They're almost completely unproven, have had very little analysis done on them, and violating the first point of defence-in-depth - fuel sealed in a solid form. The MSR would spread fission product contamination throughout the primary system (and the secondary if you're unlucky), which sounds like a safety and maintenance nightmare. It's like a reprocessing plant only much worse, as at least the fuel's left to cool off for a while before reprocessing.
#2 - The US has no fuel recycling program. If we DID have a responsible fuel recycling program, we wouldn't have to worry about the whiny idiots going "but it produces nuclear waste", nor would we be having to dig up ore for fuel - reprocessed, recycled fuel can be extracted from "spent waste" over and over again, which would take care of 95% or greater of our current "nuclear waste" in storage.
Reprocessing does not eliminate the need for waste disposal, as it does nothing about the fission products. The multi-recycling you mention is also only feasible with fast breeders, not the current light water reactors.
I remember when scientists said fission would produce power to cheap to meter. The moral is that scientists are human and not infallible.
Do you have a source? That "too cheap to meter" quote keeps getting thrown around but is never directly attributed to anyone. It seems unlikely that anyone would have thought that, as even a coal plant's electricity has a huge cost contribution (very roughly 50% if I remember right but don't quote me) from the fixed cost of the plant, rather than the fuel. And nobody would have thought a nuclear plant would be cheaper to *build* than a coal one.
Here in the UK, BT are deploying VDSL2-based FTTC despite having to open their networks to others, with the prices they charge to others being regulated. So it can be done.
Some companies even offer flat rates to various international destinations, at least where I am. But some countries do indeed charge extortionate amounts for incoming calls, yes.
Most packet loss is due to congestion, which using FEC is only going to make worse. So you'll gain your phone call clarity at the expense of other traffic.
How did the government "throw away" manufacturing? Privatisation should not have done that if the manufacturing was profitable to start with - and if it wasn't, why do we want it? There's little sense in doing ourselves that which can be done more cheaply by others, unless you want to get paranoid about self-sufficiency.
You haven't addressed his point that the previous government spent recklessly on the assumption that the boom would continue forever.
Won't increasing the number of bits per symbol as you suggest require a higher SNR, thus meaning amplifiers have to be more closely spaced? Given the desire is to get more out of the existing infrastructure, that might be a problem.
It might make funny tv, but at the heart of it is the seperation between those who do the actual work, and those that form the ruling elite.
You can hardly call Ian Hislop the ruling elite, since his most notable activity is editing a magazine whose purpose is to investigate and criticise it. For that matter, you can hardly call Bob Crow an example of the average working man, given his income of over £100k per year.
That's just one study, not an established medical fact. It does give cause for concern and merits more research but doesn't prove
anything yet.
Assuming we're referring to Kirsch's paper, It wasn't even an original study but a meta-analysis. When I read it I noticed some clear flaws with the way the data sets were combined, too - he seemed to average the placebo result from all the trials, average the drug result and then look at the difference in the average.But that would only be valid if all the tests were conducted under identical conditions, which they weren't. For example, with some fictional data (but his paper showed the same behaviour):
Trial 1: drug: no. subjects 50, average improvement -12, placebo: N=100, improvement -9. Drug better than placebo by 3 points. Trial 2: drug: no. subjects 100, average improvement -8. placebo: N=50, improvement -5. Drug better than placebo by 3 points
combined: drug: N=150, avg improvement -9.3. placebo: N=150, avg improvment -7.6. Drug better than placebo by 1.7 points.
The shielding required for a tiny reactor isn't much less than for a huge one, thanks to the exponential attenuation of radiation. Say a big reactor produces 10^9 times too much radiation to be safe unshielded and one a thousandth of the size produces 10^6 times too much - the small one will still need 2/3 of the shielding thickness of the big one.
Also, depleted uranium isn't very hazardous but fission products certainly are.
We don't have every person or entity (or, worse yet, every item that person or entity owns) assigned it's own global unique "street" address or phone number.
I don't know about you, but mail addressed to me comes to (my name, house number and road, town, post code). NAT is more like only being able to send to the house - or, for ISP scale NAT, the town - and then having to play guessing games with who the post is actually for by inspecting the contents. Anyone who wanted to implement a postal system that way would be rightfully considered insane.
Read "un-patched" version of windows. Yes, MS Windows has a great firewall that lets just about everything out. There are a few applications that will allow stuff in, the firewall allows that, too (user nearly always clicks on "Allow"). One of those un-patched applications has a serious flaw, or a zero day, and the computer is hosed, firewall or not.
True, but the same will happen right now thanks to UPnP.
The unpatched XP won't be a problem with IPv6 as IPv6 is off by default in XP.
However, suppose a flaw is found in Microsoft's firewall software. The system in question is still unpatched.
That could happen theoretically, just like it's possible for your NAT box or separate firewall to be hacked and configured to forward traffic inwards. But is it a big enough risk to limit functionality by denying incoming connections? I like being able to do peer-to-peer, VoIP games etc.
Analog has noise, but digital has aliasing and rounding errors.
Only if you do it wrong. Aliasing can be avoided by low-pass filtering your signal prior to sampling, and the rounding error can be made equivalent in character to analogue noise by dithering. The SNR for a CD is something like 30 dB better than for vinyl.
DIgital radio is wireless data. Adding a return path just seems like needless complexity - how will this a) save bandwidth and b) be beneficial except to advertisers?
Whether criminalizing kiddie porn is a good idea or a bad one (I can understand the viewpoint of the porn enabling the crimes / creating the demand), when you have thoughtcrimes on the books,
Intentionally acquiring child porn isn't a thought, it's an action. It's not a crime to be a paedophile, the crime lies in the associated activities.
For a home user, I'd say that a central firewall on the router is a bad idea, except for basic sanity checking of packets. It reduces the flexibility of your internet connection unless you know how to configure it (the average home user doesn't), and frankly any product that needs an external firewall to stop it getting rooted is defective. If that means coming with a built in firewall with appropriate firewall settings by default then so be it. Even Windows is secure by default these days from network based attacks.
Business IT is a different matter entirely though, as you have a sysadmin to adminster the device.
Most people probably don't know that USB will auto-run on Windows. I didn't until fairly recently, and was pretty surprised when I find out. Surely even Microsoft wouldn't leave such an obvious attack vector open...
The BBC Micro was a wonderful machine for tinkering. Assuming they're using the BBC Basic (which was very good compared to the other micro Basics), you can easily do the "quick fix let's test it" thing. Actually, you probably can if they're using 6502 assembly language - the system came with an assembler integrated with the Basic interpreter.
A great computer to program, in its day. All this high level abstraction of modern languages may be very practical but it's so terribly dull.
The kinda irony with this nuclear power plant is that uranium fission for electricity was commercialised partly due to a side effect: the production of plutonium. I think I am right in saying that there is more thorium in the Earth's crust than uranium, thorium is more easily available, it is fissionable, and doesn't produce the same pollutants (or explosives, if you want to look at it like that) that the fission of uranium does.
Thorium isn't fissile, so you need either uranium or plutonium in there as well. During the reaction, thorium gets converted to enough fissile uranium to replace the fissile material consumed, so it's a breeder. But we've abandoned breeders in general, not just thorium based ones.
The real reason enriched uranium is used is that it's cheaper and easier than setting up a breeding cycle, and the reason the PWR became dominant was that the technology was already well developed for submarine propulsion. Nothing to do with Pu production.
Then please enlighten us with your expert knowledge. If there's no congestion, the buffers never fill up, and every packet gets through quickly. How can prioritising help?
This doesn't buy you much though. IIRC, one study suggested that doing all this fancy QoS may let you provide the same level of service (as perceieved by the users) with half the bandwidth. Given that bandwidth is cheap [1], is it really worth all the extra administrative issues as well as the potential for abuse?
Don't forget - if there's no congestion, QoS does nothing.
But if we prioritise certain traffic, heavy use of batch net use like filesharing no longer gets in the way of real-time net use like gaming and Skype and we no longer need to care about how much any one person downloads.
It would still get in the way of other people's bulk data transfers though, so you would still need to worry about how much.
[1] Unless you're buying from BT Wholesale, which I think is the issue here for Demon.
Is that an issue for the big pipes in the core though? I may be misremembering, but I thought the buffers stayed pretty much empty up to a very high % utilisation, as the aggregation of a huge number of independent flows smooths out the fluctuations and jitter associated with any one flow. As long as the capacity in the shared section is much larger than any individual's pipe, this should remain the case.
I guess in a LAN there's still the potential for an individual machine to use up all the capacity, so this could still be an issue.
Slashdot Mirror
They're almost completely unproven, have had very little analysis done on them, and violating the first point of defence-in-depth - fuel sealed in a solid form. The MSR would spread fission product contamination throughout the primary system (and the secondary if you're unlucky), which sounds like a safety and maintenance nightmare. It's like a reprocessing plant only much worse, as at least the fuel's left to cool off for a while before reprocessing.
#2 - The US has no fuel recycling program. If we DID have a responsible fuel recycling program, we wouldn't have to worry about the whiny idiots going "but it produces nuclear waste", nor would we be having to dig up ore for fuel - reprocessed, recycled fuel can be extracted from "spent waste" over and over again, which would take care of 95% or greater of our current "nuclear waste" in storage.
Reprocessing does not eliminate the need for waste disposal, as it does nothing about the fission products. The multi-recycling you mention is also only feasible with fast breeders, not the current light water reactors.
I remember when scientists said fission would produce power to cheap to meter. The moral is that scientists are human and not infallible.
Do you have a source? That "too cheap to meter" quote keeps getting thrown around but is never directly attributed to anyone. It seems unlikely that anyone would have thought that, as even a coal plant's electricity has a huge cost contribution (very roughly 50% if I remember right but don't quote me) from the fixed cost of the plant, rather than the fuel. And nobody would have thought a nuclear plant would be cheaper to *build* than a coal one.
Here in the UK, BT are deploying VDSL2-based FTTC despite having to open their networks to others, with the prices they charge to others being regulated. So it can be done.
Some companies even offer flat rates to various international destinations, at least where I am. But some countries do indeed charge extortionate amounts for incoming calls, yes.
Most packet loss is due to congestion, which using FEC is only going to make worse. So you'll gain your phone call clarity at the expense of other traffic.
How did the government "throw away" manufacturing? Privatisation should not have done that if the manufacturing was profitable to start with - and if it wasn't, why do we want it? There's little sense in doing ourselves that which can be done more cheaply by others, unless you want to get paranoid about self-sufficiency.
You haven't addressed his point that the previous government spent recklessly on the assumption that the boom would continue forever.
Won't increasing the number of bits per symbol as you suggest require a higher SNR, thus meaning amplifiers have to be more closely spaced? Given the desire is to get more out of the existing infrastructure, that might be a problem.
It might make funny tv, but at the heart of it is the seperation between those who do the actual work, and those that form the ruling elite.
You can hardly call Ian Hislop the ruling elite, since his most notable activity is editing a magazine whose purpose is to investigate and criticise it. For that matter, you can hardly call Bob Crow an example of the average working man, given his income of over £100k per year.
That's just one study, not an established medical fact. It does give cause for concern and merits more research but doesn't prove
anything yet.
Assuming we're referring to Kirsch's paper, It wasn't even an original study but a meta-analysis. When I read it I noticed some clear flaws with the way the data sets were combined, too - he seemed to average the placebo result from all the trials, average the drug result and then look at the difference in the average.But that would only be valid if all the tests were conducted under identical conditions, which they weren't. For example, with some fictional data (but his paper showed the same behaviour):
Trial 1: drug: no. subjects 50, average improvement -12, placebo: N=100, improvement -9. Drug better than placebo by 3 points.
Trial 2: drug: no. subjects 100, average improvement -8. placebo: N=50, improvement -5. Drug better than placebo by 3 points
combined: drug: N=150, avg improvement -9.3. placebo: N=150, avg improvment -7.6. Drug better than placebo by 1.7 points.
Hmm.
The shielding required for a tiny reactor isn't much less than for a huge one, thanks to the exponential attenuation of radiation. Say a big reactor produces 10^9 times too much radiation to be safe unshielded and one a thousandth of the size produces 10^6 times too much - the small one will still need 2/3 of the shielding thickness of the big one.
Also, depleted uranium isn't very hazardous but fission products certainly are.
We don't have every person or entity (or, worse yet, every item that person or entity owns) assigned it's own global unique "street" address or phone number.
I don't know about you, but mail addressed to me comes to (my name, house number and road, town, post code). NAT is more like only being able to send to the house - or, for ISP scale NAT, the town - and then having to play guessing games with who the post is actually for by inspecting the contents. Anyone who wanted to implement a postal system that way would be rightfully considered insane.
Read "un-patched" version of windows. Yes, MS Windows has a great firewall that lets just about everything out. There are a few applications that will allow stuff in, the firewall allows that, too (user nearly always clicks on "Allow"). One of those un-patched applications has a serious flaw, or a zero day, and the computer is hosed, firewall or not.
True, but the same will happen right now thanks to UPnP.
The unpatched XP won't be a problem with IPv6 as IPv6 is off by default in XP.
However, suppose a flaw is found in Microsoft's firewall software. The system in question is still unpatched.
That could happen theoretically, just like it's possible for your NAT box or separate firewall to be hacked and configured to forward traffic inwards. But is it a big enough risk to limit functionality by denying incoming connections? I like being able to do peer-to-peer, VoIP games etc.
Is that likely to be an issue any more? IPV6 by default only came in with Vista, and Vista also brought along an on-by-default firewall.
Analog has noise, but digital has aliasing and rounding errors.
Only if you do it wrong. Aliasing can be avoided by low-pass filtering your signal prior to sampling, and the rounding error can be made equivalent in character to analogue noise by dithering. The SNR for a CD is something like 30 dB better than for vinyl.
DIgital radio is wireless data. Adding a return path just seems like needless complexity - how will this a) save bandwidth and b) be beneficial except to advertisers?
Whether criminalizing kiddie porn is a good idea or a bad one (I can understand the viewpoint of the porn enabling the crimes / creating the demand), when you have thoughtcrimes on the books,
Intentionally acquiring child porn isn't a thought, it's an action. It's not a crime to be a paedophile, the crime lies in the associated activities.
For a home user, I'd say that a central firewall on the router is a bad idea, except for basic sanity checking of packets. It reduces the flexibility of your internet connection unless you know how to configure it (the average home user doesn't), and frankly any product that needs an external firewall to stop it getting rooted is defective. If that means coming with a built in firewall with appropriate firewall settings by default then so be it. Even Windows is secure by default these days from network based attacks.
Business IT is a different matter entirely though, as you have a sysadmin to adminster the device.
Most people probably don't know that USB will auto-run on Windows. I didn't until fairly recently, and was pretty surprised when I find out. Surely even Microsoft wouldn't leave such an obvious attack vector open...
The BBC Micro was a wonderful machine for tinkering. Assuming they're using the BBC Basic (which was very good compared to the other micro Basics), you can easily do the "quick fix let's test it" thing. Actually, you probably can if they're using 6502 assembly language - the system came with an assembler integrated with the Basic interpreter.
A great computer to program, in its day. All this high level abstraction of modern languages may be very practical but it's so terribly dull.
The kinda irony with this nuclear power plant is that uranium fission for electricity was commercialised partly due to a side effect: the production of plutonium. I think I am right in saying that there is more thorium in the Earth's crust than uranium, thorium is more easily available, it is fissionable, and doesn't produce the same pollutants (or explosives, if you want to look at it like that) that the fission of uranium does.
Thorium isn't fissile, so you need either uranium or plutonium in there as well. During the reaction, thorium gets converted to enough fissile uranium to replace the fissile material consumed, so it's a breeder. But we've abandoned breeders in general, not just thorium based ones.
The real reason enriched uranium is used is that it's cheaper and easier than setting up a breeding cycle, and the reason the PWR became dominant was that the technology was already well developed for submarine propulsion. Nothing to do with Pu production.
Then please enlighten us with your expert knowledge. If there's no congestion, the buffers never fill up, and every packet gets through quickly. How can prioritising help?
This doesn't buy you much though. IIRC, one study suggested that doing all this fancy QoS may let you provide the same level of service (as perceieved by the users) with half the bandwidth. Given that bandwidth is cheap [1], is it really worth all the extra administrative issues as well as the potential for abuse?
Don't forget - if there's no congestion, QoS does nothing.
But if we prioritise certain traffic, heavy use of batch net use like filesharing no longer gets in the way of real-time net use like gaming and Skype and we no longer need to care about how much any one person downloads.
It would still get in the way of other people's bulk data transfers though, so you would still need to worry about how much.
[1] Unless you're buying from BT Wholesale, which I think is the issue here for Demon.
Of course, weight loss is a symptom of depression, so is it a side effect or the drug working as intended?
Is that an issue for the big pipes in the core though? I may be misremembering, but I thought the buffers stayed pretty much empty up to a very high % utilisation, as the aggregation of a huge number of independent flows smooths out the fluctuations and jitter associated with any one flow. As long as the capacity in the shared section is much larger than any individual's pipe, this should remain the case.
I guess in a LAN there's still the potential for an individual machine to use up all the capacity, so this could still be an issue.