[141684] in North American Network Operators' Group
Re: Yup; the Internet is screwed up.
daemon@ATHENA.MIT.EDU (Jared Mauch)
Fri Jun 10 10:18:08 2011
From: Jared Mauch <jared@puck.nether.net>
In-Reply-To: <BANLkTimkb6mpwoGidijogb_Ekyf_+LqTAw@mail.gmail.com>
Date: Fri, 10 Jun 2011 10:17:32 -0400
To: Kyle Creyts <kyle.creyts@gmail.com>
Cc: NANOG <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
On Jun 10, 2011, at 10:01 AM, Kyle Creyts wrote:
> I think the point is the ubiquity of access isn't what it should be.
I think there were several good points made in the article.
1) Data caps and how they impact software updates (or downloads) - =
hughesnet was mentioned but ..
Looking to the near future, Apple is selling a 4GB download for $30 in =
the next month or so. That will have a large impact on networks that =
day IMHO. If you have a 3G/4G/LTE/whatnot device it makes it impossible =
to pull down the image without hitting your 5GB or 10GB cap compared to =
a fixed access network.
Even assuming you go to the local Panera/McDonalds/Starbucks/Library =
access, if you get 2MB/s (16Mb/s) you're talking about 20-25 minutes. =
Those locales don't usually have that fast of a network though.
2) Last mile is expensive to install and hard to justify for people. =
This is because of a long history of universal service and =
subsidization/regulation.
In Michigan you could get a phone line installed for $42 (not sure, =
haven't installed POTS in a long time, may have gone up) regardless of =
the cost to the carrier. This isn't the case when you want to extend =
other utilities (eg Gas, electric, water...). People are willing to pay =
10k+ to install these services as part of their construction expense. =
Their other utility cost is masked in part due to the past 100+ years of =
telecom history. The cost of lighting a 20km strand of fiber at 1Gb/s =
is somewhere in the $600, including ONT, etc. Many people here on nanog =
would happily pay that amount. Now, the 12-100k per mile to build the =
fiber is the hard part to eat.
3) Certainly he did a poor job of site selection. Perhaps he was misled =
or even lied to. I've faced similar challenges when working with both =
hardware vendors and carriers out there. The sales peoples eyes get big =
once you start talking about "doing" something, but the engineers at the =
table generally start asking serious questions. (I certainly will not =
move anywhere that doesn't have a HFC or PON/FTTH network. Sorry =
AT&T/Centurylink/others but the plusses don't justify the minuses).
-
It's certainly possible that we will see improved last-mile access. The =
USDA/RUS and DOC/NTIA efforts are to be applauded. If you look at the =
current AT&T + T-Mobile merger people are talking about it will bring =
broadband to 97% of the country, and help AT&T (mobility division) with =
last-mile/local tower regulatory hurdles. They are not talking about =
how it will remove the need for data caps that are 1/30th the size of =
their 150GB cap on their mobile side elements.
I suspect there's a lot that could be improved by each market player =
here, but as happened with Verizon in the Northeast, I expect the =
less-dense markets will need to have better local service from regional =
players vs the "big guys". Overall this will be good, but the costs =
will also have to be paid for more with the local subscriber.
- Jared=
