Amusing, yes, but not entirely accurate. At best you were contained in 46 molecules, each chromosome being a separate DNA molecule. Actually, you were never less than a single cell, made up of a very large number of molecules, including complex structures like mitochondria which are critical to cell function but do not appear to be coded for in regular DNA.
I'd like to take this opportunity to thank my mother for my mitochondria, and to thank my wife for providing my kids with mitochondria, since Dad and I (and males in general) can't be bothered with such picky little details...
I don't know if they're still on the market, but you used to be able to buy a CMTS that was just a bridge, not a router. At the time, they were significantly cheaper. We used one as part of a traveling demo arrangement. A single Linux host provided all of the software that a DOCSIS system needs to function:
DHCP
TFTP serving config files
timed using UDP (some modems won't work unless they have a time server)
syslogd accepting network messages (some modems want to report problems)
Some custom software did the routing since we wanted not-quite-normal behavior. The same box was visible on the DOCSIS subnet and provided Samba file-sharing and print services. All on a laptop with a 120 MHz Pentium and 16 MB of memory!
Couldn't agree with you more. For the past 25 years I've been a systems engineer arguing that it's a software-driven world. I've lost track of the number of projects I've been associated with where the hardware folk chose the processor, chose the memory size, chose the peripherals, and then turned to the software people and said "You gotta make it work and you gotta make work on this [brain-damaged] hardware." Doing at least some of the high-level software design first would have resulted in some very different (and not necessarily more expensive) choices about hardware components and capabilities.
Defined-benefit pensions have been in the process of disappearing for years; if you're working in IT (or anywhere else, for that matter), take charge of your own life and start looking to your 401k and/or IRA.
Absolutely. Up until about three years ago, companies with defined-benefit plans were converting to defined-contribution plans at a furious rate. The government suspended that process because it appeared to be age-discriminatory (only older long-time employees see their benefit get cut). The Bush administration is in the process of issuing new rules that state that such conversions are not age-discriminatory regardless of the outcome. Expect to see the wholesale conversion away from defined-benefit pensions to resume -- big companies save LOTS of money through such a change.
If you're young and starting out, live within your means and save as much as you can in your own IRA or your company's 401(k). Compound interest in a tax-deferred account may be your ONLY friend if you intend to retire comfortably some day. And if you think that you can wait to start saving, look at the difference that 40 years of compounding makes (money saved at 25 for an age 65 retirement) versus 20 years (money saved at 45). $10,000 at 7.5% for 40 years grows to $180,000. For 20 years, it's only $42,000.
As some have pointed out, statistical measurements can be useful in IT in the proper situation. Here's a concrete example.
In a particular national network, all DNS requests went to one of two data centers, one on the East Coast and one on the West. During an exercise which measured the response time for a particular end-user service which included a DNS lookup, we discovered that the distributions of response times for DNS lookup for the two data centers were different. Further investigation (had to write some custom data collection software) showed that the problem was that about 1% of the DNS requests sent to the West Coast data center produced no response. The East Coast center produced no response in about 0.1% of the attempts, which was consistent with the known packet-loss rate in the backbone network.
I believe the problem turned out to be a faulty load balancing box sitting in front of the multiple DNS servers. As a result of the episode, however, the service-level agreement for the DNS service was expanded to include response rate and response time, as well as server availability.
At least for mass market media, I would cite the Blockbuster "outside wall" phenomenom. If estimates are to be believed (Blockbuster doesn't publish their statistics), excluding late fees, Blockbuster makes 80% of its revenue from the 20% or less of the titles on the "outside walls" of their stores. Note that in general, these are all recent releases. For the vast majority of movies, interest and the revenue stream decay exponentially over time. I would find it more likely that the studios want such long-term control on all of their products simply because they have no idea which will be the few that will be the exceptions.
I would be much happier if our copyright scheme was based on shorter time horizons that could be renewed by the holder, but the renewal process required that they keep the material on the market in some fashion. For example, Disney could renew the copyright on Steamboat Willie, but only if they re-released it.
I find it enormously frustrating that books less than ten years old are out of print and my ability to access a copy depends on networks of used book dealers or the quality of the library network to which I have access.
NEBS compliance is so much fun! Two of my favorite requirements:
There are very tough requirements about things a circuit board must not outgas when heated (eg, by a fire). Many of the outgas products from heating a board made with standard fiberglass and epoxy are highly toxic. This is dangerous for central office craftpeople who are expected to be trying to extinguish the fire without the benefit of the kind of breathing equipment a fireman would normally be using.
For every N frames more than six feet tall (I think that's the height limit, it's been a while since I really looked), the vendor must provide a stepladder made of oak. Basically it's because oak has been tested to be nonconductive, meet strength requirements, etc. No one has been willing to pay to certify some other material. Sort of like the obsolete processors in the space shuttle -- no one has been willing to pay the cost for certifying another processor to NASA specs.
The only thing I can think of off hand that might be affected by Linux (versus some other OS) would be alarms. Central office equipment is often required to provide alarm signals using relative large voltages or currents, which would require device drivers for relay boards or other ways of handling that much power.
Point #1 of my list: weight will not make your car go faster. This should be self-explanatory.
I know I'm explaining this badly. Weight will not increase the maximum velocity the car reaches at the bottom of the ramp. You are absolutely right that this is determined by the acceleration due to gravity less frictional losses. But a more massive car with the same velocity at the bottom of the ramp will have greater kinetic energy. If frictional losses for two cars of different mass are the same, they will reach the bottom of the ramp at the same time and the same velocity, but on the flat portion of the track, the lighter car will slow down more quickly. Greater mass will not increase the maximum velocity, but it can yield higher velocities at points on the flat portion of the track.
In reality, every year you see races where a car is behind at the bottom of the ramp but "catches up" on the flat portion. This is exactly the behavior you should see if the design has exchanged a small loss in maximum velocity (caused by very slightly greater friction due to increased mass) for a sufficiently bigger gain in kinetic energy.
You want to convert energy of position -- being higher in the gravity well -- into kinetic energy. Given that velocity gain down the slope due to the acceleration of gravity is identical (which is what you're saying in #1), then total kinetic energy at the bottom of the ramp will be higher if there's more mass (e=0.5*m*v**2). Quick sanity check -- does it take more energy to lift 1 unit of mass into position at the top of the ramp versus 2 units?
If frictional losses are the same, the heavier vehicle should win on the flat portion of the track.
Assume that it's a big hosting service and buys bandwidth in quantity. Say that physically they're using an OC-3, that's about 155 Mbps. Their contract is more likely to be based on usage, say $300 per month per Mbps, and the number of Mbps is determined by measuring usage for every 15-minute interval in the month and using the 95th percentile. If that figure is 100 Mbps, then they pay $30,000 for the month. Of course there's also a minimum usage that puts a floor under the payment.
If a site they are hosting gets/.'ed, it creates several 15-minute intervals with abnormally high traffic. This will raise the 95th percentile figure. If that goes from 100 Mbps to 105 Mbps, the hosting service gets charged an extra $1500 for the month. Of course they can avoid that by throttling themselves back at a gateway router, but then every site they are hosting will experience degraded service. Complex traffic shaping on this scale uses significant processing in a router, reducing its capacity and raising those costs. TANSTAAFL. Watching a network grow rapidly has renewed my appreciation for some of the problems of scale.
Actually, an OC-3 isn't all that much bandwidth. I know there are service providers connected to some of the big backbone networks that have multiple OC-12s. I have seen contracts in those cases where there's a sliding scale for bandwidth, so an incremental Mbps-month costs less as you use more.
While not taking the position that the homeland security people are right, and I certainly respect MIT for taking a position where they turn down federal dollars for projects when they disagree with the restrictions that come with the money, I feel obligated to point out that there is a substantial difference between pushing out many of your best, established citizen scientists because of their ethnic or religious heritage (Hitler's Germany) and putting relatively minor restrictions on foreign students working on government-funded research while they are temporarily in this country.
I believe the general response would be quite different if the background checks were required for citizen students of particular ethnic groups. Or if the affected people were established scientists seeking asylum from a government that was persecuting them.
If I may summarize, MS faces enormous problems in the not-so-distant future in transforming themselves from a hot-shot growth company into a mature firm. Their share price will not continue to double, so they won't be able to use options as currency (how would you like to be a relatively new hire with MS options at $120?). Some of their newer ventures may end up being profitable -- I would bet on MSN, particularly if AOL continues to screw up -- but the profit margins will not be nearly as good as those for Windows or Office. They have been unsuccessful, so far, in finding the next big thing in software that everyone wants.
MS is not the only firm with this problem. I would also add Intel and Cisco to the list of large successful tech companies whose share price is way too high for their realistic growth prospects. I have a friend at Intel who reported a rumor that Intel's upper management was shocked at a recent meeting with investment bankers who told them that they were a mature firm, not a growth company, and their share price would adjust downwards drastically as the stock market realized that fact.
Most of the comments people have made fall into the two obvious categories. Note that having to make choices like this are one of the important things to experience in college! My own opinion:
From a legal perspective, your answers are your own and you're probably entitled to do what you want with them. The one exception may be if you've signed some sort of contract with the school in which you surrendered certain specific rights, and that doesn't seem likely.
From an ethical perspective, you would seem to be in a somewhat more difficult position. Your answer may be terrific -- simple, elegant, efficient, something to be truly proud of. The same could be said for some students' essays for questions like "Summarize the economic conditions that restricted Great Britain's actions during the American Revolutionary War." Putting either on the Web adds to that enormously useful collection of information. Identifying them as answers to class assignments or test questions within your university community does a potential disservice to the professor and to future students.
You seem to have come down on the side of the decision that the good from posting the GPL source (to you as well as others) outweighs the problems of a professor having to build a new project or some students failing to learn because it was easier for them to cheat. Personally, I think I would have made the choice the other way.
Can't speak for his case, but there are cases where damn good people are laid off. I was recently put in that situation. The company was acquired, and the entire technical organization is being laid off. The decision was made by a small number of non-tech people at the acquiring company. A handful of engineering managers are being offered positions at the acquiring company's headquarters on the other side of the country. The other 100+ people will be "on the street" within the next six months. Officially I did strategic technology planning and analysis, the new company doesn't consider it important, so I was among the first to go. The fact that I can do a number of other things and do them well was never considered.
Ah, I'm just feeling bitter today. Looking for a new job after 24 years is a hassle.
In the United States, trials are adversarial proceedings. The attorneys are supposed to exercise every opportunity to "win" for their client, within the restrictions imposed by law. Hence juror profiling, motions to disallow evidence, etc. This contrasts sharply with the role of trial attorneys in France, for example, where the purpose of the court is to establish the truth. Trial lawyers in the US are excellent at invoking the "right" emotional responses in jurors -- it's one of the things that makes a good trial lawyer versus an attorney who's good at research, prepping witnesses, writing briefs, etc.
My father was called for jury duty frequently, but never sat on a jury in over 30 years. He was an accountant and an engineer by training, so defense attorneys did not want him in criminal cases because he had been trained to think in terms of the facts. He worked for an insurance company, so plaintiffs' attorneys did not want him in civil cases because there was usually an insurance angle to where the money would come from.
Just nit-picking, but I believe that you have to have made the constitutional argument in the original case -- you can't add that later. That is, you can't introduce a new legal theory (eg, that the law is unconstitutional) that was not laid out in the original defense. Appeals are based on claims that the judge presiding over the case made an error, either in his/her interpretation of the facts, or in interpretation of the law. Failure of the judge to consider constitutionality, if that argument is not introduced by the defense, is not generally an error.
If I had to bet, I'd bet that one of the important ideas that comes out of the 90's will be new versions of the transistor built from polymer semiconductors. Inexpensive better-than-LCD displays and cheap printable circuits are two of the possibilities that could come from this new technology. Would these count as a new idea, or not?
Brin didn't seem overly cynical; after all, he seems to think that the Enlightenment is still working. If you want to be really cynical about it, consider the ways that the United States is moving towards recreating the self-preserving pyramidal structures typical of feudalism.
Incumbents in Congress have a re-election rate over 95%. Top-level bureaucrats move back and forth between government office and the industries that they oversee. Officers of large corporations amass enormous personal wealth even while they lay off tens of thousands of employees. Probable repeal of inheritence taxes, so those hundreds of millions of dollars stay in the hands of a small number of people. Read some of Paul Krugman's pieces (granted Paul is a bit extreme in some of his positions) giving reasons why this may be the last generation in this country with a relatively well-off middle class that can afford a comfortable retirement in their old age.
I try to be generally optimistic, but if you want to be cynical there's a lot of bad-looking things happening these days.
So they can upgrade all they want because they know that they'll get their money back.
I think it remains to be seen whether the cable companies can grow revenues fast enough to handle the mountain of debt most of them have accumulated. Adelphia is already in bankruptcy court. Comcast, by the time they finish absorbing AT&T Broadband, will have almost $1000 of debt per subscriber. At 7.5%, they need $75 per subscriber in free cash flow per year just to pay the interest.
OTOH, if they don't care what happens to the regular shareholders, a quick trip through bankruptcy court could leave the cable operators in pretty good shape.
So what you are really stating is, because cable was laid in the 60's forward, and phone from the 20's forward, both with respective costs of laying the lines, cable is the choice because its began with better bandwidth, and possibly was planned better and benefitted from matured neighborhoods. Whereas phone became a morass with changing neighborhoods and changing technology.
Actually, cable systems as designed and deployed in the 60's were completely unusable for apps such as cable modems -- amplifiers worked in one direction only, etc. Cable companies have incurred enormous debts in the 90's in order to (among other things) rebuild their networks using individual runs of fiber to connect their headends to coax "nodes" servicing 500-1000 homes each, put in two-way amplifiers, etc.
For the phone companies to deploy DSL everywhere would require them to make much the same decision -- push fiber deep into their distribution network and use copper pairs for the last few thousand feet. IIRC (I used to do this type of study for a living), the costs associated with doing that result in a service that needs to be priced at about $100/month.
The same argument actually holds for wireless companies trying for large customer penetration (say 30% of homes in town) as well. They need to deploy fiber to neighborhood APs in order to provide enough overall capacity and performance.
Really torn about it. Part of me is happy that I won't have to work for the new owners. Part of me is horribly depressed by the prospects of being 49 years old with 24 years of service and having to go out and hit the job market.
At least for DOCSIS cable modem systems (not necessarily true for some older proprietary systems) this is not supposed to be true. The DOCSIS modem is an Ethernet bridge with some very specific additional behavioral rules. It should only learn individual MAC addresses from the customer side, and by default it should never bridge unicast packets received from the cable side unless they match a learned address.
When I run tcpdump on my household server (acts as the gateway for our LAN), I can see traffic destined for us, and ARP who-has messages from the CMTS. The ARP messages are Ethernet broadcasts that have to be bridged. If users at Adelphia can see all the traffic, and it's a DOCSIS system, something (probably the cable modem configuration file) is really screwed up.
Actually, the glut that the article refers to is long-haul fiber, and is the result of unrestrained lunatic competition rather than monopolies. There were at least 20 companies furiously building such networks in the 90s (Qwest, Level 3, Global Crossing, etc) each with a business plan that said "We'll capture 20% of the market!" and defective forecasts of the future that said the volume of Internet traffic would double every 100 days. The traffic never materialized and many (most?) of those companies have gone broke.
Keep in mind (Business 101 here) that buying the fiber, putting it in the ground, and lighting it, all costs money and sets a floor on the price that can be charged for the bandwidth. All of the companies who built these networks have similar costs, so none of them can price below that floor for long. Bankers who loan you $1B to build the network have a nasty habit of wanting to get their money back. At some point you have to show a profit.
my cable box comes with the ability to recieve all of the channels too, whats the legal implication there?
I'm not sure what point you're trying to make here. Contemporary analog and digital cable boxes will refuse to tune to channels you aren't paying for. Modified boxes will, but making such modifications is a violation of federal law. A cable-ready TV or VCR will tune to any analog channel, but the cable company will scramble any premium analog content you might receive that way and adding your own analog descrambling circuit is a violation of federal law. If your point is that cable is a precedent for delivering content to users that they are not allowed to access (without paying), you're absolutely right. If your point is something else, I missed it.
Slashdot Mirror
I'd like to take this opportunity to thank my mother for my mitochondria, and to thank my wife for providing my kids with mitochondria, since Dad and I (and males in general) can't be bothered with such picky little details...
-
DHCP
-
TFTP serving config files
-
timed using UDP (some modems won't work unless they have a time server)
-
syslogd accepting network messages (some modems want to report problems)
Some custom software did the routing since we wanted not-quite-normal behavior. The same box was visible on the DOCSIS subnet and provided Samba file-sharing and print services. All on a laptop with a 120 MHz Pentium and 16 MB of memory!Couldn't agree with you more. For the past 25 years I've been a systems engineer arguing that it's a software-driven world. I've lost track of the number of projects I've been associated with where the hardware folk chose the processor, chose the memory size, chose the peripherals, and then turned to the software people and said "You gotta make it work and you gotta make work on this [brain-damaged] hardware." Doing at least some of the high-level software design first would have resulted in some very different (and not necessarily more expensive) choices about hardware components and capabilities.
If you're young and starting out, live within your means and save as much as you can in your own IRA or your company's 401(k). Compound interest in a tax-deferred account may be your ONLY friend if you intend to retire comfortably some day. And if you think that you can wait to start saving, look at the difference that 40 years of compounding makes (money saved at 25 for an age 65 retirement) versus 20 years (money saved at 45). $10,000 at 7.5% for 40 years grows to $180,000. For 20 years, it's only $42,000.
In a particular national network, all DNS requests went to one of two data centers, one on the East Coast and one on the West. During an exercise which measured the response time for a particular end-user service which included a DNS lookup, we discovered that the distributions of response times for DNS lookup for the two data centers were different. Further investigation (had to write some custom data collection software) showed that the problem was that about 1% of the DNS requests sent to the West Coast data center produced no response. The East Coast center produced no response in about 0.1% of the attempts, which was consistent with the known packet-loss rate in the backbone network.
I believe the problem turned out to be a faulty load balancing box sitting in front of the multiple DNS servers. As a result of the episode, however, the service-level agreement for the DNS service was expanded to include response rate and response time, as well as server availability.
I would be much happier if our copyright scheme was based on shorter time horizons that could be renewed by the holder, but the renewal process required that they keep the material on the market in some fashion. For example, Disney could renew the copyright on Steamboat Willie, but only if they re-released it.
I find it enormously frustrating that books less than ten years old are out of print and my ability to access a copy depends on networks of used book dealers or the quality of the library network to which I have access.
-
There are very tough requirements about things a circuit board must not outgas when heated (eg, by a fire). Many of the outgas products from heating a board made with standard fiberglass and epoxy are highly toxic. This is dangerous for central office craftpeople who are expected to be trying to extinguish the fire without the benefit of the kind of breathing equipment a fireman would normally be using.
-
For every N frames more than six feet tall (I think that's the height limit, it's been a while since I really looked), the vendor must provide a stepladder made of oak. Basically it's because oak has been tested to be nonconductive, meet strength requirements, etc. No one has been willing to pay to certify some other material. Sort of like the obsolete processors in the space shuttle -- no one has been willing to pay the cost for certifying another processor to NASA specs.
The only thing I can think of off hand that might be affected by Linux (versus some other OS) would be alarms. Central office equipment is often required to provide alarm signals using relative large voltages or currents, which would require device drivers for relay boards or other ways of handling that much power.I apologize. I didn't interpret "overcome friction" as meaning the same thing that I was saying about increased kinetic energy.
In reality, every year you see races where a car is behind at the bottom of the ramp but "catches up" on the flat portion. This is exactly the behavior you should see if the design has exchanged a small loss in maximum velocity (caused by very slightly greater friction due to increased mass) for a sufficiently bigger gain in kinetic energy.
You want to convert energy of position -- being higher in the gravity well -- into kinetic energy. Given that velocity gain down the slope due to the acceleration of gravity is identical (which is what you're saying in #1), then total kinetic energy at the bottom of the ramp will be higher if there's more mass (e=0.5*m*v**2). Quick sanity check -- does it take more energy to lift 1 unit of mass into position at the top of the ramp versus 2 units?
If frictional losses are the same, the heavier vehicle should win on the flat portion of the track.
If a site they are hosting gets /.'ed, it creates several 15-minute intervals with abnormally high traffic. This will raise the 95th percentile figure. If that goes from 100 Mbps to 105 Mbps, the hosting service gets charged an extra $1500 for the month. Of course they can avoid that by throttling themselves back at a gateway router, but then every site they are hosting will experience degraded service. Complex traffic shaping on this scale uses significant processing in a router, reducing its capacity and raising those costs. TANSTAAFL. Watching a network grow rapidly has renewed my appreciation for some of the problems of scale.
Actually, an OC-3 isn't all that much bandwidth. I know there are service providers connected to some of the big backbone networks that have multiple OC-12s. I have seen contracts in those cases where there's a sliding scale for bandwidth, so an incremental Mbps-month costs less as you use more.
While not taking the position that the homeland security people are right, and I certainly respect MIT for taking a position where they turn down federal dollars for projects when they disagree with the restrictions that come with the money, I feel obligated to point out that there is a substantial difference between pushing out many of your best, established citizen scientists because of their ethnic or religious heritage (Hitler's Germany) and putting relatively minor restrictions on foreign students working on government-funded research while they are temporarily in this country.
I believe the general response would be quite different if the background checks were required for citizen students of particular ethnic groups. Or if the affected people were established scientists seeking asylum from a government that was persecuting them.
If I may summarize, MS faces enormous problems in the not-so-distant future in transforming themselves from a hot-shot growth company into a mature firm. Their share price will not continue to double, so they won't be able to use options as currency (how would you like to be a relatively new hire with MS options at $120?). Some of their newer ventures may end up being profitable -- I would bet on MSN, particularly if AOL continues to screw up -- but the profit margins will not be nearly as good as those for Windows or Office. They have been unsuccessful, so far, in finding the next big thing in software that everyone wants.
MS is not the only firm with this problem. I would also add Intel and Cisco to the list of large successful tech companies whose share price is way too high for their realistic growth prospects. I have a friend at Intel who reported a rumor that Intel's upper management was shocked at a recent meeting with investment bankers who told them that they were a mature firm, not a growth company, and their share price would adjust downwards drastically as the stock market realized that fact.
From a legal perspective, your answers are your own and you're probably entitled to do what you want with them. The one exception may be if you've signed some sort of contract with the school in which you surrendered certain specific rights, and that doesn't seem likely.
From an ethical perspective, you would seem to be in a somewhat more difficult position. Your answer may be terrific -- simple, elegant, efficient, something to be truly proud of. The same could be said for some students' essays for questions like "Summarize the economic conditions that restricted Great Britain's actions during the American Revolutionary War." Putting either on the Web adds to that enormously useful collection of information. Identifying them as answers to class assignments or test questions within your university community does a potential disservice to the professor and to future students.
You seem to have come down on the side of the decision that the good from posting the GPL source (to you as well as others) outweighs the problems of a professor having to build a new project or some students failing to learn because it was easier for them to cheat. Personally, I think I would have made the choice the other way.
Ah, I'm just feeling bitter today. Looking for a new job after 24 years is a hassle.
My father was called for jury duty frequently, but never sat on a jury in over 30 years. He was an accountant and an engineer by training, so defense attorneys did not want him in criminal cases because he had been trained to think in terms of the facts. He worked for an insurance company, so plaintiffs' attorneys did not want him in civil cases because there was usually an insurance angle to where the money would come from.
Just nit-picking, but I believe that you have to have made the constitutional argument in the original case -- you can't add that later. That is, you can't introduce a new legal theory (eg, that the law is unconstitutional) that was not laid out in the original defense. Appeals are based on claims that the judge presiding over the case made an error, either in his/her interpretation of the facts, or in interpretation of the law. Failure of the judge to consider constitutionality, if that argument is not introduced by the defense, is not generally an error.
If I had to bet, I'd bet that one of the important ideas that comes out of the 90's will be new versions of the transistor built from polymer semiconductors. Inexpensive better-than-LCD displays and cheap printable circuits are two of the possibilities that could come from this new technology. Would these count as a new idea, or not?
Incumbents in Congress have a re-election rate over 95%. Top-level bureaucrats move back and forth between government office and the industries that they oversee. Officers of large corporations amass enormous personal wealth even while they lay off tens of thousands of employees. Probable repeal of inheritence taxes, so those hundreds of millions of dollars stay in the hands of a small number of people. Read some of Paul Krugman's pieces (granted Paul is a bit extreme in some of his positions) giving reasons why this may be the last generation in this country with a relatively well-off middle class that can afford a comfortable retirement in their old age.
I try to be generally optimistic, but if you want to be cynical there's a lot of bad-looking things happening these days.
OTOH, if they don't care what happens to the regular shareholders, a quick trip through bankruptcy court could leave the cable operators in pretty good shape.
For the phone companies to deploy DSL everywhere would require them to make much the same decision -- push fiber deep into their distribution network and use copper pairs for the last few thousand feet. IIRC (I used to do this type of study for a living), the costs associated with doing that result in a service that needs to be priced at about $100/month.
The same argument actually holds for wireless companies trying for large customer penetration (say 30% of homes in town) as well. They need to deploy fiber to neighborhood APs in order to provide enough overall capacity and performance.
Really torn about it. Part of me is happy that I won't have to work for the new owners. Part of me is horribly depressed by the prospects of being 49 years old with 24 years of service and having to go out and hit the job market.
When I run tcpdump on my household server (acts as the gateway for our LAN), I can see traffic destined for us, and ARP who-has messages from the CMTS. The ARP messages are Ethernet broadcasts that have to be bridged. If users at Adelphia can see all the traffic, and it's a DOCSIS system, something (probably the cable modem configuration file) is really screwed up.
Actually, the glut that the article refers to is long-haul fiber, and is the result of unrestrained lunatic competition rather than monopolies. There were at least 20 companies furiously building such networks in the 90s (Qwest, Level 3, Global Crossing, etc) each with a business plan that said "We'll capture 20% of the market!" and defective forecasts of the future that said the volume of Internet traffic would double every 100 days. The traffic never materialized and many (most?) of those companies have gone broke.
Keep in mind (Business 101 here) that buying the fiber, putting it in the ground, and lighting it, all costs money and sets a floor on the price that can be charged for the bandwidth. All of the companies who built these networks have similar costs, so none of them can price below that floor for long. Bankers who loan you $1B to build the network have a nasty habit of wanting to get their money back. At some point you have to show a profit.