Cable Co's Want More Control Over Your Network

Moonshine Coward writes: "'The CAT and the NAT' in latest issue of www.cedmagazine.com discusses Cable labs and their efforts to come up with a 'better' protocol than NAT that allows them more control over devices behind your cable modem. Their upside on this...$4.95 per IP per mth. Their #1 concern...people putting in 802.11b hubs and sharing with their neighbors. Fine in principle and if it gets them drooling enough to speed up the deployment of fiber to the home it might be a good thing. However I can see way too many downsides...not least of which is being nickled and dimed to death..my webcam, cable ready microwave, refrigerator, pictureframe that shows revolving jif's ... each costing me $4.95 p.m. -- all on top of regular $39.95 cost." Note: the article is written from an interesting point of view -- it's aimed at the people who want to collect the additional per-IP charges.

Wrong way to meter usage

What relevance does the number of devices behind the cable modem have? The reality is that the real load on their system is gross throughput, and if there really is a problem of abusers then the natural solution will be in the realm of additional bandwidth costs: Joe will be a lot less likely to set up a 802.11 network if it costs him $5 / GB past 5GB or whatever.

As a bit of perspective here: I hope they didn't have to do any of this, but the reality is that the "honest" among us end up paying when people abuse these sort of commercial services : i.e. they price based upon the requirements to support the average Joe's bandwidth, so when BillyBob opens up his cable modem to 10Mbps with SNMP and then sets up a warez FTP site and shares his connection with his apartment complex, then that ends up cost ME more in the long run (or alternately, and worse, the service is withdrawn entirely because it isn't economically viable).

Cable Modem's Real Constraints / Openness

The real constraint isn't running servers - it's Upstream Bandwidth and Complaints/Publicity which are the problems.

• The technology is highly asymmetric - it can handle lots more downstream bits from the head end to the user than upstream bits from the user to the head end. (Beyond that, it's really symmetric, and since most users are couch potatoes, upstream isn't a problem from there on up.) Early cable modem hardware couldn't limit bandwidth, so the "No servers" threat policy was used as a substitute for technology. Newer hardware can do bandwidth limitation, but by now many of the cable companies have forgotten why they set that policy, and they're clueless about how desperately they need high-bandwidth customers and how this is losing them business. There are still performance concerns, but using a webcam to video-call grandma burns a lot more cumulative upstream bandwidth than easily-limited web service.

• The real problem is that they can't risk bad publicity about service quality, like the Obnoxiously dishonest but effective telco "Web Hog" commercials, and in the early years, service quality was really variable, and often bad, and they suffered a big PR hit about overloaded service in their beta-test cities that turned out to be because of some bad hardware, but the public perception of bad performance in overloads overshadowed the later explanations of what went wrong. So they really don't want some pr0n server dogging some neighborhood's network performance and leading to customer complaints, plus that kind of thing gives them a bad image. And face it, while it makes much more sense for you to use your 20GB disk drive for pictures of your kids and cats than to upload it to a colo-based web service, just so grandma and your friends and a few random servers, can see it, which needs to run a much higher service quality than you do, popular web sites (ok, pr0n, warez, and pirate music) are much more likely to have an impact on upstream bandwidth than your much more respectable uses. Obnoxious as it is, it doesn't make economic or social-policy sense for the cablemodemco to waste its time debating with you about whether your server is a Politically Correct Family-Oriented Site or a Politically Incorrect Too-Popular Web Server. So they really need to get even better at bandwidth control - which they could do pretty easily if they were technically smarter folks who weren't in deep financial trouble right now.

The whole Cable Openness debate a couple of years ago was bogus, and ISPs and Cable Companies both mishandled it. Until PPPoE, the technically right architecture for a cable modem service was to do routing from the head end on up, which makes the traditional ISP's bundled service (modem access, routing packets to Rest Of Internet, and mail/web support) much less competitive, because it's Already Too Open - the cableco will route your packets anywhere you want them to go, without the ISP's bottleneck, and that leaves them competing with free email and web services (including the cableco's portals), so their only value adds are personalized service quality and avoiding advertising banners. The other two openness issues are wholesale pricing / billing, and the afore-mentioned service restrictions. PPPoE strikes me as an ugly kluge that's mainly designed to make it easier to shut off accounts for non-payment, charge extra for some services, and force traffic into bottlenecks like some ISPs, and it's a bad idea as are most of the different NAT options cablecos play with.

What the cablecos should have done is realize that they desperately need customers and use two ways to get them:

• Cooperate with the small ISPs, or at least with large marketing-oriented ISPs, giving them some cut of the bill to bring in customers and maybe handle billing. Much better than fighting with them, which cost several years of businesses opportunity, though they did spend part of that time fixing the ugly pre-IP technical infrastructure that many local cable TV companies had.

• Encourage unrestricted development of cool applications that will make people want to buy broadband to get them - whether it's things that depend on always-on, or things that need higher bandwidth, or locality-based things like neighborhood-watch cameras, or peer-to-peer games or whatever - which are much more likely to come from some random Internet users or random industry than just the cableco's own development efforts. Two classic applications are ICQ and Napster. (Some cablemodemco folks do get it about Napster, so they alternate between saying "Napster Bad, Servers Bad, Intellectual Property Licensing Good, Piracy Bad" and saying "Well, Duuhhh, of course we like killer apps that make everybody want to buy broadband, as long as they can make the locality work better so it doesn't dog our network, we just say bad things about Napster because our lawyers tell us we have to, and at least it's in better taste than pr0n.")

I've found the whole "Stop the Nasty Thieving Bandwidth-Sharers" publicity campaign to be in bad taste and a tremendous display of lack of imagination - not only do the cablecos have to cope with the reality of cheap radio and NAT hardware and NAT and routing software, but they Still desperately need ways to bring in many more customers, and should figure out how to use this technical opportunity to get them. Of course, cluelessness isn't a new problem for these folks :-) See: Use a Cable Modem, Go To Jail [geocities.com] and the Slashdot Ensuing Discussion [slashdot.org].

Lots of Disclaimers - I'm posting this as Anonymous Coward, because I do work in this industry and my opinions are Extremely Not My Employer's, especially the bit about Napster which I just didn't say at all, and you didn't read it here. But hey, I've been ranting like this for a while, and I'm not mentioning their names, because it's strictly my own opinions, not theirs, and besides, as a stockholder of several of these companies I'd appreciate it if everybody in the computer and communications industries could start to get some clues again. We need to start doing synergy, not fighting each other, so we can make some money. And there are several other rants I left out of this one, like how they've dropped the ball on totally transforming the voice telephony industry :-)

Bill The Anonymous Coward

It's actually quite reasonable

There is (was) money in hosting. People need access to the Internet to send data. You can warehouse your servers or you can rent thick-pipes (T1+) that gives you bi-directional bandwidth. Therefore, hosting companies buy large amounts of bandwidth (bidirectional) or are big enough to carry it themselves with peering.

Now home users want downstream bandwidth.

Solution? Buy the bulk bandwidth, and sell the upstream via hosting and the downstream via broadband.

It's not a rude situation.

If you want bidirectional bandwidth, you can get it. Get a T1 or SDSL at home.

It costs more?

Of course it does! Upstream bandwidth is expensive, downstream is cheap.

Therefore, ADSL is priced based upon the little bit of upstream used and you get a high speed downstream connection.

It's economics. If you want upstream bandwidth, buy it. You aren't entitled to it.

Alex

I'm not sure I see the real argument

(Well, okay, the real argument is probably that the providers see a way to make more money but....)

I pay for a certain amount of bandwidth. Why do they care how it gets used? If I spend my 10 MB/s downloading porn or if I only use half of it and then let my neighbor use the other half...seems like the problem is not people "stealing" bandwidth but the providers not provisioning correctly.

Re:I'm not sure I see the real argument

Well the problem is that you don't really pay for that amount of bandwidth. They are banking on the fact that most people won't max out the connection all the time, so you're paying for a certain maximum amount of bandwidth. If everyone does max out their bandwidth all the time, then they'd be forced to actually charge you for the full bandwidth - likely it'd be significantly more than you currently pay. Then your arguments would hold up, but not with the current $40/month or whatever you pay.

Re:I'm not sure I see the real argument

Here is why they don't charge based on bandwidth: they would make less money. How much bandwidth do you think the "average" person uses? A little web browsing, a little email, *maybe* a total of 3MB/day, for around 90MB/month. Any reasonable system of paying for bandwidth would have to be in the ballpark of $10/GB or less (really far less would still be generous to the cable companies, a Commercial T1 line capable of 300GB/month is $400), and so the majority of users paying $40/month now would instead be paying $1/month. Ouch, there goes the profit.

Re:I'm not sure I see the real argument

Damn right about that. I pay for a set amount of bandwidth and I should get it. My site (above) had just upgraded its bandwidth and was deciding between colos and a T1...funny thing was that several of the folks that had talked about doing the T1 had mentioned an increased price for 'Excessive' Bandwidth.

What the F*ck????

I was willing to pay for a set amount of bandwith...if I saturate it 24/7 why should it matter? I can't imagine how I could have excessive bandwidth...do these guys sense that my lines are saturated and go out and install a second line (actually we had enough lines run to install several....when ya go up 14 flights, ya don't want to do it again).

Ok, that is a business situation, so it is *slightly* different. Then again, when I look at my DSL contracts, it really is the same wording. I am paying for 512k SDSL. They promise 512k of bandwidth ready to use...I'm going to damn well use it. I'm not paying per meg, thats not what they advertised, they advertised always on 512k of service.

Yeah yeah, I know the reality of this, and ya'll all know the reality is that the companies are banking on the fact you won't be using the service 24/7. This is the same reason biz phones cost 10x what residential phones do because the assumption was that residential phone users don't use the phone as much as businesses (we're ignoring the modem users out there because this is still a model predated from the advent of the 'internet' as the former AOL User know it).

If companies want to charge per meg, let them advertise that...it doesn't sound as sexy to say 10Gigs of downloads a month as it does 512k UNLIMITED Downloads.

You will get far less folks buying it if it sounds limited in any way. Too f*cking bad. Stop advertising it as unlimited and let us do what we will with our lines. I don't have more than one computer hooked up to my DSL...I have my little linux box hooked up to it...I have several other computers hooked up to the box, but none of them actually touch the lines, so its none of their business what else I have in my home.

Again, the problem I have with this is truth in advertising...don't get mad when people take you for your word...

clif marsiglio

This won't solve any problems

The proposed CAT doesn't sound like it breaks NAT but simply replaces it (or works with some sort of enhanced NAT). As long as folks have a way to run a NAT service (i.e., running a Linux router behind the cable modem), the "nightmare scenario" of bandwidth sharing won't be stopped other than through bandwidth usage monitoring, which can be done now.

CAT might be helpful to manage sanctioned home-networking schemes, but it won't solve the problem the article addresses.

What is the TRUE value of an IP Address?

from the article:

"What's the value of the stolen goods? Revenues associated with additional IP addresses, for one. Let's say one in 10 of the 5 million U.S. cable modem subscribers are usurping IP addresses without paying the $4.95 per month fee that's typically charged (beyond a pre-specified limit, which varies MSO to MSO.) Right off that bat, that's just shy of $30 million lost, annually."

I've never ran an ISP, so i'm not familiar with how IP addresses are doled out to the "big" guys. Interesting that they calculate the "losses" at $5.00 a month.

A long time ago, weren't different classes of IP addresses handed out for free? How does one put a price on these things?

Furthermore, i thought there was a shortage of IP addresses now. If they're going to implement some funky $5.00/month additional IP charge, i actually wonder if these IPs are going to be routable ones, or an IP on some cheezy intranet, unaddressable to the outside world (as if the cable companies were themselves NATting the connection for you from your private $5.00/month address.)

Two computers makes me a thief?

This article is a misleading justification of price gouging. "The good news is, the dishonest people who know how to do it are already doing it..."; clearly anyone with two computers must be a dishonest thief.

They discuss sharing amongst neighbors, but what they are really upset about is not being able to charge for every device I own or sharing amongst roommates. Nowhere is the fact that even toasters are getting IP addresses mentioned, and none of the technology they are looking forward to will allow the provider to differentiate between my toaster and my neighbor's computer.

So the interesting question to me is, why does my service provider deserve more $$$s because I own three computers, a net-connected TiVo, and an internet enabled toaster or stoplight? Aren't they still just providing me a single connection and some bandwidth? What right do they have to charge for my toaster? Do they have a contract with *me*, or with *my device*? They seem to think they are providing my computer with a service; I happen to believe my computer can't sign a contract, so the service is provided to me, and this price gouging shouldn't be allowed.

Re:Two computers makes me a thief?

> What right do they have to charge for my toaster? Do they have a contract with *me*, or with *my device*?

Depends on who you ask.

If you ask a /.er, they have a contract with you.

If you ask a pigfscking marketroid who believes (in the words of the article), that "[a] crucial part of the success or failure of broadband home networks will be the set-up and ongoing care processes used to link PCs and consumer-electronics gear", then no, they have a contract with your devices.

...or rather, that "If we can find a way to charge you $4.95/month for your TiVO and another $4.95 for your toaster, we will."

Personally, I have no problem with saying "thou shalt not 802.11 thy neighbors onto thy cablemodem" -- cablemodem subscriptions really aren't priced with a full pipe in mind. If you need a full pipe 24/7, buy a T1 or T3.

But the solution to that problem is monitoring of bandwidth and peak usage. (And yes, the article even acknowledges this -- "until then [when we have the brave new world of us charging for your toaster], all indicators point to DOCSIS 1.1, which includes methods to monitor bandwidth consumption [...] and speed [...]".

Meantime, if CAT asks my firewall "Pardon, NAT, but what's that behind you?", I'll tell my firewall to tell the CAT to go stick itself in a sealed box with a poison bottle and a hammer hooked up to an intrusion detection system, and as far as they're concerned, my network can remain in a superposition of states until observed.

(Of course, that's redundant. Any BOFH knows that every computer network remains in a superposition of states between "up" and "down" until they actually try to accomplish something on one. ;-)

First Gripe!

The other day I went to my brother's house with my laptop. I couldn't remember a few commands to release and renew my IP address for some reason so I decided to call Road Runner tech support. For those that don't know, Road Runner is a cable modem service provided on a franchise basis by companies such as Time Warner.

In any event, they were slow but helpful. I noticed during the help call they asked a million silly questions that had nothing to do with my issue. The call should have taken about 2 minutes but it actually took about 8-10 minutes because of these questions (e.g., What is the brand of your cable modem?, What is the serial number on your cable modem?, When is the last time you called us?, and so forth). These questions were asked after I got the command that I needed. It was actually painful to get the guy off the phone. He wanted to check and verify basically the entire setup of my brother's computer and cable connection.

Now, I don't know about you, but this kind of thing really rubs me the wrong way. It isn't support. And, despite what many companies think, it is not Customer Relationship Management (CRM). It is 100% hassle. I am pretty sure this kind of "support" is used to control users and ultimately squeeze more money out of them.

On the one hand, I am not happy about this kind of user support. On the other hand, I am glad that I can even get a good high speed connection. It does cost more than dial up, but it is worth it to me given my career. In any event, I really wish there was more competition. I don't have a choice but to suck it up and quietly complain on Slashdot.

ISPs should be ISPs!

How long is it going to take before ISPs start realizing that Internet Service Provider means Internet Service Provider? I just want a pipe with some bandwidth, to use as I want. This seems a simple enough notion, but the ISPs are all into "we'll sell you a piece of a pipe, as long as you don't use it much, and not for things we don't like."

Clue to ISPs: Sell the pipe. Don't worry about what goes through it unless you're sitting on a subpoena or something. Everything else is silly optional garbage.
--G

Re:ISPs should be ISPs!

Yes, exactly.

In Chicago, we got so sick of sucky internet providers that we banded together and created a Coop, where you pay for only the pipe, and you get what you pay for.

www.ISPFH [ispfh.org].org

The drawbacks?

It ain't cheap.

a bigger problem than you realize

I used to work for a cable modem ISP (until they went out of business last January). People sucking up an inordinate amount of bandwidth on "consumer" accounts were a huge drain on our resources. Usually it was spammers or people running high volume websites at home, but we also had a few folks with as many as 30 computers on one cable modem. We were only charging them $50 a month, but they were eating up almost an entire T1 all by themselves. Losing $1000 a month to one customer is not a good way to stay in business.

It got so bad in one area we actually started putting together a database of MAC addresses, trying to map them to individual customers (even with NAT, the MAC address of the original computer is in the packet). Unfortunately, that project was just starting when the company filed for bankruptcy.

That said, an easier and more effective solution would be to put QOS restraints on people. Who cares how many devices are hanging off one network connection? It's the bandwidth they're using that's important. And if bandwidth were limited to cable modem customers they wouldn't be so eager to share what they have with all their neighbors.

Cory

Re:a bigger problem than you realize

The problem is on both ends though. Until Febrary when the cable modem service provider I was using went belly up (ISPChannel) I was reasonably happy with the service. Four fixed IP's, 500k/sec downstream, 200k/sec upstream. Reliability was less than pristine, but at least some fraction of that was in the ancient cables run through the city.

My issue here is with the bandwidth. The cable modems were all throttled to restrict the upstream and downstream speeds we could utilize. I was limited to 500k/sec as mentioned, but the entire city was fed by 4 T1 connections. We had roughly 1000 users, each throttled to 500k/sec sharing a 6M/sec pipe.

You do the math. There are similar cases with DSL providers hanging 8000 ADSL users at 1+M/sec of a Redback serviced by a single DS3.

The replacement service, Excite@Home, was no better. Worse, in fact, since they had a No Servers policy and used to aggressively scan for them. No improvement in service or bandwidth. Just a loss of freedom to use the bandwidth we were already paying for.

The providers are complaining about people "stealing" bandwidth when they are massively over-subscribing their systems. If I am paying for bandwidth, I expect to get it. This "they're stealing IP's and sharing the pipe!" line is just a feint to cover the fact they are so massively over-subscribed they can't possibly support the userbase they have.

If my link is throttled, then HOW I use that link is realy no business of my ISP's - unless I'm doing something that's actually against the law. If they don't have the infrastructure to support the bandwidth I'm paying for, that is not MY problem . If they can't support X users at Y bandwidth, then they have no business SELLING X users Y bandwidth.

In other venues, it's called fraud.

Sorry, the ISP's aren't getting my sympathy.

What's with the Dr. Seuss quotes?

I do not want it in my box.
Not on my hard drive's precious blocks.
I do not need it in my house.
I will not click it with my mouse.
My packets fly throughout the air,
I use my laptop anywhere.

I will not switch my NAT with CAT.
I will not switch, and that is THAT!

:)

-prator

Keep Services Separate

As a consumer with a long term view, I'd much prefer a commodity market for packet delivery - just as I would for any other essential utility such as phone or electric.

I'd be willing to pay based on Quality of Service parameters, time of day, mean bandwidth, maximum latency, etc., but definitely don't want the service provider reaching into the guts of my home network as part and parcel of the service. Naturally, services based on open standards are subject to greater rigor in the competitive marketplace than closed "standards".

While I realize that no stone goes unturned in the marketing departments seeking to

• "provide solutions" ,
• "add value" ,
• "open new revenue streams",

it would be as if my electric company were billing me for every circuit in my house instead of just the 200 A service to the meter! As another example, it would be as if your trucking company started to provide warehousing and inventory control of your goods.

It's fine to provide and charge services for a separate business of Home LAN Construction and Management (assuming you trust your vendor), but artificially mixing packet transport providers with this other service seems to me to be just another attempt to provide a gratuitous lock-in in the guise of and end-to-end "solution".

Alas, people will probably fall (again) for a well-marketed scheme to reduce apparent complexity, even as they remain unaware of the long-term consequences of their choices.

The costs of simplification are greater than many realize.

LOL my cable company tried this

they charge $5.00 a month per IP. It took a while but the way they do it is by mac address verification. They know your modem mac and if the see anything elso online they halt service and require a remote reset. I was able to get my modem's mac address and using my Linksys router, assign IT the same mac :) Now I've got dhcp running and they are none the wiser. My sdsl connection is superior in ping time and reliability but it is hard to beat 2.5 mb download of of astound cable. PL sucks so I game on the SDSL and pir8 my muzak from the cable :)

[e-mail]

That's right a "Technology Analyst" with an AOL address. Fuck, I wonder how much this person gets paid, an easy job, easy money, and you don't have to know shit about what you're talking about.

Someone needs to smack this person with a cluestick. Has this person heard of cable companies that encourage you to use NAT? What does this person think that a gateway running NAT would look like to this fancy new computer counting technology? Has this person actually neworked two computers together, or did (s)he just read "Wired's history of the Internet and NAT, for dummies?"

Culture Shift...

What I found most interesting about this article was the amount of time it spent name calling-- in particular, the IPS's users 'thieves'.

Of course any movement has its particular socialization. The OSS movement in particular hangs on the 'Information Wants to Be Free' slogan.

It's a little more extreme in this case. The author of the article, and probably the magazine that published it, has a definite agenda to push. The agenda here is to try to limit the amount of bandwidth any one user uses per month. In this case, they're pushing their new 'standard' (*snicker*), and are trying to convince the readers of the article that it's not only right to force that on their users, but that the users need have done something wrong and criminal that they need to be punished for.
Personally, when I pay for cablemodem service, I figure that if I pay $50/month for 384kbyte/s service, then I'm paying $50 for
384kbyte * 2678400 and whatever I don't use is just a bonus for the cable co.

It's obvious that Cable providers would have a different viewpoint, but to criminalize their oppozing viewpoints is altogether more than is called for.

The problem is in the charging paradigm...

The problem is that the cost structure of ISP services doesn't match the pricing structure. Charging per bit moved wouldn't work, because for most residential service the main cost is infrastructure support (the cost of maintaining the pipe, regardless of whether it's used). But charging only for access, as is currently done, doesn't reflect the scarcity of the actual resource -- bits moved.

The only reason we (residential customers) have to sign no-resale agreements is that the ISP's pricing structure is a poor match to the cost structure. Think about it: if the match were better in the high-demand case, then no agreement would be necessary. Does the power company forbid you from reselling your power? No -- but it doesn't make economic sense for you, because the price structure matches the cost OK in the high demand case.

The no-redistribution agreeent is a kludge that doesn't even work to limit customer bandwidth in all cases. Typical ISPs might oversell their pipes by a factor of 50, so each user must stay below 1/50 of their long-term-average bandwidth or else the ISP loses money. I just upgraded my DSL connection to 640kb symmetric, and one use I'm putting the pigger pipe to is listening (at work) to my home mp3 jukebox. That uses 128kbps, or just over 1/5 of my pipe -- so my ISP, who charges only for access, loses out on the deal if I leave the stereo running all day.

A low-volume NATted subnet doesn't affect the fan-out rate nearly as much as a heavy data mover like my mp3 stream -- though it does use slightly more bandwidth. A high-volume NATted subnet increases the spikiness of the load on the ISPs pipe and requires beefier infrastructure -- so you should pay for it.

It seems to me that the ISPs that charge nothing up to some volume of data flow, then a fee per gigabyte above that, have the right idea. That charging scheme matches well with the actual cost of high-volume users. (Cell-phones work that way too...)

What a load of crap

What's the value of the stolen goods? Revenues associated with additional IP addresses, for one. Let's say one in 10 of the 5 million U.S. cable modem subscribers are usurping IP addresses without paying the $4.95 per month fee that's typically charged (beyond a pre-specified limit, which varies MSO to MSO.) Right off that bat, that's just shy of $30 million lost, annually.

Except there aren't any additional IP addresses being used. And of course, as with most speculative damages, this fails to take into consideration the fact that many of these additional computers would not be networked for internet access at $5 per month if there were no "free" alternative available. Consumers gaining functionality does not automatically equate to companies losing profits, especially if the service offered is not the one desired (IP addresses vs. just a data pipe).

With NAT-based hubs, cable providers won't be able to see into all connected devices-making remote troubleshooting difficult-because, again, the NAT is speaking for all connected devices.

Oh no, my cable company won't be able to mess around with the equipment without my knowledge. I'm so worried.

CAT could replace NAT altogether, at least in equipment hand-picked by MSOs for home-network service packages. ... At the very least, cable MSOs involved in CableHome want a counting mechanism, with parameters set by them, that specifies a maximum number of connected devices.

Um, why should my cable company be able to penalize me for having devices that aren't routinely (or ever) used for internet access? So I guess I'll need NAT in the CAT... This whole article is one big piece of misinformation and FUD. My cable company doesn't need to know what I have on my private network - they provide the pipe, I use it. They might be able to monitor some of the data that goes through their network, but anything more invades my privacy (ethical argument, not legal argument) and puts my network at risk of attack. NAT will be around until the cable companies buy a law banning it, and then it will still be around illegally.

You can do that?

Hell, I never even thought of it. They can't even detect that I'm sharing the connection? I'll get with the neighbors now! Thanks, CED Magazine!

My views, plus a future problem

First my view:
I used to work in the cable modem industry, and my beliefs made it very hard to me to tell people that they needed to cough up an extra $4.95 per computer they wanted online.

I always looked at it like every other cable or electricity or phone service. You pay a certain amount of money for a line that goes up to your house, and the ability to use the service provided in general.

Think about it. I can have 1 phone, or 10,000 phones all connected to the same phone line. The phone company doesn't care, so long as I pay for the number of calls I make. I can have 1 outlet, or 10,000 outlets. (Or one desk lamp, or 10,000 desk lamps.) The elctric company doesn't care, so long as I pay for the amount of electricity used.

The cable company will let me connect 1 or 10,000 televisions up to their CATV service, so long as I pay my monthly bill for the channels I recieve.

Similarly, I should be able to have 1 computer, or 10,000 computers, so long as I pay for the bandwidth and IPs I use. In my case, I use 1 IP amongst 4 computers, and have opted to pay for the fastest cable modem service available, making it easy for all 4 computers to be using the service without noticable speed problems.

I see absolutely nothing wrong with my setup.

Now for the problem:
IPv4 has a limit number of valid IP's available. Many of the class A ranges are already taken by telco's and large network companies. If everyone obeyed the cable company's silly policies about 1 IP per computer, they WOULD run out of IP space. Yes, it would be a while, but if everyone that could have cable television had cable internet, and they all had an average of 1.5 PC's in their homes, you're looking at more than likely more IPs than are currently available.

MSOs, revenue, DOCSIS 1.0, and DOCSIS 1.1

I'm sure many of you have seen those hilarious DSL commercials that cast cable-based broadband access in a bad "shared access" light. That's because the current Data Over Cable System Interface Specification (DOCSIS), 1.0, is a best-effort packet delivery system and thus has no guarantees for Quality-of-Service(QoS). Thus, the cable operator (MSO) has no way of throttling bandwidth, especially upstream bandwidth. That's why the MSOs don't like NAT and want to be able to bill their subscribers on a per IP basis. Enter DOCSIS 1.1, essentially a QoS add-on to DOCSIS 1.0 . With a DOCSIS 1.1 Cable Modem Termination System (CMTS) sitting at the MSO's cable head-end and a DOCSIS 1.1 cable modem (CM) sitting at your house, QoS can be guaranteed. That is, the MSO can both limit you to a certain upstream and downstream bandwidth as well as guarantee a minimum upstream and downstream bandwidth. So, given a DOCSIS 1.1 deployment, I see no need for the MSOs to agitate customers with this intrusive CAT proposal, since they now have a way to bill you by bandwidth. Two months ago, the first set of DOCSIS 1.1 products were certified by CableLabs. However, I don't expect DOCSIS 1.1 deployment and replacement of DOCSIS 1.0 systems to happen in large numbers until the end of 2002. Another insider note: CableLabs, the entity pushing CAT, is funded by the MSOs, but has no authority to push its proposals into implementation. Only vendors building CAT products and MSOs buying those CAT products have the power to deploy this ludicrous CAT proposal.

Exsqueeze me?

With NAT-based hubs, cable providers won't be able to see into all connected devices-making remote troubleshooting difficult-because, again, the NAT is speaking for all connected devices. It's the data communications equivalent of, "You wanna talk to her, you go through me"-except you don't even know she's there to talk.

Uhm, Cable droids, that's what my firewall IS THERE FOR!!! Damn skippy you ain't gonna see what's behind my NAT device, you and every NetBus packing, snot-nosed, loser script kiddie out there. My provider has this little numeric string that can be used to gain access to my machines if need be: My phone number.

letter I sent to the author...

Your article makes a number of assertions that reasonable
people could dispute. First is that there is anything illegal
about using NAT; Second is that what NAT is being used for is
unintentional. The gist of my complaint is that you could have
addressed the real issues without waving the red flags of "illegal"
behaviour and "unintentional" consequences.

To the first incorrect assertion: You claim that it is "illegal"
to use NAT. This has never been suggested or proven in a court of
law. It is not a "theft of service" in any event -- the service
of a single ip address to the subscriber is not being stolen from
the service provider. There remains only the single publicly
visible IP address. If there are restrictions in the SP ToS
limiting single computers to be connected, they would need to
be pretty carefully worded to rule out NAT use, and would at
worst create a ToS violation.

To the second point 8 years ago when NAT was created, there was
great concern about IP address shortage, which remains true today.
Contrary to your article, people were at the time very concerned
about the trend towards every electronic appliance in a house needing its
own IP address. NAT was one of the solutions to the problem.
Creating "sort of private, sub-network running datagrams to and
from invisible end devices" as you put it was the point of NAT.

The real issues for connectivity providers are (a) bandwidth
utilization by subscribers; (b) market penetration/revenue. (c) abuse
accountability. We can agree that a huge network hidden behind a NAT,
using a home cable connection provisioned for fractional use can use a lot
of unexpected bandwidth, but so can a spammer using a single machine, or
a teenager dedicated to downloading mp3s. So to address
issue (a) the problem is regulating traffic use in a way that offers
reasonable service to customers on low priced tiers with low provisioning.
This is a ToS issues with price/demand curve and competitive implications.
You don't have to drag NAT into the bandwidth hog issue at all.

Issue (b) is the penetration/revenue question: if one house buys the
connection and 802.11's the neighborhood, how does the installation pay
for itself? The answer is cruel: the service providers need to provide
enough value to justify subscriptions. If a shared connection using 802.11
is acceptable and worth $5/month, the service provider should provide a
supported, reliable $5/month service, not a $29.95 service.
In this case, tiered pricing (see issue (a)) may stabilize the
situation - if the neghborhood 802.11 connection is saturating the cable
connection

For abuse issue (c), the problem is that if someone drops into a private
802.11 domain and disrupts the network, who do you blame, and how do you
sanction them? The same as before, under ToS/bandwidth conditions.

In conclusion, NAT isn't a problem for which service providers need a solution.
SPs need bandwidth and abuse controls, and pricing commensurate to the
perceived value of their product in an area of rapid change. If one had
bandwith control, and the extra $4.95 month bought an additional increment
of allowed utilization, then there might be a value proposition that could
be tolerated by the public.

For the record, I had no access to ADSL or cable modem. I have a 144k
IDSL connection behind which I use NAT to attach 10 computers on my property.
I'm already paying for 24/7 use of my 144k, and I am completely guilt free.

cheers,
-dB

Unbelievable Spin

This is like the electric company charging me per light.

It's really about metered service

"At the very least, cable MSOs involved in CableHome want a counting mechanism, with parameters set by them, that specifies a maximum number of connected devices. Until then, all indicators point to DOCSIS 1.1, which includes methods to monitor bandwidth consumption (how much is used per customer) and speed (who's bursting at what rates)."

They want to protect the revenue stream from additional IP addresses. This will fail, because...

1. It will cost money
   
2. A fair percentage of the installed base will walk, especially if this means no Linux
   
3. Any software that runs on the client is open to all sorts of hacking fun. Perhaps the cable geniuses will get their software written by the MPAA masterminds who created CSS for DVD players.
   
4. The hardware manufacturers will not be pleased.
   
5. There are easier ways to make money
   

As soon as they have the ability to easily track bandwidth utilization, they will use that to drive the billing. Far better to charge per megabyte than to waste time trying to figure out how many toys the customer has and how many of them are really using the Internet. Besides, bandwidth measurements are [almost] fraud-proof, whereas this address counting stuff is a losing battle for them. They will use metered service to drive home the mother of all rate hikes, so that [among other things] AT&T can pay for @Home's sins.

Of course, metered service brings up the spam problem. Instead of the benign tolerance that most ISPs have, they will need a massive crackdown on spam unless they want all kinds of billing disputes regarding unsolicited bandwidth consumption. It's not just spam, there is also the issue of unsolicited pinging, port-scanning, and unauthorized telnet/ftp logins. If they want to measure my consumption, I intend to pick and choose which packets I pay for.

For the record, I set up my NAT-based LAN in the old days, when the cable company had no intentions of selling additional IP addresses. My continued use of this arragement is non-negotiable. I'll pull the plug before tolerating any of this CAT crap.

I wonder what these cable geniuses plan to do when they over-sell their IP allocations and need to take back the addresses. The whole concept of selling additional addresses is really wasteful. The government should have some kind of whopping tax (like 500%) on secondary residential IP addresses, so as to make the problem go away. The cable companies have never been great thinkers, they obviously need the governement to think for them.

Open letter to cable companies

Your CAT NAT replacement technology is based on the faulty assumption that you're selling a 'subscription' to the Internet. That is an extremely cable providerish way of looking at things, and precisely the reason I avoid cable (and tell my friends to as well) like the plague.

What you're selling me is a connection to the Internet. You're selling me bandwidth. That's all you're selling me. That's it. You can't care what I have on the other end of the pipe anymore than the water company can care whether or not I have a dishwasher plugged in or water a neighbors lawn.

If you're basing your pricing and bandwidth provisioning on expected usage, it's cheaper and easier to implement traffic shaping and aggregate (as opposed to burst) bandwidth limiting than it is to develop a whole set of proprietary protocols that people will just get around anyway, thereby starting a technology war (which cable companies will ultimately lose) with your customers. Then you can charge people if they want to exceed your expectations. This model is enforceable, will be seen as reasonable, and doesn't require expensive proprietary and invasive technologies to implement.

I find it kind of amusing (and scary) how so many companies want to have broken business models, call customers criminals when they don't work, and try to implement invasive technological solutions that give the service provider immense control. It's stupid and wrong, and you should know better than to have written an article advoacting such iodiocy.

Cable will never enter my home until you guys get a clue and stop trying to make me into a passive consumer instead of a happy customer.

Why I use NAT

Yeah...I have a 512K cable modem, and I can usually get around that. About the only high bandwidth I use is pulling down files from work.

Personally I like the low latency.

But, the damn cable modem gets addicted to one machine's MAC. My house is wired and if I wanted to use my notebook in the living room, it is about a 45 minute process to get the cable modem to understand that the machine behind it changed.

So, by using NAT, it is always just one machine to the cable modem...and behind the router, it is usually just only one machine on at a time anyway. I guess that makes me a thief.

Oh yeah...there is the other reason that I use NAT. Half the time if I don't keep the connection constantly going, when I go to get on, the DHCP server doesn't have any IP addresses left - so this way I don't have to worry about that. And THEY want to provide me more IP's?

So are they going to count TV Components?

If the Cable Companies want to charge for each computer, they should at least be consistant.

If have 2 televisions, but they charge me for one, does that make me a dirty thief?

What about my VCR? It has a TV reciever, so that's another conection they should charge seperately.

If I run sound out to my sterio, that's another connection. I have Dolby 5.1. Better charge extra for each speaker.

Sometime people watch TV with me. Better shake them down.

Reply to Leslie Ellis

Dear Leslie Ellis,

I just finished reading your CED article regarding NAT and cable modem service, and I would like to throw my viewpoint back at you (as countless others have likely already done, since your article was mentioned on Slashdot today).

I think you clearly and rightly stated your comparison of NAT to cable TV theft. In this argument, I would not accuse you of expressing only the point of view of the cable company, because you are also addressing some simple concepts of what is fair.

However, I think the analogy to cable TV theft is an inaccurate representation, and that it makes some assumptions as to the service being purchased by "Customer Bob" that doom him and his neighbors to being defined as abusers.

In the world of TV cable theft, sharing your subscription with your neighbor had no detrimental effect on your own service, unless you were bad at splicing and damaged your own connections; the neighbor's stolen cable would normally be identical to the service to which paying subscribers were entitled. There was no noticeable issue of bandwidth.

However, in the world of cable modem service, the subscriber is renting a connection and purchasing bandwidth from the cable company. Unless prohibited (some would say arbitrarily, or in a slippery attempt to hedge off potential revenue loss) in the service agreement, it is not dishonest for Customer Bob to share that single connection and bandwidth with his neighbors, as he is not consuming ISP resources that he would not otherwise potentially have used. Bob's sharing of his own connection and bandwidth is very different from Bob somehow jury-rigging an independent cable or DSL connection at his neighbor's house using his neighbor's own cable or phone line.

Should such a standard as CAT be implemented, I would certainly hope that the cable companies using it would reduce their rates as they applied to single computers, as they would be reducing the service provided and severely limiting the customers' options as users of that reduced service.

Please understand that I approach this issue from the viewpoint of my own NATted network, all within my own home, using a DSL connection, with an ISP who has no qualms with the full usage by customers of their paid service.

Thank you for your presentation of this issue, and thank you for your attention. This reply is also being posted to the Slashdot thread where your article's URL appeared this morning.

David A. Mason
[e-mail]
Network Administrator
Center for Nonproliferation Studies
Monterey Institute for International Studies
http://cns.miis.edu/

Classic Absurdity

The hypothetical numbers this articles uses are priceless examples of industrial chicanery: "Let's say one in 10." No, let's not say one in ten. Let's be realistic instead.

Okay, I'll bite... WHY IS EVERYTHING SO $$$$????

I'll state this up front. I am not a networking expert, network programmer, or even a guest on The Battle of the Network Stars. *

What I am, tho, is someone who has been on this scene since '81. I remember the advent of fiber optic lines, and the promise of immense bandwidth Some Day, maybe in ten years...

In the mid Eighties, the talk was of laying the mighty fiber trucklines through major cities. I remember the day that downtown Chicago got it's first, GASP, fiber line down the middle of State Street (I think).

Speculation was rife about fiber to the house. Of course, the holdup was that it would cost roughly 500 -- that's five hundred -- dollars per household in '86 dollars to fiber the country up. No one wanted to shoulder that expense. No company wanted to do it -- the profit model couldn't be made to show it working as a business proposition.

I remember debate about letting it become a governemnt service, like water, or a regulated utility. Let taxpayer cash fund the structure of the net; the benefit would be laser beams for all, forever and ever, amen.

Well, the '80's marked the ascendency of the capitalist as a god, and business was our new religion. Public anything was communism, anti-profit, and besides, private biz could do it cheaper, faster, and without the bureaucracy.

We went ahead. Modems reached dizzying speeds of 28.8k, 56k... and the businesses who would pay the premium got T1/T3 lines. No fiber ever reached the citizen, except for a few private projects.

Curiously, as hardware became commodity priced, switches, routers, and their humongous bigger brothers became a cash cow for the companies that made them. Shakeouts occured, companies merged, profits stayed pretty high. Small ISPs couldn't compete with ever-bigger competitors, and died.

Here we are. 2001. And we still are using modems over 1890 Bell wire. And the phone bills still keep climbing, tho why is a mystery...

Here's the bad math. If we had fiber, say, 50 million homes and apartment complexes in the late '80's at guvmint expense, the total would have been:

$ 500.00 US * 50,000,000
= 25,000,000,000 bucks.

Let's adjust it a bit by assuming:

1. That even tho the per home cost of equipment should have dropped with that scale of manufacturing, the cost would have stayed about the same due to the enormous physical work necessary to lay glass pipes over entire cities and burbs.
2. That inflation would make it, say for the fun of it, about $50,000,000,000 US in today's dollars.
3. The project would have taken, say, fifteen years.

Okay then. Per annum, 3 1/3 billion a year to fiber every one of fifty million homes. Hell, there weren't even that many PC's yet, so I'm overshooting.

For 3.33 bil a year, we could have replaced the phone system with a packet-switched digital model. Had video phones. Cable TV with thousands of channels. Video cameras on neighborhood networks, so that everyone could see what was going on around town. Cheap ways for bizes to connect with each other.

The upkeep cost of the system would be in the billions every year, not to mention the cost of fibering new customers all the time. Obsolesence would be a major pain, but we'd get by by standardizing on newer equipment using old standards, and do Good Enough overall.

Okay, so by today, we would all be connected by laser, running at rather interesting speeds. The equipment would become obsolete, but mostly at the neighborhood switch level and higher -- the customer setup would become commodity priced pretty quickly.

What do we have instead?

Okay, let's just say we have, um ten million cable modem subscribers now. Each pays $50 US a month.

That's 500,000,000 mil a month. For 128, 256, whatever, bandwidth.

That multiplied by 12 is $6,000,000,000 - six billion a year we shell out.

And under that biz model, there is no profit incentive, ever, to fiber our homes.

Think about it. Twiddle the numbers around. Don't forget businesses pay far higher prices for their connectivity as well. I left out the modem users and what THEY pay to the phone companies and ISPs.

How much has the free market cost us, and what have we gotten for it?

Shangri La: we had spent 3 billion or maybe more a year, in today's bucks, over a long period of time, to fiber everyone. Yay us.

Too expensive? What about all that Dark Fiber laid down in the last few years? Why innanameofGawd is everything so expensive when it wasn't all that hard to drop that fiber?

Reality: the mega-companies that are buying up and/or creating bandwidth are never going to fiber us, not at prices we can afford. And they also are becoming the same companies that additionally own the entertainment giants, so they want to monitor our net usage to make sure we don't steal their "property". They don't want us sharing bandwidth, or using too much bandwidth, because their profit models would be ruined.

That's business? A small group of rather wealthy companies get it all their own way, and we gave up fiber for this? 'Cause biz was better and cheaper?

I've watched the Great Experiment of the dereg of the telcos (now remerging), of the degreg of media, and I see that we are getting absolutely robbed, of not only our cash, but what the future should have been.

Hell, not the future, the PRESENT.

* Battle of the Network Stars was a really, really bad show in the '70's. Forget I mentioned it.

Two Marketing Fallacies

What's the value of the stolen goods? Revenues associated with additional IP addresses, for one.

The author is assuming that, if the users weren't "stealing" (rhetoric 101: apply perjorative terms to things you don't like) bandwidth, they would be buying it for whatever the seller cares to charge. Doesn't work that way. There are many things that I get free (the vast majority of Webpages I look at, for instance) that I wouldn't be willing to pay anything at all for.

And certainly, no one had fully imagined that the resources shared by a single, wirelessly-networked residence would also be shared among other devices, at other residences, within 300 feet.

This is simply a failure of market research. The cable providers assumed that the "typical" user would look at graphics-heavy news sites (cnn.com or suchlike) and send a bit of e-mail, and that would be it. When the "typical" household has Mom watching movie trailers, Dad looking at pr0n, and the kids swapping MP3s, it's no wonder that the pipe gets jammed. Instead of saying "Oops!" and figuring out how to deal with it, they want to go back and cram the usage pattern into their marketing model.

Basically, the whole thing is a marketing error, compounded by abysmal ignorance of things Internet on the part of the cable providers. There are any number of technical fixes that don't involve dealing with anything behind the firewall. Unfortunately, this is "too much like work" for the cable providers.

Re:More Slashdot Sensationalism at Work

Except that their solution, like CSS or any other "anti-piracy" solution, is not going to punish merely the offenders. It is also quite likely to catch a lot of innocent people in its claws. The article itself seems to have a very negative view on NAT, which indicates to me that they think plain-old-honest-sensible address translation is a criminal behavior if it deprives them of revenue. Serious questions need to be asked and answered before we who are technologically savvy allow this sort of thing to become widespread (if we even have a say in the matter).

Most importantly, does this portend a future in which NAT or ip chains are deemed a violation of our user agreements? If so, I would have never signed up (well, maybe I would have, but given the criminal penalties provisioned in the DMCA and that NAT could be deemed a circumvention device if the cable company only approves this proposed CAT nonsense...). So the real question is, would you like to occupy the cell next to Dmitry simply for having a firewall and a class C network?