Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

Dark fiber everywhere but not a last-mile solution to be found

By hillct in Op-Ed
Mon Sep 03, 2001 at 06:49:43 PM EST
Tags: Internet (all tags)

It's time once again to re-evaluate the last-mile problem - the problem of getting high bandwidth connectivity that last mile to the home. It's not going to be solved by large telecom equipment makers. It's going to be solved by some innovative start-up that may or may not already exist. Let's take a closer look at why this is the most probable course of events.

The last mile problem is that issue of getting broadband connectivity into the home. It has been solved thus far by DSL and cable modems, but neither of these technologies seem to be getting the job done for the majority of Americans.

The other side of this issue is that over the past two years, investors have shelled out $35 billion to finance the running of maybe 100 million miles of fiber all across the United States. Most of this fiber remains dark - it's capacity unused, because of the limits of the last-mile solutions developed and deployed thus far. The problem is, due to the collapse of the tech bubble in the stock market, there is little or no money left (that anyone wants to risk in the telecom market) to finance development of the next generation last-mile solution.

DSL is limited by several factors - some technical and some economic. Customers must be within a couple miles of the CO (telco central office) and must have copper lines - not fiber - covering the full distance from the CO to the customer. This is a particularly substantial problem in rural developments built between 5 and 15 years ago, where many telcos chose to run fiber to the curb and then split out lines to customers in the developments, starting right outside the development. This has the result of making it impossible for any of these customers to receive DSL service, without a substantial equipment upgrade by the telco.

The economic problems of DSL are many. First, the telco must replace the line card (the hardware at the other end of your phone line) for each DSL customer, and - in the case of the developments mentioned above - replace the DSLaM (Digital Subscriber Line Multiplexor) outside the developments, with one which can split off the DSL signal appropriately. These are expensive pieces of equipment for which the telco is footing the bill, and in which they'd rather not be investing, since DSL is a technology that runs along copper lines, making it unusable across all the fiber that has been laid with a view toward the future - giving DSL a planned obsolescence, and making it a vary poor investment.

Regardless of this, there were many entrants into the DSL provider market, and most non-telcos have gone out of business. The body count is steadily increasing but at the moment is includes Covad Communications Group, NorthPoint Communications, and Rhythms NetConnections. These companies were positioned to benefit from supposedly open loops which federal regulations required local telcos to provide them, but at the same time, these local telcos were competing with them, providing their own DSL services which resulted in nightmarish customer service and poor pricing for consumers, as well as vary high infrastructure costs and low margins for the providers.

That explains the demise of the independent DSL providers, but let's look now at telco-provided DSL service. Telcos have two high infrastructure cost categories. First, they're paying for some local runs of the afore mentioned dark fiber, and they're paying for - the completely incompatible - DSL hardware, which is replacing each customer's line card. After paying this cost, now they have to pay for additional call center technical support staff and have to be able to field perhaps twice as many technical support calls. all for a measly $50/customer/month on average. It just isn't economical, so many local telcos were staying away from the DSL market as long as they could.

This is why Cable Modem service has seen the popularity that is has. The primary problem with this service is again the costs of infrastructure and support - but since they have a much broader potential audience than DSL which had range and other technical limitations. Still, Cable Modem service providers like Excite@Home are struggling.

Who, then should we look to in order to solve the last-mile problem? The largest telco equipment makers and their customers have invested billions in the fiber infrastructure needed for the next logical evolutionary step in broadband access, but they did so without a next generation last-mile solution, and now they have limited funds to finance the development of that solution. It's a catch-22. Without the last-mile solution these companies will have lost billions of dollars, but since they've spent that money, they have no means to develop it. This leaves an opening for a start-up like Actelis Networks to come in and solve the problems the large telcos created by moving forward without an end to end solution. Actelis claims to have a technology that vastly increases the performance of copper-carrier broadband but this doesn't address the real issue of how best to move beyond copper. It does extend the lifespan of our existing copper last-mile infrastructure, but it doesn't address the problems of those customers with a split copper/optical loop.

There are two other last mile technologies we haven't examined here so far: Bidirectional Satellite service like Star Band and RF technologies such as 802.11b or the older but still usable packet radio (used on ham radio frequencies).

One of the important points that should be taken from this is that no-one has suggested running fiber for the last mile. At some point either fiber or a satellite solution or some combination of both will most likely supplant any technology we make use of in the interim.

So, what will the new last-mile solution look like? Who's to say... The big telecom equipment makers aren't in a position to develop it - perhaps a new player will burst into the marketplace with an entirely original solution; then again perhaps not.


Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure


What will the dominant last mile solution be?
o Copper carrier - some variant of DSL 19%
o Optical Carrier 36%
o Non telco carrier - like Cable Modem 14%
o RF technology like 802.11 28%
o Satelite Technology like Starband 1%

Votes: 57
Results | Other Polls

Related Links
o 100 million miles of fiber all across the United States
o Actelis Networks
o copper-car rier broadband
o Star Band
o Also by hillct

Display: Sort:
Dark fiber everywhere but not a last-mile solution to be found | 30 comments (29 topical, 1 editorial, 0 hidden)
on the way (4.28 / 7) (#1)
by Sikpup on Mon Sep 03, 2001 at 02:06:42 PM EST

Several major markets will soon have choice in the cable tv market. One example is WinFirst (www.winfirst.com) is building fiber to the curb, to provide cable, internet, and voice communications in a bundled package.

In San Jose, AT&T has been tearing up the streets and laying fiber, because there will be competition here in the near future. The thought of getting rid of both pac bell and att in one move should delight much of silicon valley.

Rural developments (3.00 / 6) (#3)
by marx on Mon Sep 03, 2001 at 02:47:46 PM EST

To me, the fiber-to-the-curb problem appears to be the only one which does not have an immediate solution. How many people does this involve though? I don't know the demographics of the US very well, but normally a very large percentage of the population lives in urban areas. If we're only talking about a few percent, I don't see how this is a big problem. I guess these people will have to find other solutions which are a bit more expensive. Two-way satellite for example is approaching affordability for normal people.

Still, while this problem is a bit interesting, it is not very important. The information infrastructure problem in Africa is much more important (and interesting).

In Africa, 70% of the population lives in rural areas (see here for more data). Something like 1% of the population in Africa has a phone. Here you're not talking about a fiber-to-the-curb problem. How are you going to distribute phone access to an almost entirely rural population over an entire continent? Do you think the phone companies are going to have any return on the investment of their expensive equipment in this case?

It is very unfortunate, but as we can see in this case, the mechanism of capitalism places much greater importance on solving the problem of getting broadband access to the rich people in the rural developments in the US than phone access to the general population of Africa.

Join me in the War on Torture: help eradicate torture from the world by holding torturers accountable.

Data to the rescue...(2) (4.66 / 3) (#7)
by scriptkiddie on Mon Sep 03, 2001 at 07:35:35 PM EST

According to the U.S. Census Bureau, 25% of Americans live in towns of less than 2,500 people. So while the U.S. isn't like Africa, with very few people living in close proximity, the population is broadly distributed, unlike Europe.

Remember, though, that more than 99% of U.S. citizens have phone service, and even the most isolated areas have electricity. So it's not like building a whole new infrastructure grid would be unprecedented. P.S.: Sorry about last post, this crazy web browser spontaneously submitted the form....

[ Parent ]

Dark horse in the race: the power utilities (4.33 / 9) (#4)
by Jonathan Walther on Mon Sep 03, 2001 at 03:35:10 PM EST

Not only do the power utilities have their own copper to everyones home, in Washington state they also have fiber to EVERY home. What? Yes, thats right. Everyones power meter is hooked up to home office by fiber. Apparently with all the marijuana growers and survivalist hippies they felt they had to put special solutions in place to prevent power stealing, and the fiber is part of that plan, don't ask me how.

So, look at the power companies. I think they are going to come out of left field and suprise us all some day.

(Luke '22:36 '19:13) => ("Sell your coat and buy a gun." . "Occupy until I come.")

Re: Power utilities (3.50 / 6) (#6)
by mmcc on Mon Sep 03, 2001 at 07:27:34 PM EST

Electric power lines have inherent problems with noise, and require an expensive modem at both subscriber and exchange ends of the power cable.

If you insist on using copper, you might as well use xDSL over a twisted pair... twisted pairs have better immunity to noise.

I think the main problem with fibre is that it's hard to work with... you can't just solder two pieces together, you have to use and expensive splicing machine. However, if it's already been done, they that power company is in a very good position to offer broadband services to their customers.

[ Parent ]

Last-mile solution in rural Pacific Northwest ... (4.54 / 11) (#5)
by joegee on Mon Sep 03, 2001 at 03:51:42 PM EST

A few months ago I read about a Pacific Northwest power company consortium, called NoaNet, that had installed fiber-cored cabling to monitor remote power meters. When they realized their grossly underutilized bandwidth rich fiber monitoring system was a untapped revenue source they began leasing bandwidth to local ISPs, who now provide megabit per second and better internet service to mostly rural and smaller-town end users at extremely affordable rates.

<sig>I always learn something on K5, sometimes in spite of myself.</sig>
How can I find out about my own power supplier? (3.00 / 2) (#9)
by avatarxy on Mon Sep 03, 2001 at 08:34:27 PM EST

How can I go about finding out if my power supplier (Niagara Mohawk) has fiber run to meters? And, if they do, what's the best way to suggest they offer internet service/lease bandwith? Thanks, Jason

[ Parent ]
Give 'em a call ... (3.00 / 1) (#11)
by joegee on Mon Sep 03, 2001 at 09:38:59 PM EST

Check with their linemen (call their office.) AFAIK this capability is something that has only recently been added as power companies upgrade their infrastructure.

This has nothng to do with the transmission line schemes being deployed in Germany.

<sig>I always learn something on K5, sometimes in spite of myself.</sig>
[ Parent ]
The problem isn't technical at all. (4.81 / 11) (#8)
by bgalehouse on Mon Sep 03, 2001 at 07:42:38 PM EST

If a power company can afford to run fiber to every power meter, running the fiber to the curb is hardly a problem technically or financially. I've also heard of running in through water/sewer lines. The problem isn't the last mile, or the last 10 feet. The problem is that last inch into the house.

I mean, then you would be a telecommunications carrier. And then you would be regulated like one. And the regulations are tough; they were formed to keep monopolies in their place. So tough, that nobody can figure out how to become a carrier without being one to begin with.

I leave it to your imagination who all was involved in drafting these tough, anti-monopoly regulations.

So our salvation may be wirless communication, but not for any technical reason.

I was at a .com with a CEO who knew James Gostling. When he came to have lunch and give advice and talk tech, we talked about may things. In one of our meadering discussions, he explained that it was the regulations which were the problem. This is probably paraphased, but the jist was "If CISCO saw a way to get around the legalities, they'd be doing it themselves in a heartbeat". (This was before the crash, so maybe it wouldn't be quite so rosey)

Bullshit (1.19 / 21) (#10)
by jungleboogie on Mon Sep 03, 2001 at 09:11:55 PM EST

This article is basically someone complaining
that they don't have lit fiber to their house.

Wait for it, Technology marches on buddy.

ok (4.66 / 3) (#20)
by zhermit on Tue Sep 04, 2001 at 12:52:46 PM EST

you try living in a nice suburban community, where the next street over can get DSL (but you're too far), and cable modems have been vaporware for at least 3 years. This article makes complete sense to the millions of us who just can't get anything above 56k.

I have a sig?
[ Parent ]
DSL is not evil. (4.00 / 7) (#12)
by sety on Tue Sep 04, 2001 at 12:30:53 AM EST

since DSL is a technology that runs along copper lines, making it unusable across all the fiber that has been laid with a view toward the future - giving DSL a planned obsolescence, and making it a vary poor investment.

DSL is not a poor investment. The vast majority of the population will still have copper phone lines in 15 years. This easily makes DSL a good investment.

I believe you use the word fiber too loosely. Telecommuications companies today use fiber for SONET rings. This takes very expensive equipment and must have high up times. This is why miles of long haul fiber lays dorment Fiber to the home (last mile) will not run SONET. I would guess people would like to see IP protocol directly to the home for phone/TV/internet access. I have yet to see a land-line phone with a fiber plug in the back. There really would be no need for this as the bandwidth is so low. There is no last-mile fiber standard. It is not simply a matter of running fiber to your door. Do you run one fiber to every home or DWDM things to a seperate box? It is one thing to twist a few wires, run a line test and have a copper line up and running. It is quite another to in someones house or up on a pole splicing fibers together. This takes much more,care,time,experience then copper does today. That means a higher cost. fiber itself is cheap.

Your argument can be applied exactly the same to cable modems. Cable modems run over Coax (higher bandwidth copper wires). Most cable networks use fiber to split up there network and keep the signal loss low. They are not wasting fiber.

The whole reason DSL and the internet is even possible is that fiber is run to CO's. So saying that DSL doesn't use fiber is true in the strictist sense. But all those packets flowing over DSL hit fiber a few miles from someones house. So not all the fiber that is laid now is wasting away is is actually being used for SONET rings that carry the DSL traffic.

Just some random thoughts, but DSL is a good solution. fiber to the home needs an ITU or IEEE standard.

Another problem with optical fiber (3.66 / 3) (#13)
by Cironian on Tue Sep 04, 2001 at 03:50:52 AM EST

Something I was recently made aware of is that the optical properties of most (dont think it affects all materials; anyone know specifics of that problem?) fiber lines degrade substantially after just 10 to 15 years of use, so a couple of years from now we would hit a wave of having to replace all those lines laid in the recent past.

So "laid with a view toward the future" isnt quite right. Again, this is not as much a problem if you are only wiring up you network backbones with this. But if you want to lay fiber to every home you will face major construction work again and again. (Though I guess 10 years from now the problem will most likely be solved so it wont have to happen again after that)

[ Parent ]

age of fiber (4.50 / 4) (#15)
by sety on Tue Sep 04, 2001 at 07:46:58 AM EST

I'm not really aware of anything that phyisically happens to the fiber over time that degrades it.

The big problem today is that old fiber (15 years old) has things like mechanical splices. It was put in before fusion splices were trusted. So if you want to put 10 or 40 Gb/s systems on this you have to dig up the fiber every 2km and replace the splice with a fusion so that the loss is less.

Older fibers always have the problems of not being:

1. Dispersion shifted.

2. High dispersion ( partly because of 1)

3. High amounts of the non-linear dynamic effect PMD.

This could make the majority of old fiber in the ground unuseable for fast systems in the future. Aerial fiber is even worse. There are ways to correct some of these things, but they all cost money. It is too early to tell how feasible they will be.

[ Parent ]

Ethernet in the MAN. (4.00 / 1) (#22)
by zak on Tue Sep 04, 2001 at 06:28:02 PM EST

A new area of development is Metro Ethernet (see the newly formed Metro Ethernet Forum). Several high-profile startups are working on solutions which involve optical ethernet connections in the Metro Area Network. This will provide customers and providers/carriers with Ethernet all the way, from the RJ45/GBIC on the consumer's premises to the net core. The beauty of Optical Ethernet solutions is that it's Ethernet all the way. No nasty SONET or DS*/E* overhead and management headaches.
I work for one of these companies as a driver developer. Our network edge device can be configured with either 24 Fast Ethernet ports, or a combination of Gigabit Ethernet and E1/T1 ports (with more adapters being developed). The provider can configure virtual circuits between edge devices (or the internet backbone), with true quality of service (1..1000 Mbits on a 1Mbit resolution). VCs can be configured on a per-port or per-port and VLAN basis, very very easily.
These solutions are currently a bit too costly for non-business needs. However, placing such devices in residential areas may be closer than we think, and may be the only cost-effective way to provide true 100Mbit connectivity for the (discerning :) ) home user.

[ Parent ]
Fiber is a tough business (3.80 / 5) (#14)
by halr9000 on Tue Sep 04, 2001 at 06:44:16 AM EST

I am lucky and unlucky enough to have fiber to my house. Ok, to the curb really. I live in Georgia, and Bellsouth over the past several years has been playing with IFITL, Integrated Fiber in the Loop.

Great idea in theory. They have fiber from the CO to my neighborhood, then a 2-3 month project had fiber to the curb in front of every house in the neighborhood. Then for those who subscribe, you get a fresh new copper RJ-45 into the house and then into a NIC in your computer. It works great, I cannot complain.

But, according to some friends in Bellsouth, the cost to deploy IFITL has killed the project. The subscription price is $50/month, and due to competitive pressure, Bellsouth rarely could charge an installation fee.

I don't know what the best solution is. I personally hope that wirelsss wins out. Maybe UWB Wireless or 802.11b.

Some links for you. (3.85 / 7) (#16)
by Surial on Tue Sep 04, 2001 at 08:31:48 AM EST

Here are some interesting links relevant to this subject.

"is a signature" is a signature.

Why broadband is failing (3.33 / 6) (#17)
by wiredog on Tue Sep 04, 2001 at 10:03:30 AM EST

From the Washington Post:
"Unlike Internet companies, most of which never had a credible plan to make money, the telecom start-ups generally had proven leaders, real assets and business plans that made a lot of sense. But so many companies flooded in that they slaughtered each other."

The idea of a global village is wrong, it's more like a gazillion pub bars.

Here's a hint: (4.60 / 5) (#18)
by Pink Daisy on Tue Sep 04, 2001 at 10:05:38 AM EST

Two companies that just may have solutions:

SOMA Networks makes high-speed last-mile wireless solutions.

Qualcomm is one of the key companies developing CDMA technologies that will enable high speed third generation mobile wireless services.

There's only so much you can do with copper, considering the variable quality of copper and the number of wires running to each home. Wireless technologies are reaching the price point that makes them attractive, and they are much cheaper for installing a completely new infrastructure than new fiber or copper wires. Also, I've only mentioned two of the many companies working on solutions; there are a lot of them out there!

Well.. while we're plugging companies.. (none / 0) (#28)
by mindstrm on Thu Sep 06, 2001 at 11:11:24 AM EST

http://www.waverider.com for a last-mile solution.

SOMA seems light on any details, and claims 'line of sight and distance are not factors'.
Sounds kinda funny... anyone see more details there? what bands? What tech?

[ Parent ]
International Fibercom (none / 0) (#29)
by lvogel on Thu Sep 06, 2001 at 07:12:36 PM EST

IFCI is a company that specializes in last-mile solutions, as well as most-to-all other aspects of the cable-laying industry :)
-- ----------------------
"When you're on the internet, nobody knows you're a dog!"

-a dog
[ Parent ]
Cheap fibre (3.75 / 4) (#19)
by ttul on Tue Sep 04, 2001 at 11:42:58 AM EST

TeraSpan Networks has figured out how to cheaply lay fibre without digging the street up. They just cut a thin groove in the pavement using a diamond saw and lay in a bundle of fibres. Copper will reach its limit soon. Innovative solutions such as TeraSpan's will bridge the gap.

stop mixing your layers, fiber isn't the solution (none / 0) (#23)
by gps on Tue Sep 04, 2001 at 08:11:47 PM EST

Fiber and copper are both physical layer things. Stop assuming that you -need- fiber anywhere to get a decent amount of bandwidth. You don't.

As someone else has already pointed out, fiber is extreemly expensive to install and maintain compared to copper. This is why fiber is destined to go no further than the cable or telco boxes in your neighborhood and not to your home itself.

You can easily get several megabits per second through a pair of copper wires over a long neighborhood distance to a box like that.

Until there is a serious large company "piggy bank stuffing application" that consumers will fork over loads of money for, you won't see net access to the home faster than a few megabits downlink. Media companies that control the content and stuff it full of advertising or charge you your next of kin to view it are the only ones likely to be interested in doing this.

[ Parent ]
Copper oxides, suffers interference... (none / 0) (#30)
by C0vardeAn0nim0 on Mon Sep 10, 2001 at 01:18:41 PM EST

Fiber don't.

Try setting up an ethernet LAN in your home/office with coaxial copper cables, then pass it near an eletrical engine or fluorescent light and start counting the number of droped packets. You'll count lots of them, thanks to the electromagnetical interference in the copper cable.

Add to this that copper is a metal, and metals are subject to oxidation. I don't know if there's a metal oxide capable of conducting electricity, but copper oxide certainly can't.

There's lots of thing that afects copper lines that prevents it from reaching high bandwidths at long distances. Wheather, powerlines, cars, oxidation, electrical resistance, etc. Copper can achieve high bandwidth only in controled environments with short cables, and it needs lots of maintenance.

But fiber... Fiber is made of glass, and glass is eternal. The only maintenance it needs is on it's plastic/kevlar coating and terminators, and of course, a fix if something breaks it, but it certainly doesn't break all the time.

At longer distances (like, 10 km) it can deliver a much grater bandwidth than coper. a single pair of fiber can transmit/receive at 155 Mbps between two places separated by a long distance. Try that with copper.
Sooner or later people will start demanding things like 10 Mbps or more at home, and this will be impossible to achieve with copper if the distance between the two points (home and the nearest co) is larger than 300 m (the limit for a 10 Mbps ethernet running on CAT 5). In this case youll need to run fiber to the home or at least very close to it (less than 100 m, maybe.

Yes, they are physical layer things, but that's where the similarities stops. Fiber is much more eficient, delivers more bandwith in the same pair, is almost imune to external interference. Unless other alternative shows up, I bet everithing on it.

[ Parent ]
Good Article (4.50 / 4) (#21)
by PhillipW on Tue Sep 04, 2001 at 01:47:11 PM EST

Very good article. There are a few solutions that are being developed and implemented.

The one that comes to mind with me, because I work with it on a daily basis, is a DSL solution called Project Pronto. It is being implemented by SBC's local subsidiary out here, Pacific Bell(the rest of SBC may be implementing this as well, I'm not sure). The longest loop length that DSL will run over and still be stable is around 15,000 feet. The telephone company, in this case Pacific Bell, cuts it off at 14,000, just to be safe. This means a lot of customers are too far away from the CO to get DSL. So what Pacific Bell has done, is build a Remote Terminal out 14,000 feet from the CO, doubling it's range. In some areas that are further than that, The build the RT 14,000 feet from the community, and after the RT it is all run over fiber. This eliminates the distance problem, and utilizes existing infrastructure. Of course, since this is new it is not perfect. Fiber anywhere down the line that is not clean will screw the connection up. But it is still a good idea, and, over time, will improve.

Local DSL (none / 0) (#26)
by tjb on Wed Sep 05, 2001 at 08:12:19 PM EST

I program DSL modems for a living, and while I generally concentrate on the DSP side of things, from what I've heard from our marketing guys, you're right on the money about remote CO deployment.

Lucent makes a DSLAM, known as the Stinger, which does exactly what you said: its a very compact DSLAM with remote administration capabilities meant to be deployed away from the CO and close to the customers. And, as far as anyone can tell (precise numbers are a bit hard to come by), the Stinger is the best selling DSLAM in the world now.

(and this makes me happy :)


[ Parent ]
Fibre to the Curb in Australia's Capital (3.00 / 1) (#24)
by bg on Tue Sep 04, 2001 at 08:41:37 PM EST

TransACT Communications here in Canberra is rolling out a fibre network that will connect home users to their ISPs. They've hooked up 5 suburbs thus far, with 10 more to be done by the end of the year.

They don't provide the internet connection themselves, they just run a fibre connection between the users and their ISPs. Other services going over the fibre include TV, Video on Demand and Telephony.

It's an "open" network, so if a business wants to get into people's homes at high speeds they just need to connect to TransACT, and then they're free to provide services.

It's the first network of its kind in Australia- I'm not sure about the rest of the world.

The TransACT network has fibre up to a node, then copper going out (in a star topology) to between 45 and 65 homes. Node to home distance is never more than 300 metres. This is in contrast to the cable networks which have a fibre upto a node and then they service somewhere between 500 and 2,000 homes off a coax cable on a bus-like network topology.

Allowing for various overheads, bandwidth is about 36 Mbps downstream and 1 Mbps upstream. Or there's the option of symmetrical capacity (13 Mbps downstream and upstream).

Pretty cool. Line rental is about $70/month AUD.

- In heaven, all the interesting people are missing.
It's been solved.... (2.00 / 4) (#25)
by ckm on Tue Sep 04, 2001 at 10:55:07 PM EST

This article is a red-herring. The problem of fiber to the home has been solved by a company called CityNet Telecomunications. They are laying fiber through sewers, avoiding digging up roads and all that other overhead.

You can read about it here [usa today]. and find the company here.

Since the majority of US residents live in cities, it's safe to assume that you will have fiber to your home in the near future.

Also note that Sprint is currently delivering point to point fixed wireless broadband in San Jose, CA (and probably other places) with a 5mbs top speed. They will be rolling out test areas of their new 1.5mbs cellular technology in the next 6 months.

And there are a whole raft of new 802.11b home grown networks, as well as commercial providers.

Note that this is not such a problem in Europe. The Netherlands has had a fully digital telephone system for at least 7 years. I remember when they switched off the last analogue exchange. France has had a largely digital infrastructure for at least 10 years. When I was living there 10 years ago you could get an ISDN line at no additional cost.

In summary, I think that there may be some lag between backbone broadband and residential broadband, but that's likely to diminish greatly in the next couple of years.

It was Bill Gates that said that you'd never need more than 512k of RAM, and someone before him thought that the entire computing needs of Earth could be handled by just a few mainframes... This is the same thing, it may take some time, but the technologies to do it are already here and being deployed.


Comments. (none / 0) (#27)
by mindstrm on Thu Sep 06, 2001 at 11:08:09 AM EST

We should stop calling it 'broadband' and start calling it 'high-bandwidth', it's a better description. Broadband does not mean high speed. That's my nitpick for the day.

Cable, as you put it, had a better start; DSL involves more work. DSL as a tecnhology is fine, but running DSL to every home is not practical.

802.11b is *NOT* a last-mile solution.. let's not call it that. It's wireless ethernet. Plus, you would saturate a neighborhood quickly. Sure, you can USE it for distance transmission, but there are other protocols better suited. Packet radio is way too slow, and not a contender.

Fiber to the home *IS* an answer, and it's been suggested many times before; it's just not economical. The costs involved in bringing fiber to, say, 100,000 homes is staggering, nevermind the support. You can bet that, if it were economical, it would be done already.

What's the new last mile solution? It's all of the above. It's what makes the Internet work... the fact that you can use anything.

Live in a city? You'll get Cable and DSL.
Live in the country, or city with no wired infrastructure, you'll get fixed wireless.

Fiber? That'll be for the select few with buildings wired for it.

Dark fiber everywhere but not a last-mile solution to be found | 30 comments (29 topical, 1 editorial, 0 hidden)
Display: Sort:


All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!