Even with hardware crypto devices, the key can often be found in memory or on disk. The card has to get the key from somewhere. Only the super expensive FIPS compliant cards have on-board key storage (key your private key safely offline on a floppy in a safe) and generation (so noone ever knows the private key.)
A top of the line opteron is good for ~1000 RSA key exchanges per second. With "cheap" hardware cards (where cheap is in the 500$ range), that number goes from 10k to 28k per second. FIPS certified hardware is a lot slower, but still much faster than a CPU -- ~4k with a $4k card. (And yes, these are all fully supported (openssl) linux hardware solutions.)
No it doesn't. I run WSUS on a cheap Dell 4600 running Windows 2000. The problem with WSUS is the large number of things machines MUST ask Micro$oft DIRECTLY to even know about, much less download. And if your machines are not asking Microsoft about updates for the update service AND Windows Genuine Advantage(tm), then there are even more updates it'll never know about or be able to install if it does. For example, I have a few XP machines that don't even know IE7 exists because they don't have up-to-date WGA crap installed. (And yes, they are legal/genuine Dell machines. Nobody has pointed IE to windowsupdate. Btw, that bypasses WSUS.)
DSLAM contention statistics are reasonably well published here in the UK.
We aren't talking about the UK, genious. Most non-US ISPs have been metering services for a long time. We've never had metered connections in the US; US consumers aren't going to accept it without an argument.
Some customers only want to make light use of the network whilst others want to make very heavy use... It's not abuse if you are using the service you were sold in the way the ISP intended... Provide a tiered pricing strategy...
They already have tiered options from $29.95 "Lite" (1.5M/128k) up to $89.95 "Turbo" (8M/768k?). They just don't want to provide the service they have sold to us. The "intent" of the ISP is hard to read -- and they change their minds all the time. If someone opts for the $80/month "up to 8M" connection, they will assume they can actually use it. However, if they use their 8M for any measurable amount of time, they're seen as abusing the network, "disruptive to other users experience." There have been a number of reports of the ISPs dropping customers like that.
No, Time Warner simply doesn't want to support what they have astoundingly over-sold. They don't want to cut off paying customers because that's money gone. They don't want to throttle these users to limit the "disruption", which their user agreement has always allowed them to do, because that would cost a lot of money to implement -- new hardware, or expensive upgrades to their existing hardware -- without generating a single penny of revenue. They've opted for the "free" solution that lines their houses with cash... they can stup monitoring with a few simple configuration changes on their existing hardware (rmon, netflow, etc.) (processing the output is 99% of the work which I'm pretty sure they'll still have messed up a year from now.)
[PS: I've collected as-aggregate netflow stats from what is a gnat compared to TW. And that was over 100MB per hour.]
If they cared about the network, they would do what many non-US ISPs do... throttle your connection. Instead, they gladly let people continue to "abuse the network". The only difference is TW will now profit from those users by increasing their bill. The network will still "suffer" -- it's still unproven that there really is as much of bandwidth crunch as they want us to beleive; and while it's hard to say on the grand scale, I've not noticed "slow internet" anywhere I've been.
History has already shown they are far more interested in pocketing the cash than spending potentially billions on network upgrades.
Bandwidth monitoring per user is much more involved than you think it is. But by all means, keep your head in the sand until they incorrectly bill you.
As for "light users", you just don't understand how ISPs work -- and have worked for decades... they oversubscribe everything by at least 10:1. They actually prefer customers that pay the bill but never use the service. When you oversubscribe your network by a factor of 20+, you have to have those low use customers. Back in the days of dialup, an ISP needed 3-5 customers PER LINE just to pay for the line; obviously 5 people cannot use the same modem at the same time. Today's DSL and cable modems hang dozens to hundreds of people off the same link at the same time... hundreds of dsl modems into one DS3/OC-3 feed DSLAM, hundreds of cable modems on a single 256-QAM (38.8Mbps) channel. At best a half dozen can use their full connection speed at the same time. There's no way the entire network can support even a fraction of it's users using their full connection. Only your ISP can tell you how congested a node actually is -- but they won't, because it's BAD... I worked for an ISP once that sold 1.5M DSL connections from DSLAMs feed from 768k links, one DSLAM with 192 IDSL ports feed from a single T1, etc. Do they tell the customer that? Of course not.
Those grandmothers most likely have the lowest service teir (19.95/mo for 1.5M/128k service.) And if my circle is any indication, it's there because one of the children had it installed -- and is paying for it. I've worked for several ISPs; I know first hand how much of a mess this will be when first turned on. The first overage bill someone gets will very likely be the last time you see them as a customer. They take their business, including those grandmothers, elsewhere. This being cable, they stand to lose much more than a "bandwidth pig"... cable tv, cable phone, and cable modem -- potentially $100+ per month gone, forever.
But there's the rub... TW isn't doing this in the name of network health. They're doing it to generate revenue. See, they aren't slowing your network down. They're increasing your bill. And I have money on them screwing this up. Accurate bandwidth monitoring is not as hard as it used to be, but it's still a long way from simple.
There are no requirements one way or the other w.r.t. analog signalling on cable. The only reason analog cable still exists is generally money: in the absence of an analog teir they cannot charge a premium for digital cable. And in many places, that would make digital cable a price regulated service. In fact, the FCC has helped to push everything to 100% digital... a digital only cable system doesn't have to convert broadcast stations back to analog. And cable system that still has an analog teir will have to convert their OTA digital stations to analog for their analog teir -- which is usually the cheapest ("almost free") service teir.
It's about money first, control second. Actually, that pretty much sums it up nicely.
CC installs really are that simple. They don't want you doing it yourself because a) they can charge you to have a tech do it, b) they want to ensure a licensed device is being used (not that you could build a functional non-licensed device), and c) they want the process to be as big a pain in the ass as possible. However, they will happily let you walk up and get one of there boxes without any of that.
Personally, I'd like to see the FCC go one step further... outright BAN cable operators from providing hardware. While they are now required to use the same cableCARD technology as CE manufacturers, they are still building and using proprietary, custom built -- and NOT Cable Labs tested and certified -- hardware. Make them use the same hardware they make us use and all this bull will stop. But this will never happen because the FCC is too spineless and the cable co's make far too much money renting boxes to let go of it.
It's a matter of "spirit" vs. "letter"... The FCC is mad, but there's nothing they can do about it. (and they should've seen it coming.) The cable companies have implemented a seperable security system as per the FCC mandate. It's not their fault almost every current Cable Labs certified receiver is unidirectional. (actually it is, since at the time cable labs had no other certification track -- UDCP was it.)
Btw, all cableCARDs are bidirectional. They are, however, at the mercy of the host into which they have been placed.
All the boxes and most cable-cards are staged at the warehouse prior to going to the customer.
That is indeed an unique operation. From all the reports I've heard, nobody likes doing that because "unpairing a card is a pain in the ass." Which, I'll add, is completely opposite the CC1.0 and CC2.0 specs; the card will (re)pair with any host that asks. It'll only pair with one host at a time, 'tho. I think they were referring to their headend systems being difficult to add, remove, and then re-add a CC. (which might be true, that's not part of the spec:-))
the complication also comes that I don't believe the TIVO gives any input that the card is actually downloading code. With the Moto or SA settop boxes you'll see a countdown on the LED's as it downloads, so you can at least tell it's doing something.
Correct. Because it has no way of knowing. Maybe you should ask Moto/SA how they're showing that status. (they're bidirectional, so they can ASK the headend how much has been sent.) Motorola cards get their firmware via the OOB modem, so the host has no idea what the data means -- assuming the host is even involved. SA gets it's firmware inband, directly from a tuned digital channel; so the host has no idea at all what the CC is doing. This is also why you cannot just shove both cards in at the same time. If a SA card needs to do a fimrware update, it'll take over the system to do it -- it needs "interactive" control of the tuner to tune to (or search for) the firmware channel.
Technically, you can insert both cards at the same time. Sometimes it'll work. A lot of times it won't. Don't blame Tivo, Inc. because you won't follow the (provided) instructions. One would think an installer would be happy to be handed such clear, simple instructions given the hundreds of different certified devices and problems they may face. Also, it's not as if you have to wait an hour before insterting the second card. It only takes a few seconds for tivo recognize the card and take you to the CC screen. From there select the card, and then the host info screen. The tivo can only present as much diag info as the card will give it -- notice all the "RESERVED" lines on the CC screen. The time consuming process is the firmware upgrade; it should only take a few minutes, but a lot of times it cannot find the firmware (headend isn't broadcasting it?) [go look at the tuner diag screen to see what it's doing. if it's searching for a channel, get a new card:-)]
... they pretty much said that after 7/07 all systems must use cablecards.
You've leaving out almost a decade of the story. (and I'm likely to mis-remember most of it:-)) The FCC demanded a seperable security system in order to open the cable TV market to consumer electronics manufacturers over a decade ago. At which point Cable Labs, et. al., drug their feet for years coming up with the 1.0 cable card specs and certification process. The FCC had to set an implementation deadline just get them to do that. They would have spent several more decades puttering around if a gun hadn't put to their head. Then they asked for a delay of the "integration ban" to finalize the second revision of the CC spec -- multi-stream support. The FCC, being the spinless p***ies that they are, agreed. And once again, as soon as the deadline approached, they wanted more time to work out yet an other set of specs. This time DFAST/DCAS, a software version of the CC so they wouldn't have to change out all their hardware. (and make the CE industry have to redesign everything.) Again, the FCC agrees. Yet again, when the deadline started approaching (07/2007), they wanted more time... this time it's OCAP, and work out "issues" with CC deployment. However, this time the FCC had had enough and they're being made to eat their tainted dog food.
But your right, it's more a political process than a technological process. The FCC is messing in cable's sandbox, and they don't like it. So, they're slow to develop standards, slow to deploy new standards, and generally do everything to make it a huge mess. Most people don't realize cable cards have been around for years; many more years than people have been hearing about them. Only recently (2 years +/-) have they become an advertized option. And yet, very, VERY few cable techs know much about them. Had the cable industry been forced to start using this stuff about 8 years ago, then we wouldn't have the mess we have today. (we'd've had a smaller version of the same mess back then.)
This encryption helps to prevent unauthorized access to the digital signal. The most obvious (without getting into conspiracys or opinions on greed and whatnot) reasoning for encrypting a channel is so that little johnny doesn't stumble across hardcore sex in the clear with his QAM tuner TV.
Simple answer: Channels are encrypted in order to control who receives them - period. In the old analog days, this was done with a notch filter physically attaced to the line. Those are expensive and time consuming to install, remove, and repair. With digital cable, who receives a channel is as simple as giving them the key to decode it -- which changes several times per minute.
The issue with Tivo's is entirely that of impatient installers. The instructions are very simple and easy to follow. And if you follow them, they work. Of course, the tivo falls into the same boat as an other multi-card devices... improper card setup where each card is authorized for a different set of channels. That's not tivo's fault; that's your fault over there at the headend.
OCAP is an application environment. It's like saying your STB has to run linux vs. windows ce vs. beos... The CE manaufacturer provides the hardware and an "OS" to run applications provided by the MSO. The UI, features, and functionality of the STB depends entirely on what the MSO sends it. So, no matter who makes the box, they'll look exactly the same until it's moved to a different cable network and given a different set of applications to run. And there's the blanket "no" to OCAP: zero consumer choice. And with specs being what they are, and cable companies being who they are, I seriously doubt OCAP will be any better. It'll work just fine on their hardware, but random issues will exist on customer owned CE boxes. We see it today despite extensive testing done by Cable Labs in the certification process of both hosts and cards.
There's plenty of capacity in 1GHz -- anywhere from 300 to 1200 channels depending on the amount of HD. Step one, dump the GD analog channels. Most cable operators are still wasting -- yes, WASTING -- over half their spectrum with analog channels. But what about all those people with basic/standard cable? *cough*screw em*cough* Basic cable is ~12 channels, and they pay almost nothing for them. Standard cable is an additional ~50 channels, and people pay ~40-50$ for them. Rent them a STB, which is what most cable companies want anyway; in just over a year, they won't have any other choice... buy a new TV, or live with a "box." I don't know of any cable operators willing to risk losing any customers by dropping analog channels; that will change in '09 when consumers have no other analog TV choice. And as soon as MSOs have no analog tier, they'll no longer be allowed to add a surcharge for digital cable -- which is even more lost revenue. Yeah, I know "greed"... walks like a duck, talks like a duck...
If you want to post a useful link, go find the specifications document for the supposedly "open standard" developed by SA and TWC for SDV. I see references to it all over the place (and have for years.) I can even find the document ID. But even the mighty google has not spotted the actual pdf.
Do you have to pay the phone company for a special box to use their phone lines? do you have to pay the power company for special box? do you pay the water company for a special box? no you dont.
You used to... a few decades ago, nobody owned their own phones. You don't directly pay the power company for and "special box", but included in your bill are costs associated with the meter on your house and the trained squirrels reading that meter. And YES, I DO pay the city (water company) for a "special box" -- there's a direct line-item for the water meter.
The whole point is, you don't need the cable company's box. They KNOW that. And they see the revenues they've been collecting on rented boxes over the past decades withering away. 90% of the issues MSOs have today with capacity are directly attributable to all the d***ed OnDemand crap. DVRs mean you don't need that crap... I've used tivos for a decade; I've never wanted or needed any ondemand content. I can watch whatever I want, whenever I want, because I have hundred (thousands?) of hours of previously recorded content. I schedule things from the future, not from the past.
You seem to be unaware of the economics of game consoles... almost all of them are sold at a loss. They games are where the money is made. But yes, the buy-it costs of cable and sat hardware are insanely over the top.
Cable card installer? It's not about putting a card in a slot. It's all about making sure you're putting the card in the certified device you said it's going in. And taking more of your money. Btw, I've yet to meet a TW tech who's had any training at all on cable cards... plug it in, call a number, read 'em back the numbers; it either works or it doesn't. (most of the time it doesn't) They have zero issues with their own hardware -- a) it's bidirectional, and b) it doesn't have to cable labs certified; that's right, they are free to use what ever non-standard, proprietary crap they want.
Because we're the USA. The whole world is expected, and required, to bend over and do our bidding. Or that's how US industries think and act.
Think about it. If the US adopted DVB-C/DVB-T, then the rest of the world -- already manufacturing equipment to those standards -- would have an advantage over our own hometown businesses. This, of course, ignores the fact that almost none of this crap is even remotely "made in america".
Here we go again... The cableCARD specification has been bi-directional since day one. Put another way, ALL CABLE CARDS ARE BIDIRECTIONAL; and they always have been. It's the host (the thing you plug the card into) that's only licensed for one-way communications. There wasn't an two-way license or standards track until very recently -- and it's f'ing OCAP.
CableLabs, the cable industry's research and development arm, which Roberts leads, was to announce Monday that its OpenCable platform, first developed in 1997, will now be branded as "tru2way."
Is it too much to ask people to actually read the article before asking stupid questions that are already answered? Oh right, this is slashdot.
It's "OCAP" -- basically a JAVA engine running what ever your cable provider sends it. And the CE (Comsumer Electronics) industry doesn't like it. Nor is the FCC particularly fond of it either since the cable industry has been constantly reinventing this wheel as a means to push back the "integration ban" -- which the FCC finally stopped doing... cable providers have had to use "seperable security systems" (a.k.a. cableCARDs) sine July 1.
That's because Windows XP doesn't support WPA2 without the hotfix. The wireless card's driver and management application(s) might even if Windows does. Just keep the hotfix on a thumb drive, and always be prepared.
Most modern wireless routers have barely enough flash and RAM to do what they're supposed to do. Replacing that firmware with custom firmware that a) continues to function as the router, and b) can scan and attack other routers is simply not going to fit in 2M flash/8M RAM. Plus, there are a lot of different hardware platforms around. Most of them are not Linux based, so the code and details of their operation aren't so easily replicated.
Sure, it's an interesting THEORY, but until they show a working model, it's simply not a threat. It's an academic exercise.
Oh, and the article says even WEP won't protect you. This is also very much in error. Show me even ONE cheap little home wireless router with the CPU power to attack WEP. Sure almost any modern computer can recover a WEP key in just a few minutes, however they have two things few wireless routers have... very fast cpus, and a loads of memory (RAM, hard drive, etc) in which to hold the thousands (or millions) of packets necessary for the process. You'd be far more successful driving around the city with a laptop running airsnort.
That CD doesn't do anything useful. It's a free-standing ethernet device. It doesn't require drivers or proprietary setup programs. Point a browser at it and go.
Some older netgear and 3com hardware wouldn't work until you completed their lame "setup". It intercepted all web traffic and blocked everything else. And wireless wasn't enabled until you intentionally turned it on. Today's Linksys crap is designed for complete brainless idiots -- no shit, some of them cannot even figure out how to plug it up (power maybe, ethernet no.)
I've gotta ask: Who the h*** builds a CO without battery and generator backup? (The CO itself should already be battery powered... -48VDC) I've never noticed in other states, but in NC I've never seen a CO without a generator sitting outside.
Diesel vapor pressure is too low to be anywhere near as dangerous as you're making it out to be. Unless it's under pressure or boiling, it won't burn. Get yourself a cup of diesel, a match (lots of them), and a blow torch and go see just how hard it is to get it to burn (and keep burning on it's own.) Likewise, it doesn't evaporate quickly so there's little issue of fumes.
The big problems with generators are their exhaust and noise. Noise being what most people take issue with -- i.e. why many HOA's ban them (in addition to being ugly.)
You make it sound like it "goes bad" in just a few days. That's not at all true. It takes months, if not years. And it has nothing to do with microbes. It chemically degrades over time -- the result of modern high volume processing (see also: "cracking".) There are numerous commercial stabalizers that can keep fuels viable for decades. (some better than others.)
So you don't think that the unauthorized access to the secretary's office with a stolen key would be charged as breaking and entering?
Nope. That's unlawful entry and/or trespass. If they have a key, it's not BREAKING and entering. Picking the lock, jimmying the door, etc. is breaking and entering.
Slashdot Mirror
Even with hardware crypto devices, the key can often be found in memory or on disk. The card has to get the key from somewhere. Only the super expensive FIPS compliant cards have on-board key storage (key your private key safely offline on a floppy in a safe) and generation (so noone ever knows the private key.)
A top of the line opteron is good for ~1000 RSA key exchanges per second. With "cheap" hardware cards (where cheap is in the 500$ range), that number goes from 10k to 28k per second. FIPS certified hardware is a lot slower, but still much faster than a CPU -- ~4k with a $4k card. (And yes, these are all fully supported (openssl) linux hardware solutions.)
No it doesn't. I run WSUS on a cheap Dell 4600 running Windows 2000. The problem with WSUS is the large number of things machines MUST ask Micro$oft DIRECTLY to even know about, much less download. And if your machines are not asking Microsoft about updates for the update service AND Windows Genuine Advantage(tm), then there are even more updates it'll never know about or be able to install if it does. For example, I have a few XP machines that don't even know IE7 exists because they don't have up-to-date WGA crap installed. (And yes, they are legal/genuine Dell machines. Nobody has pointed IE to windowsupdate. Btw, that bypasses WSUS.)
No, Time Warner simply doesn't want to support what they have astoundingly over-sold. They don't want to cut off paying customers because that's money gone. They don't want to throttle these users to limit the "disruption", which their user agreement has always allowed them to do, because that would cost a lot of money to implement -- new hardware, or expensive upgrades to their existing hardware -- without generating a single penny of revenue. They've opted for the "free" solution that lines their houses with cash... they can stup monitoring with a few simple configuration changes on their existing hardware (rmon, netflow, etc.) (processing the output is 99% of the work which I'm pretty sure they'll still have messed up a year from now.)
[PS: I've collected as-aggregate netflow stats from what is a gnat compared to TW. And that was over 100MB per hour.]
By your reasoning, everything is conjecture.
If they cared about the network, they would do what many non-US ISPs do... throttle your connection. Instead, they gladly let people continue to "abuse the network". The only difference is TW will now profit from those users by increasing their bill. The network will still "suffer" -- it's still unproven that there really is as much of bandwidth crunch as they want us to beleive; and while it's hard to say on the grand scale, I've not noticed "slow internet" anywhere I've been.
History has already shown they are far more interested in pocketing the cash than spending potentially billions on network upgrades.
Bandwidth monitoring per user is much more involved than you think it is. But by all means, keep your head in the sand until they incorrectly bill you.
As for "light users", you just don't understand how ISPs work -- and have worked for decades... they oversubscribe everything by at least 10:1. They actually prefer customers that pay the bill but never use the service. When you oversubscribe your network by a factor of 20+, you have to have those low use customers. Back in the days of dialup, an ISP needed 3-5 customers PER LINE just to pay for the line; obviously 5 people cannot use the same modem at the same time. Today's DSL and cable modems hang dozens to hundreds of people off the same link at the same time... hundreds of dsl modems into one DS3/OC-3 feed DSLAM, hundreds of cable modems on a single 256-QAM (38.8Mbps) channel. At best a half dozen can use their full connection speed at the same time. There's no way the entire network can support even a fraction of it's users using their full connection. Only your ISP can tell you how congested a node actually is -- but they won't, because it's BAD... I worked for an ISP once that sold 1.5M DSL connections from DSLAMs feed from 768k links, one DSLAM with 192 IDSL ports feed from a single T1, etc. Do they tell the customer that? Of course not.
Those grandmothers most likely have the lowest service teir (19.95/mo for 1.5M/128k service.) And if my circle is any indication, it's there because one of the children had it installed -- and is paying for it. I've worked for several ISPs; I know first hand how much of a mess this will be when first turned on. The first overage bill someone gets will very likely be the last time you see them as a customer. They take their business, including those grandmothers, elsewhere. This being cable, they stand to lose much more than a "bandwidth pig"... cable tv, cable phone, and cable modem -- potentially $100+ per month gone, forever.
But there's the rub... TW isn't doing this in the name of network health. They're doing it to generate revenue. See, they aren't slowing your network down. They're increasing your bill. And I have money on them screwing this up. Accurate bandwidth monitoring is not as hard as it used to be, but it's still a long way from simple.
There are no requirements one way or the other w.r.t. analog signalling on cable. The only reason analog cable still exists is generally money: in the absence of an analog teir they cannot charge a premium for digital cable. And in many places, that would make digital cable a price regulated service. In fact, the FCC has helped to push everything to 100% digital... a digital only cable system doesn't have to convert broadcast stations back to analog. And cable system that still has an analog teir will have to convert their OTA digital stations to analog for their analog teir -- which is usually the cheapest ("almost free") service teir.
Simple. Buy one of the new Linksys Draft-N routers and put it in 40MHz mode. It'll stomp all over them.
It's about money first, control second. Actually, that pretty much sums it up nicely.
CC installs really are that simple. They don't want you doing it yourself because a) they can charge you to have a tech do it, b) they want to ensure a licensed device is being used (not that you could build a functional non-licensed device), and c) they want the process to be as big a pain in the ass as possible. However, they will happily let you walk up and get one of there boxes without any of that.
Personally, I'd like to see the FCC go one step further... outright BAN cable operators from providing hardware. While they are now required to use the same cableCARD technology as CE manufacturers, they are still building and using proprietary, custom built -- and NOT Cable Labs tested and certified -- hardware. Make them use the same hardware they make us use and all this bull will stop. But this will never happen because the FCC is too spineless and the cable co's make far too much money renting boxes to let go of it.
It's a matter of "spirit" vs. "letter"... The FCC is mad, but there's nothing they can do about it. (and they should've seen it coming.) The cable companies have implemented a seperable security system as per the FCC mandate. It's not their fault almost every current Cable Labs certified receiver is unidirectional. (actually it is, since at the time cable labs had no other certification track -- UDCP was it.)
Btw, all cableCARDs are bidirectional. They are, however, at the mercy of the host into which they have been placed.
Technically, you can insert both cards at the same time. Sometimes it'll work. A lot of times it won't. Don't blame Tivo, Inc. because you won't follow the (provided) instructions. One would think an installer would be happy to be handed such clear, simple instructions given the hundreds of different certified devices and problems they may face. Also, it's not as if you have to wait an hour before insterting the second card. It only takes a few seconds for tivo recognize the card and take you to the CC screen. From there select the card, and then the host info screen. The tivo can only present as much diag info as the card will give it -- notice all the "RESERVED" lines on the CC screen. The time consuming process is the firmware upgrade; it should only take a few minutes, but a lot of times it cannot find the firmware (headend isn't broadcasting it?) [go look at the tuner diag screen to see what it's doing. if it's searching for a channel, get a new card
But your right, it's more a political process than a technological process. The FCC is messing in cable's sandbox, and they don't like it. So, they're slow to develop standards, slow to deploy new standards, and generally do everything to make it a huge mess. Most people don't realize cable cards have been around for years; many more years than people have been hearing about them. Only recently (2 years +/-) have they become an advertized option. And yet, very, VERY few cable techs know much about them. Had the cable industry been forced to start using this stuff about 8 years ago, then we wouldn't have the mess we have today. (we'd've had a smaller version of the same mess back then.)
The issue with Tivo's is entirely that of impatient installers. The instructions are very simple and easy to follow. And if you follow them, they work. Of course, the tivo falls into the same boat as an other multi-card devices... improper card setup where each card is authorized for a different set of channels. That's not tivo's fault; that's your fault over there at the headend.
OCAP is an application environment. It's like saying your STB has to run linux vs. windows ce vs. beos... The CE manaufacturer provides the hardware and an "OS" to run applications provided by the MSO. The UI, features, and functionality of the STB depends entirely on what the MSO sends it. So, no matter who makes the box, they'll look exactly the same until it's moved to a different cable network and given a different set of applications to run. And there's the blanket "no" to OCAP: zero consumer choice. And with specs being what they are, and cable companies being who they are, I seriously doubt OCAP will be any better. It'll work just fine on their hardware, but random issues will exist on customer owned CE boxes. We see it today despite extensive testing done by Cable Labs in the certification process of both hosts and cards.
There's plenty of capacity in 1GHz -- anywhere from 300 to 1200 channels depending on the amount of HD. Step one, dump the GD analog channels. Most cable operators are still wasting -- yes, WASTING -- over half their spectrum with analog channels. But what about all those people with basic/standard cable? *cough*screw em*cough* Basic cable is ~12 channels, and they pay almost nothing for them. Standard cable is an additional ~50 channels, and people pay ~40-50$ for them. Rent them a STB, which is what most cable companies want anyway; in just over a year, they won't have any other choice... buy a new TV, or live with a "box." I don't know of any cable operators willing to risk losing any customers by dropping analog channels; that will change in '09 when consumers have no other analog TV choice. And as soon as MSOs have no analog tier, they'll no longer be allowed to add a surcharge for digital cable -- which is even more lost revenue. Yeah, I know "greed"... walks like a duck, talks like a duck...
If you want to post a useful link, go find the specifications document for the supposedly "open standard" developed by SA and TWC for SDV. I see references to it all over the place (and have for years.) I can even find the document ID. But even the mighty google has not spotted the actual pdf.
The whole point is, you don't need the cable company's box. They KNOW that. And they see the revenues they've been collecting on rented boxes over the past decades withering away. 90% of the issues MSOs have today with capacity are directly attributable to all the d***ed OnDemand crap. DVRs mean you don't need that crap... I've used tivos for a decade; I've never wanted or needed any ondemand content. I can watch whatever I want, whenever I want, because I have hundred (thousands?) of hours of previously recorded content. I schedule things from the future, not from the past.
You seem to be unaware of the economics of game consoles... almost all of them are sold at a loss. They games are where the money is made. But yes, the buy-it costs of cable and sat hardware are insanely over the top.
Cable card installer? It's not about putting a card in a slot. It's all about making sure you're putting the card in the certified device you said it's going in. And taking more of your money. Btw, I've yet to meet a TW tech who's had any training at all on cable cards... plug it in, call a number, read 'em back the numbers; it either works or it doesn't. (most of the time it doesn't) They have zero issues with their own hardware -- a) it's bidirectional, and b) it doesn't have to cable labs certified; that's right, they are free to use what ever non-standard, proprietary crap they want.
Because we're the USA. The whole world is expected, and required, to bend over and do our bidding. Or that's how US industries think and act.
Think about it. If the US adopted DVB-C/DVB-T, then the rest of the world -- already manufacturing equipment to those standards -- would have an advantage over our own hometown businesses. This, of course, ignores the fact that almost none of this crap is even remotely "made in america".
Here we go again... The cableCARD specification has been bi-directional since day one. Put another way, ALL CABLE CARDS ARE BIDIRECTIONAL; and they always have been. It's the host (the thing you plug the card into) that's only licensed for one-way communications. There wasn't an two-way license or standards track until very recently -- and it's f'ing OCAP.
It's "OCAP" -- basically a JAVA engine running what ever your cable provider sends it. And the CE (Comsumer Electronics) industry doesn't like it. Nor is the FCC particularly fond of it either since the cable industry has been constantly reinventing this wheel as a means to push back the "integration ban" -- which the FCC finally stopped doing... cable providers have had to use "seperable security systems" (a.k.a. cableCARDs) sine July 1.
That's because Windows XP doesn't support WPA2 without the hotfix. The wireless card's driver and management application(s) might even if Windows does. Just keep the hotfix on a thumb drive, and always be prepared.
Most modern wireless routers have barely enough flash and RAM to do what they're supposed to do. Replacing that firmware with custom firmware that a) continues to function as the router, and b) can scan and attack other routers is simply not going to fit in 2M flash/8M RAM. Plus, there are a lot of different hardware platforms around. Most of them are not Linux based, so the code and details of their operation aren't so easily replicated.
Sure, it's an interesting THEORY , but until they show a working model, it's simply not a threat. It's an academic exercise.
Oh, and the article says even WEP won't protect you. This is also very much in error. Show me even ONE cheap little home wireless router with the CPU power to attack WEP. Sure almost any modern computer can recover a WEP key in just a few minutes, however they have two things few wireless routers have... very fast cpus, and a loads of memory (RAM, hard drive, etc) in which to hold the thousands (or millions) of packets necessary for the process. You'd be far more successful driving around the city with a laptop running airsnort.
That CD doesn't do anything useful. It's a free-standing ethernet device. It doesn't require drivers or proprietary setup programs. Point a browser at it and go.
Some older netgear and 3com hardware wouldn't work until you completed their lame "setup". It intercepted all web traffic and blocked everything else. And wireless wasn't enabled until you intentionally turned it on. Today's Linksys crap is designed for complete brainless idiots -- no shit, some of them cannot even figure out how to plug it up (power maybe, ethernet no.)
I've gotta ask: Who the h*** builds a CO without battery and generator backup? (The CO itself should already be battery powered... -48VDC) I've never noticed in other states, but in NC I've never seen a CO without a generator sitting outside.
Diesel vapor pressure is too low to be anywhere near as dangerous as you're making it out to be. Unless it's under pressure or boiling, it won't burn. Get yourself a cup of diesel, a match (lots of them), and a blow torch and go see just how hard it is to get it to burn (and keep burning on it's own.) Likewise, it doesn't evaporate quickly so there's little issue of fumes.
The big problems with generators are their exhaust and noise. Noise being what most people take issue with -- i.e. why many HOA's ban them (in addition to being ugly.)
You make it sound like it "goes bad" in just a few days. That's not at all true. It takes months, if not years. And it has nothing to do with microbes. It chemically degrades over time -- the result of modern high volume processing (see also: "cracking".) There are numerous commercial stabalizers that can keep fuels viable for decades. (some better than others.)
Their corp office is in Texas.