[LibreQoS] Enabling a production model
dandenson at gmail.com
Wed Mar 29 10:54:41 EDT 2023
> > Always a mistake to generalize from a sample of one, but in my case I
> have four, because I live in four places. So I like to think that, to some
> degree, I represent a kind of market demand.
> > All those places—Santa Barbara (CA), New York (NY), Bloomington (IN),
> and San Marino (CA)—are served by cable monopolies (Cox, Spectrum,
> Comcast/Xfinity) that provide (or at least claim) 1 Gb service...
> downstream of course. One (Cox) provides 36 Mb of upstream capacity. The
> other two provide just 10 Mb. Because of that, residents have no option to
> do much work, or to store large amounts of data, in clouds (to mention just
> one grace of upstream capacity). The market is rigged for consumption, not
> production, on the TV model. Same as it has been since commercial activity
> began to explode in 1995, when John Perry Barlow wrote Death From Above.
> It's killer. Please read it:
> I have been citing that piece left and right lately.
The problem is that this 'FiWi' model or the municipal backhaul model
FORCES this model. The reason you are stuck with those providers is
because there is a monopoly designed into the system. Without competition,
10Mbps is good enough. There is no way for consumers to 'vote' with their
money because they can't pick another product or provider.
> I think the smartest thing any city can do to start with, is to
> establish a good ole-fashioned internet exchange point there, require
> those providing service in the city to interconnect,
> > See what you think.
> > For me, the promise of fiber is a huge attraction to living and working
> here. And I am not alone.
This makes the municipality the internet provider. Even if you get to pick
who does the upstream on the bits, it's ultimately the muni to repair the
lines, handle the CPE, and handle the switching infrastructure in the
exchange. So an ISP run by a city council? a council who got elected to
'Karen' away about how cell towers give them 5G poisoning? Disaster.
Take any city listed about and look at the water and waste facilities. The
pockets of the city that are not served or are poorly served. The
Flint Michigans with one source of water that is contaminated. how those
services just stop, homes beyond are on septic tanks and hauled water.
When you've destroyed all the ISPs, whos going to bring services to those
beyond the core? The county? not sure if you've ever dealt with county
This entirely removes all choice. The entire job of the ISP is the last
mile, there is no point in selling bits to individual users at the
exchange. Take that away and the city itself is necessarily the ISP. The
'exchange' model is fundamentally flawed because there's no money in it.
The city is going to have to raise taxes or charge for the last mile at the
same rates as the ISPs do, except more because government inefficient and
inflexible. The upstream connectivity is the simplest and cheapest part
of being an ISP.
The solution to having monopolies control internet service isn't to create
a different monopoly to control internet service.
The obvious solution is to foster competition. Anywhere you overlay cable
companies with fiber BOTH companies remain and compete against each other
and the cable company increases upload speeds. If fiber was so naturally
superior, the cable companies would be erased. I have MSP customers in
multiple markets with competing techs and it's VERY nice to be able to get
fiber and cable or terragraph and cable to a business for resilience. I
cannot do that on single product dominated markets. The 'exchange' model
above doesn't do it because of that single point of failure of the
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the LibreQoS