlGFEH51OD/2C2p3XMVcYfZvD+
FitWJ4Ft9sJobtsUNzOlKOb3aCCyrxzRz/Ians1D1U3OEBfi0ZTw0w2KJnuM5FBp
zCWiwvfOUJcnbsZiftL90n9delrVf/COYawlaKmfWzKI5TRrx10JcSxY46YWql87
U6DyzfgA7p62j1Fok8t2oEgsT8eWDEVZ4ZqlCld0WxsjUGw5HUz9rns8jNHXd70k
dxdYsESvQcvHWRCFF0dYub2lmGTL9AwmfB/r+4EJYahClDdkDt264MYTBn+90fGL
ZK3GedSScCmmNsueQy20qfIc6bP8XIfT68EwStiSsuKSNcoSMtxmR8G/6Sdmp+j2
ZXgAO1xb2i8JugmHwMOq
=63jE
-----END PGP SIGNATURE-----
--Qxx1br4bt0+wmkIi--
