I read an article on DSL Reports the other day (
http://www.dslreports.com/shownews/FCC-Please-Define-Broadband-104056), in
which the FCC has a document requesting feedback on the definition of
Broadband.
What are your thoughts on what the definition of Broadband should be going
forward? I would assume this will be the standard definition for a number of
years to come.
Thanks.
It's not a technical question, it's a political one, so feel free to
squelch this for off-topicness if you want.
Technically, broadband is "faster than narrowband", and beyond that
it's "fast enough for what you're trying to sell"; tell me what you're
trying to sell and I'll tell you how fast a connection you need.
<ul><li>If you're trying to sell email, VOIP, and lightly-graphical
web browsing, 64kbps is enough, and 128 is better. <li>If you're
trying to sell wireless data excluding laptop tethering, that's also
fast enough for anything except maybe uploading hi-res camera video.
<li>If you're trying to sell talking-heads video conferencing, 128's
enough but 384's better. <li>If you're trying to sell internet radio,
somewhere around 300 is probably enough. <li>If you're trying to sell
online gaming, you'll need to find a WoW addict; I gather latency's a
bit more of an issue than bandwidth for most people. <li>If you're
trying to sell home web servers - oh wait, they're not! - 100-300k's
usually enough, unless you get slashdotted, in which case you need
50-100Mbps for a couple of hours. <li>If you're trying to sell
Youtube-quality video, 1 Mbps is enough, 3 Mbps is better. <li>If
you're trying to sell television replacement, 10M's about enough for
one HD channel, 20's better, but the real question is what kind of
multicast upstream infrastructure you're using to manage the number of
channels you're selling, and whether you're price-competitive with
cable, satellite, or radio broadcast, and how well you get along with
your city and state regulators who'd like a piece of the action.
<li> If what you're trying to sell is "the relevance of the FCC to the
Democratic political machines", the answer is measured in TV-hours,
newspaper-inches, and letters to Congresscritters, which isn't my
problem. </ul>
Historically, narrowband was circuit switched (ISDN etc) and broadband was packet switched. Narrowband was therefore tied to the digital signaling hierarchy and was in some way a multiple of 64 KBPS. As the term was used then, broadband delivery options of course included virtual circuits bearing packets, like Frame Relay and ATM.
The new services I am hearing about include streamed video to multiple HD TVs in the home. I think I would encourage the FCC to discuss "broadband" to step away from the technology and look at the bandwidth usably delivered (as in "I don't care what the bit rate of the connection at the curb is if the back end is clogged; how much can a commodity TCP session move through the network"). http://tinyurl.com/pgxqzb suggests that the average broadband service worldwide delivers a download rate of 1.5 MBPS; having the FCC assert that the new definition of broadband is that it delivers a usable data rate in excess of 1 MBPS while narrowband delivers less seems reasonable. That said, the US is ~15th worldwide in broadband speed; Belgium, Ireland, South Korea, Taiwan, and the UK seem to think that FTTH that can serve multiple HDTVs simultaneously is normal.
Fred Baker wrote:
What are your thoughts on what the definition of Broadband should be going
forward? I would assume this will be the standard definition for a number of
years to come.Historically, narrowband was circuit switched (ISDN etc) and broadband was packet switched. Narrowband was therefore tied to the digital signaling hierarchy and was in some way a multiple of 64 KBPS. As the term was used then, broadband delivery options of course included virtual circuits bearing packets, like Frame Relay and ATM.
of or relating to or being a communications network in which the bandwidth can be divided and shared by multiple simultaneous signals (as for voice or data or video)
That's my humble opinion. Let them use a new term, like "High Speed Internet".
Paul Timmins wrote:
Fred Baker wrote:
What are your thoughts on what the definition of Broadband should be going
forward? I would assume this will be the standard definition for a number of
years to come.Historically, narrowband was circuit switched (ISDN etc) and broadband was packet switched. Narrowband was therefore tied to the digital signaling hierarchy and was in some way a multiple of 64 KBPS. As the term was used then, broadband delivery options of course included virtual circuits bearing packets, like Frame Relay and ATM.
of or relating to or being a communications network in which the bandwidth can be divided and shared by multiple simultaneous signals (as for voice or data or video)
That's my humble opinion. Let them use a new term, like "High Speed Internet".
Seconded
In the applications I wrote earlier this month for BIP (Rural Utilities Services, USDA) and BTOP (NTIA, non-rural) infrastructure, for Maine's 2nd, I was keenly aware that broadband hasn't taken off as a pervasive, if not universal service in rural areas of the US.
I don't think the speed metric is the metric that will make non-adoption in sparce clustered demographics distinguishable from adoption in denser demographics. I suspect that issues like symmetry of state signaling, latency, jitter, ... metrics that resemble what I looked for from MPI runs when benchmarking parallel systems, will characterize applications that may be distinguishable from the adoption, market penetration, renewal criteria from the applications that for reasons I can only conjecture, the standard "triple play" killer apps, which simply aren't driving broadband (whatever that is) adoption in rural areas. And no, I don't know what those better-than-triple-play-killer-apps-in-suburbia are.
Eric
Luke Marrott wrote:
In the applications I wrote earlier this month for BIP (Rural Utilities Services, USDA) and BTOP (NTIA, non-rural) infrastructure, for Maine's 2nd, I was keenly aware that broadband hasn't taken off as a pervasive, if not universal service in rural areas of the US.
I don't think the speed metric is the metric that will make non-adoption in sparce clustered demographics distinguishable from adoption in denser demographics. I suspect that issues like symmetry of state signaling, latency, jitter, ... metrics that resemble what I looked for from MPI runs when benchmarking parallel systems, will characterize applications that may be distinguishable from the adoption, market penetration, renewal criteria from the applications that for reasons I can only conjecture, the standard "triple play" killer apps, which simply aren't driving broadband (whatever that is) adoption in rural areas. And no, I don't know what those better-than-triple-play-killer-apps-in-suburbia are.
My meta-point is that I suspect there are two "broadbands", one where triple-play sells recurring subscriber drops, and one where it doesn't, and for the later a better definition would be more useful than a definition that reads (in fine print) "not available here".
Eric
Luke Marrott wrote:
I read an article on DSL Reports the other day (
FCC: Please Define Broadband - Baby steps. Baby steps. | DSLReports, ISP Information), in
which the FCC has a document requesting feedback on the definition of
Broadband.What are your thoughts on what the definition of Broadband should be going
forward? I would assume this will be the standard definition for a number of
years to come.Thanks.
From MAILER-DAEMON Wed Aug 26 17:25:51 2009
Received: from outbound-mail-30.bluehost.com ([69.89.17.212])
by s0.nanog.org with smtp (Exim 4.68 (FreeBSD)) id 1MgMFr-0001kJ-FV
for nanog@nanog.org; Wed, 26 Aug 2009 17:25:51 +0000
Received: (qmail 8313 invoked by uid 0); 26 Aug 2009 17:25:50 -0000
Received: from unknown (HELO host308.hostmonster.com) (74.220.215.108)
by outboundproxy2.bluehost.com with SMTP; 26 Aug 2009 17:25:50 -0000
Received: from c-24-5-230-26.hsd1.ca.comcast.net ([24.5.230.26] helo=Honkin)
by host308.hostmonster.com with esmtpsa (TLSv1:AES128-SHA:128)
(Exim 4.69) (envelope-from <richard@bennett.com>)
id 1MgMFq-0001OA-7a; Wed, 26 Aug 2009 11:25:50 -0600
Fred,
Historically there was no such thing as a "narrowband" Internet
connection. We used "bandwidth" as slang for the speed of an Internet
connection, possibly because in communications in general you can send
more information in a wide frequency band than you can in a narrow
frequency band and we knew that a phone line used a 4khz frequency
band while a T1 used a 1.5mhz frequency band.
When we started selling residential Internet connections that were
significantly faster than a modem (i.e. DSL, cable modems) some
marketing guru somewhere came up with the idea that if Internet speed
is bandwidth then fast internet must be -broad- bandwidth. The same
marketing gurus wouldn't be particularly guruish if they had then
started referring to their modem products as "narrowband." So the
choice was "dialup or broadband" not "narrowband or broadband."
As the term caught on, it was the expanded by various marketing and
salesfolk to encompass any kind of commodity Internet connection
(commodity = not custom, that is not doing anything uncommon like
dynamic routing or multiplexing) which was better than a dialup modem.
When you start assigning CIDR blocks and what not, that's generally a
business service rather than "broadband."
So historically speaking, broadband is anything faster than POTS dialup.
What it -should- mean for stimulus purposes is another matter... But
I'd personally prefer to see the stimulus money only used for
delivering rural high speed. The telcos and cable companies are in a
race to deliver fast residential Internet access in any densely packed
area where the governing authority isn't making it a costly PIA to
install. Where they need the swift kick in the tail is in the low
density areas.
Really where they need the swift kick in the tail is in the product
tying where you can't buy a high speed connection to J. Random ISP,
you can only buy a high speed connection to monopoly provider's
in-house ISP. Which means you can only get commodity service since
monopoly provider isn't in the business of providing low-dollar custom
solutions. But it sounds like that's outside the scope of what
Congress has approved.
Regards,
Bill Herrin
As tedious as the downstream can be, engineering the upstream path of a cable plant is worse.
A lot of older systems were never designed for upstream service. Even if the amps are retrofitted, the plant is just not tight enough.
Desirably, fiber should be pushed deeper; the quantity of cascaded amps reduced, coax fittings and splitters replaced and so on.
In a message written on Mon, Aug 24, 2009 at 10:17:02AM -0600, Luke Marrott wrote:
I read an article on DSL Reports the other day (
FCC: Please Define Broadband - Baby steps. Baby steps. | DSLReports, ISP Information), in
which the FCC has a document requesting feedback on the definition of
Broadband.What are your thoughts on what the definition of Broadband should be going
forward? I would assume this will be the standard definition for a number of
years to come.
I'm not sure the defintion of Broadband matters, what matters is that we
keep moving forward.
We should set a goal of all American's having a speed twice as fast in 5
years. And after that twice as fast again in another 5 years.
There is no bar we reach where we are "done".
An interesting question: as the population gets sparser, the average trench
mileage per subscriber increases. At some point this renders fibre deployment
uneconomic. Now, this point can change:
1) as we deploy fibre we'll get more efficient at it - I think VZ's cost per sub
has come down quite a lot since they started the FIOS rollout.
2) the flip side of the cost to serve a subscriber is of course revenue, and if
you can find other services to sell'em you can go further. may also be scope
for tiered pricing
3) public sector investment
Going the other way, as the population gets denser, it becomes harder to
provide an acceptable broadband wireless service because of spectrum
limitations. You either need more and more cells (=more and more sites and
more and more backhaul), or more and more spectrum.
Where's the crossover point? There are clearly places where some fibre
investment (like L(3)'s proposed deployment of many more POPs) would make it
possible to get good service out using radio from the end of the fibre,
precisely because they are sparse. There are clearly places where fibre to the
home will eventually arrive.
Is there a broadband gap between the two groups, however, where it's not dense
enough to ever deploy fibre and too dense to deploy good wireless? Or can we
rely on FTTH for one lot and RTTR (Radio to the Ranch) for the other?
In a message written on Thu, Aug 27, 2009 at 09:58:22AM +0100, Alexander Harrowell wrote:
An interesting question: as the population gets sparser, the average trench
mileage per subscriber increases. At some point this renders fibre deployment
uneconomic. Now, this point can change:
This statement makes no sense to me.
The cost to dig a trench is cheaper in rural areas than it is in
urban areas. A lot cheaper. Rather than closing a road, cutting
a trench, avoiding 900 other obsticals, repaving, etc they can often
trench or go aerial down the side of a road for miles with no
obsticals and nothing but grass to put back.
So while mileage per subscriber increases, cost per mile dramatically
increases. The only advantage in an urban enviornment is that one
trench may serve 200 families in a building, where as a rural trench
may serve 20 familes.
But more puzzling to me is the idea that fiber becomes uneconomic.
This may have once been true, but right now you can buy 10km or
even 40km lasers quite cheaply. Compare with copper which for even
modest speeds requires a repeater every 2-4km.
If you have to reach someone 20km from the CO, the cost of running
the ditch-wich down the road in a rural area is not the dominate
cost over the next 20 years. It's equipment. If the copper plant
takes 4 repeaters to do the job, that's 4 bits of equipment that
can fail, and will need to be upgraded at some point. Running
something as simple as point to point fiber they can be provided
with GigE speeds today with no intermediate equipment; the cost of
a 20km GBIC is far less than the cost of installing 4 repeaters.
The problem with all of these is ROI, not cost. Somewhere along the
line we've decided very short ROI's are required. Do you work at a
company where an ROI of over a year is laughed at? When the original
rural telephone network was pushed ROI's of 50 years were talked about.
There's plenty of infrastructure built every day with ROI's of 20 years.
So it would cost $2000 per home to put in fiber. The margin on the
service is $5 per month. It's a 33 year ROI. That's ok with me, it's
infrastructure, like a road, or a bridge. We're still using copper in
the ground put in during the 60's, 70's, and 80's.
In a message written on Thu, Aug 27, 2009 at 09:58:22AM +0100, Alexander Harrowell wrote:
An interesting question: as the population gets sparser, the average trench
mileage per subscriber increases. At some point this renders fibre deployment
uneconomic. Now, this point can change:This statement makes no sense to me.
The cost to dig a trench is cheaper in rural areas than it is in
urban areas. A lot cheaper. Rather than closing a road, cutting
a trench, avoiding 900 other obsticals, repaving, etc they can often
trench or go aerial down the side of a road for miles with no
obsticals and nothing but grass to put back.So while mileage per subscriber increases, cost per mile dramatically
increases.
I think you meant, "decreases", here.
Marshall
Leo Bicknell wrote:
If you have to reach someone 20km from the CO, the cost of running
the ditch-wich down the road in a rural area is not the dominate
cost over the next 20 years. It's equipment. If the copper plant
takes 4 repeaters to do the job, that's 4 bits of equipment that
can fail, and will need to be upgraded at some point. Running
something as simple as point to point fiber they can be provided
with GigE speeds today with no intermediate equipment; the cost of
a 20km GBIC is far less than the cost of installing 4 repeaters.The problem with all of these is ROI, not cost. Somewhere along the
line we've decided very short ROI's are required. Do you work at a
company where an ROI of over a year is laughed at? When the original
rural telephone network was pushed ROI's of 50 years were talked about.
There's plenty of infrastructure built every day with ROI's of 20 years.So it would cost $2000 per home to put in fiber. The margin on the
service is $5 per month. It's a 33 year ROI. That's ok with me, it's
infrastructure, like a road, or a bridge. We're still using copper in
the ground put in during the 60's, 70's, and 80's
Seems like a good idea to the technical side of me, but the business side sees a problem: that the employees like to eat in the 33 year span wherein the company isn't making a dime on its customers.
-Paul
Once upon a time, Leo Bicknell <bicknell@ufp.org> said:
When the original
rural telephone network was pushed ROI's of 50 years were talked about.
There's plenty of infrastructure built every day with ROI's of 20 years.
How much of that was built in the last 15 years though (where now it
needs to be replaced before it has been paid for)? In the 1990s,
BellSouth pushed hard here, rolled out fiber to the neighborhoods, and
deployed ISDN-capable equipment everywhere. ISDN was available at every
single address in town by around 1995 (allegedly we were one of if not
the first moderate-sized city with ISDN everywhere).
Then it turned out ISDN was a flop, and DSL came along, which wouldn't
run over that nice big fiber plant. They had to start rolling out
remote DSLAMs all over town. Shortly after they had most of the city
covered, ADSL2 came along, and they had to start upgrading again.
Granted, the cable plant (whether copper, fiber, coax, or avian
datagram) is not quite the same, but the bean-counters look at it as "we
were supposed to have <bignum>-year ROI on project 1, 2, and 3, and we
didn't get it; why should I believe we'll get it on project 4?".
Leo Bicknell wrote:
So while mileage per subscriber increases, cost per mile dramatically
increases. The only advantage in an urban enviornment is that one
trench may serve 200 families in a building, where as a rural trench
may serve 20 familes.
Cost per subscriber is the only cost that matters. That is what defines your recoup time and profit margins. BTW, in many cases it's actually cheaper to bore the entire way then intermix boring and trenching. And out here, they are heavily against you trenching right through someone's driveway or a road. Then there's the rivers and creeks.
But more puzzling to me is the idea that fiber becomes uneconomic.
This may have once been true, but right now you can buy 10km or
even 40km lasers quite cheaply. Compare with copper which for even
modest speeds requires a repeater every 2-4km.
Maintenance. The reason rural companies prefer active equipment in the plant is because of maintenance. 20 splices to restore service to 20 customers vs 1 splice to restore service to 20 customers. This is oversimplified, in reality, many of the FTTH comments in this thread imply bringing all customers back to the CO to keep active equipment out of the plant. This will tend to imply large fiber bundles leaving the CO and breaking down smaller and smaller as you get further from the CO. A large fiber cut may mean 128+ splices to restore service at 1 splice per customer.
In addition, it throws away all the money and investment of plant already in the ground from key points to the customers. I haven't seen an installation running repeaters for copper. More common is a remote system fed by a fiber ring (so when the 20km fiber is cut, service isn't lost while repairs are done) and the last 1.5 miles fed by copper which is already there.
with GigE speeds today with no intermediate equipment; the cost of
a 20km GBIC is far less than the cost of installing 4 repeaters.
If someone is setting up like this, I'd agree. More common:
Traditional POTS was often served off double ended carrier and load coils, which later became fiber fed integrated carrier with gr303 and load coils. Cheapest solution, replace carrier with DSL capable carrier, remove load coils when not necessary and extend from there for closer carriers where applicable (shorten copper loops, and removal of more load coils).
Here locally, we dropped over 90% of our load peds. Only the furthest reaches still have them and of course cannot get DSL.
There's plenty of infrastructure built every day with ROI's of 20 years.
Hope they have disaster insurance. A good tornado or wildfire (or backhoe) can do some serious damage. I had both this year in Lone Grove. Fun. Fun. Fiber rings to remote field equipment still gives the best redundancy and maintenance cost (as there is less to splice over the longhaul to the remote system).
service is $5 per month. It's a 33 year ROI. That's ok with me, it's
infrastructure, like a road, or a bridge. We're still using copper in
the ground put in during the 60's, 70's, and 80's.
You bet. We're also using fiber and copper put in the ground yesterday. Copper is amazingly resilient. Most of the copper that has to be replaced is old aircore in the ground (which is why aircore shouldn't be in the ground, as it collects water and leads to shorts over long distances) or rehab of aircore in aerial due to bad boots that weren't maintained. The switch to fiber fed remote systems abandoned most of the problematic copper, though.
Jack
In a message written on Thu, Aug 27, 2009 at 10:47:01AM -0400, Paul Timmins wrote:
Seems like a good idea to the technical side of me, but the business
side sees a problem: that the employees like to eat in the 33 year span
wherein the company isn't making a dime on its customers.
The last letter of ROI is Investment. When Ford decides to build
a new car it sinks in billions of dollars over a 5 year period where
it makes nothing. It then starts selling the new model, and finally
reaches a point where it makes a profit, and uses that to find the
next Investment.
What Telecom companies have done is confused infrastructure and
equipment. It would be stupid to plan on making a profit on your
GSR over 30 years, after 10 it will be functionally obsolete. When
it comes to equipment the idea of 1-3 year ROI's makes sense.
However, when it comes to fiber or copper in the ground or on a
pole it has a 20, 30, 40, or even 50 year life span. To require
those assets to have a 1-3 year ROI is absurd.
I remember when Cable TV was "new". I lived in a neighborhood
without it, and the men with ditch wiches came through and wired
the entire neighborhood. I don't think it had an ROI of a year,
or even 5, but it has now, 30 years later, spawned a multi-billion
dollar industry and allowed us to have things like Cable Internet,
which weren't even invented at the time. Someone loaned them the money
to do it, and it appears to me the investment performed well, overall.
I don't think ISDN was a flop.
In the middle of the 90 years. The most KMU and bigger Companies have ISDN. At home it was at 1997 a trend two with Internet. Ok in Europe we haven't till begin of the 2000 no Clip Informations on a analoge line. This will be come to begin of the 2000.
With ADSL and Clipinformations has the most people at home chanched back to an analog Line. For the companies is ISDN allready a must.
You must see. At the End of the last century the most people has a phone, has a fax, has a Modem... The best way was ISDN.
Now The childern are skypeing... or take an other IP Fon. Fax doesn't exist at home. The people has E-Mail. And Internet we have on ADSL or VDSL. with many speed. For phoneing 2/3 of the people take the handy (celuar phone) or IP fon.
I think this is the bigest part in the last 10 years.
Greetings
Xaver
In a message written on Thu, Aug 27, 2009 at 09:57:56AM -0500, Jack Bates wrote:
oversimplified, in reality, many of the FTTH comments in this thread
imply bringing all customers back to the CO to keep active equipment out
of the plant. This will tend to imply large fiber bundles leaving the CO
and breaking down smaller and smaller as you get further from the CO. A
large fiber cut may mean 128+ splices to restore service at 1 splice per
customer.
The interesting technology here of course is split optical networks.
A single fiber from the CO to a remote splice box, split to 10-100
customers. I'm not really up on this technology, but my understanding
is that development is rapid in this space.
Hope they have disaster insurance. A good tornado or wildfire (or
backhoe) can do some serious damage. I had both this year in Lone Grove.
Fun. Fun. Fiber rings to remote field equipment still gives the best
redundancy and maintenance cost (as there is less to splice over the
longhaul to the remote system).
I hate to say it, but this was an advantage to "Ma Bell". Insurance
is about spreading risk out over many participants. An alternative
strategy is to pool everything into one company! 
My perception is that the rural telecom market is fragmented by many
smaller players, which amplifies this problem.
Leo Bicknell wrote:
My perception is that the rural telecom market is fragmented by many
smaller players, which amplifies this problem.
I have 12 ILEC and 1 CLEC under my umbrella. I can guarantee that not a single one is the same at the plant, equipment, or business level.
That being said, I think we are luckier than Bell, who has only a few dozen concentrated CO's in the state which feed the smaller CO's, or in some cases, entire towns are fed by double ended carrier (where 14.4 is considered the best connection one can hope for). We see unlicensed wireless in a lot of these places, but their customers honestly beg for better service.
Jack