Its fairly simple to keep track when all you're looking at is cash in and cash out.
Now imagine you've got $10 million spread across maybe 1000 different investments. There will be a line for each dividend received, any tax already paid (don't know exactly how the US works but in the UK 10% tax is taken at source before you get the money).
Then you'll have to account for cheques written but not yet presented. It's no good saying "Ah yes, that looks about right" when there's about $10K of cash in your account if there's an unpresented cheque of 100K and fraud of 98K approximately compensating.
I have problems balancing my equity investments with what is reported in statements and I've not got much invested directly. For example, my statements include a "capital account" and an "income account". Money seems to get transferred randomly between them. So I might get a dividend of 100 pounds appearing in the income account, then there is a management fee deducted, and then money is transferred to the capital account. Effectively I have to work through two statements eliminating the common items. And that's just to balance the cash.
There are times, e.g. takeovers where you get cash and shares, when the only way I've been able to make the accounts balance is to look at the numbers on my statement, look at my numbers, and then adjust the commission or tax. For example, back in March Alfred McAlpine was taken over by Carrillion. I got some cash and some Carrillion shares. I know what I should have got for my shares. But the only way I can make the numbers balance is to put in a random number for commission charged by my broker.
I have NEVER managed to get my calculations to match my pension exactly after I've made a payment into it. Say I pay in 1000 pounds and I look up the price of the fund I'm investing in. I cannot make the numbers balance at all. The discrepancy is small but it's there. I've got one fund I don't pay into any more, but the number of units I hold changes monthly. It's only a few pounds per month but it makes checking the accounts impossible.
Which will return the size of a char * (pointer to character) on your system (typically 4 bytes), _not_ the length of the array. There is no way in C to get the length of an array after it's been allocated. Arrays are 'stupid' chunks of memory, not objects with properties.
Huh?
char x[9]; printf("%d\n", (int)sizeof x);
will print 9 exactly as required.
There are a handful of cases where arrays do not decay to pointers. This is one of them.
There is some detectable C-14 in some coal (and oil). AFAIAA nobody is absolutely sure where it comes from, there are several possibilities (or a combination of more than one).
But even for the coal with the highest levels of C-14, releasing the CO2 into the atmosphere will dilute the existing C-14 in the atmosphere, not increase its concentration.
Atmospheric testing of nuclear weapons caused a significant spike in C-14 concentrations in the 50s/60s.
To me, a website who has let their certificate expire or is too cheap to spend $10 a year to get a real certificate is not a website that I want to be doing business with in the first place.
Presumably you use unencrypted sites. So why would you refuse to use sites that have a self signed certificate? Does the fact that you have some additional protection against some attackers (e.g. someone with a packet sniffer at your isp) but not total protection immediately stop you using the site?
IMO there should be little difference between http:/// and https:/// when using self signed certs. If anything there should be a small warning for the http:/// that the connection is unencrypted.
There isn't a coal plant on the planet that could get an operating license as a nuclear plant, given the amount of radioactive carbon they dump into the air.
While coal plants release more radioactivity into the environment[1] than a nuclear plant is allowed to, radioactive carbon is one thing they do NOT release. In fact, one of the reasons we can be sure that the increase in CO2 in the atmosphere is fossil, rather than some natural effect of the biosphere is because it's _less_ radioactive than it would naturally be.
This has also caused problems for carbon dating - the relative reduction in C-14 has caused 20th Century objects to look older than we would expect due to having a lower C-14 activity which we would normally attribute to C-14 decay but is, in fact, due to the dilution from fossil CO2.
[1] Much of the radioactivity ends up in the ash heap but a significant amount goes up the smoke stack, particularly radioactive gases like radon.
But if you crashed and became brain damaged, you would be very expensive to look after, and I (and a whole bunch of other people) are going to pay for that with our insurance premiums and taxes.
Head injury rate is positively correlated with helmet wearing. Cycle usage is negatively correlated with helmet wearing.
Everywhere where helmets for cyclists have been made compulsory (and enforced), the decline in cycling has been larger than the decline in head injuries, i.e. head injury risk has increased.
and in BillG's case the money almost certainly won't disapper but will be paid to someone else. I doubt he's going to start turning $100 bills into paper aeroplanes and throwing them into the sea just to see how far he can throw them - but he might give his kids some quarters (I assume they're 1/4 dollar, not 1/4 cent) to try skipping.
Lets say BillG has $10b of disposable money (i.e. not stuff he actually uses e.g. home, car etc) Lets say you've got $1m in the same way (you've probably got a lot less).
BillG blowing $1000 is like you blowing 10 cents (is that a dime?)
One of your mates says "hey, anyone got a dime to try?" and you're the only one with one in their pocket. Are you seriously saying that you would never even consider throwing a dime into the sea just to see how well it went?
(And I suspect in practice $1000 to BillG is more like 1c or less to you - indeed, it won't be surprising if, not counting the value of your home, you're actually in debt due to a mortgage.)
Now lets say you give $1000 per year to a charity and BillG gives $10m.
Are you really going to forgo trying to skip the dime and up your donation to $1000.10? Do you think BillG is going to up his donation to $10,001,000 instead of, when he's on holiday with his kids, doing something "silly" that costs $1000 "because it's fun".
And that's also the strategy that top human chess players use to beat top computer chess players.
When Kasparov lost there was some surprise that he played an "active" game[1]. This plays into the computers strength because it can easily spot bad moves (by both players) and so it dramatically reduces the search space.
Tim.
[1] he made some bizarre decisions throughout the series of games - my own gut feeling is that he started the series assuming he would win easily and wasn't sufficiently prepared. In particular, his first move on the sixth game might have surprised a human player who would probably have spent more time researching other moves but this "shock" was completely ineffective against Deep Blue, and, indeed resulted in a "book" position for Deep Blue at move 8
Wouldn't feeding everything into a news server and then reading it with a newsreader that supports scoring do this?
I've used this in the past to read mailing lists, gateway the emails onto a local news server and then you (mostly) get proper threading.
Set up the local group as moderated and you can even use the news server to reply (but be careful if you do this as a small bug can result in you echoing every message you receive back to the mailing list which will not make you popular)
The data looking random or not has nothing to do with the information capacity of the channel. Shannons definition of information capacity is simply the maximum amount of information that can be recovered by the receiver on the channel. A channel where the receiver can't recover data under any circumstances is a 0 capacity channel, that reason could be interfering noise or the fact that the channel doesn't exist. Which poses a problem with this theory, it basically says 2 channels that don't exist can transmit information, which is intuitively incorrect.
I've only skimmed the paper but that's not what they are saying.
In classical information theory, if you have two channels, each with a capacity of 0 then the combined capacity of the channels is 0. It doesn't matter why each of the channels has a capacity of 0, whether that's noise, broken cable, etc. If you know the capacity of the channels you know the combined capacity.
Once we start transmitting quantum information things are harder. Just because each channel individually has a capacity of 0 doesn't *necessarily* mean that the capacity of the combined channels is also 0.
While this is only of theoretical interest at the moment its the sort of thing that could end up having practical consequences in the future. Someone has a quantum channel that's reaching capacity so they need to commission a new channel. By correctly specifying the new channel relative to the existing capacity they can actually add more new capacity than the new channel is capable of supporting.
This concept isn't completely unknown in the classical world. If you have a factory that needs electricity, you can supply the power on three phases using less conductor area than a single phase supply would need.
A "secure" and encrypted connection to a compromised or malicious server is worthless.
Exactly! My accountant needs some documents from me. Rather than email them I have them up on a secure site. If my accountant connects to the wrong site I really don't care, he's not going to find the documents he needs so he's going to give me a call and ask where they are.
Self signed certs are for when you want to do the encryption but you're doing the authentication via other means.
I've used this in the past (although not to my accountant).
At the very worst, a self signed certificate is no worse than a plain HTTP connection.
If we didn't have plain HTTP at all then we would consider sites using self signed certificates as secure (or insecure) as a plain HTTP connection.
Again as mentioned before LDPCs are not useful in these situations they are only useful in overcoming erasures within data communications.
Why aren't erasure codes any good here? What errors are we trying to recover from?
Don't we have known bad sectors from the disk?
I don't know if, when you get a bad sector the drive returns nothing or whether the drive can be told to return its best guess so I can see it might depend whether your code word is a sector or something smaller.
What happens when the pressure is relieved? Does your theoretical neutronium (or high-density Lithium!) go back to normal?
Define normal. If matter gets compressed down to neutronium it is still the same electrons, protons, neutrons, etc. that it originally was.
I wasn't sure from your original post but this is wrong.
If you compress the nuclei enough you get fusion. Fusion, as well as merging the two nuclei, usually also involves transforming some of the protons or neutrons into the other type. But the electrons are uninvolved (other than you might get electrons or positrons created in the fusion)
Neutronium is when you've not only merged the nuclei but you've squashed the electrons down into the nucleus as well. The electrons combine with the protons to give just a mass of neutrons.
Neutrons themselves are unstable with a half life of about 12 minutes. If you reduce the pressure then neutronium will start to decay. In a neutron star the electrons cannot escape the gravity well so the star is stable.
Very easily. The US and Europe are different markets. Analytics for pricing have shown time and again that Europeans and Britons are willing to pay more for consumer electronics and for software. Hence, suppliers charge more.
As time goes on and the "global" market homogenizes, this will change. But until then, pricing decisions based upon local markets will continue to create situations like those described in the summary.
There's another complication - if you import something into the UK (as a private citizen), it's IMPOSSIBLE to find out how much tax will be charged at import until after the item has cleared UK customs.
So the first you know is when you've ordered something from the states for $1000 that costs GBP1000 and when the courier arrives, theres a 300GBP tax bill to pay before the courier will give you the goods.
And this is on something that you also ordered two weeks ago, exactly the same paperwork, that got charged 15GBP.
Additionally, the courier will then probably charge you another 10% of the tax to collect the tax on your behalf (minimum 10GBP) non negotiable - you can't pay the taxman directly.
The rules are impenetrable. There's VAT, import duty, VAT on import duty, import duty on VAT, particular taxes on particular goods. The rates depend on what order the taxes are applied. If your bicycle arrives with the pedals separate then it's 1% import duty. If the pedals are attached then it's 20% (I made that bit up but there's definitely something funny with importing bicycles). Whether the import duty is charged on the "value" of the goods (i.e. the price you would have to pay in the UK) or the price you paid in the US.
Of course, the companies that sell in the UK (I assume it's similar in Europe) are perfectly happy with this state of affairs because it discourages parallel imports and allows them to maintain their margins.
And then, of course, we have crazy laws - a shop cannot buy Levi's jeans from wholesalers in Europe, import them into the UK (paying whatever taxes are required) and then resell them without the permission of Levi (US).
Is it any wonder that US companies screw over European customers when the European courts enforce the price discrepancy even when the ONLY difference to getting the same goods cheaper is using the supply chain in a different manner.
It would be like being charged to use a non toll road because your alternative route was the toll road.
In an infinitely long thread, you are absolutely certain to have at least one mention of every single concept, object, philosophy and idea ever known to humanity, because of the way probability works.
Only if the thread is irrational (just like you can find any combination of numbers in pi or e). In a nice, rational thread, you'll eventually get repetitions and the thread will loop back to itself.
This doesn't follow at all.
Liouville's constant is not only irrational, it's transcendental. But it only contains the digits 0 and 1.
IIRC, using oyster it's 90p for a single bus journey. But your bus travel is capped at 3GBP (about $6)
Likewise, tube travel. In central London it's 1.50GBP per journey but the cap (for zones 1 and 2) is 4.80GBP. Also the bus travel counts towards this. So, if you're staying outside central London as a tourist then get a bus into the centre, travel about the centre by bus and tube and then just get a bus back to your hotel and your travel will be capped at a maximum of 4.80GBP (so long as you always touch in and out on the tube)
You can also buy a travelcard. This will be 50p more than the capped oyster fare but there's no risk of forgetting to touch in or out.
Even from the mid 80's to the early 21st century it dumbed down. I even still have a few magazines from about 2003 that I haven't bothered to take out of their plastic mailing wrapping.
I recently (few years ago now) let my subscription lapse.
But it's not just SciAm, or American publications. I used to subscribe to Maplin Electronics. It used to have interesting projects and articles that you could learn from. But the projects degenerated into building kits, no information on how they worked, what you might like to try modifying etc, and the articles just became reviews of high street consumer gear.
Presumably, however, this move was more profitable for Maplin, and it's more profitable for SciAm.
Solving for t involves solving for the doppler shift even if you don't explicitly export that information.
The receiver generates the C/A chip code (I don't know why GPS calls it chip and not bit) and then finds the best correlation to the received signal in order to calculate the time. I think it's finding the time of the start of the (1023 bit) code down to an accuracy of about 1% of one bit.
I've noticed slow lock even on a bicycle. The worst case is when you start moving before it has spotted any satellites at all. It's not so bad once it's got one. (IIUC, once it's got one satellite, it will have the almanac for the rest within about 30 seconds which makes it easier to find the other satellites). Obviously, if the GPS was last used within about six hours then the previous almanac might be good enough. Modern receivers get a lock much faster than old ones, my Etrex Legend HCx will get a lock when moving as quickly as my Summit will when standing still, but gets a lock when standing still even more quickly.
The GPS has to factor out both the effect from the satellite and the effect from your moving. The satellite effect is easy to factor out because the satellite broadcasts its own ephemeris. (It also broadcasts the almanac for the other satellites so the GPS receiver knows where to start looking for the signals from the other satellites) But the GPS has to simultaneously solve for both your position and speed. When your speed is zero there is no "unexpected" factor to be solved for. I expect that probably, if you were driving in a straight line then it wouldn't matter too much, but typically, if you turn the GPS on in a car as you pull out of your drive you're going to be making lots of speed and direction changes while the GPS receiver is trying to get a lock.
I'm actually amazed that the separate satellite signals can be separated out at all. It always strikes me as incredible that you can simultaneously shout up to 32[1] different things all at the same time on the same frequency and the receiver can can detect which shouts are present and which aren't. If I had never seen how it was done I'd have said it was impossible to do.
Tim.
[1] I think this is the upper limit for the GPS C/A PRNG coding but I'm not certain
so you have to do differentials on position for speed, and it gets ugly.
No you don't. It's doppler, exactly the same as the radar gun.
Infact, it's that doppler effect that can make handheld GPS slow to lock on if you're moving. If you're planning to take one in a car (or on a bicycle), let it get a lock before you start moving.
If the gun was a laser type - then possibly the laser tracked across the car. Lets say the car was 12 feet long and initially the reflection was from the rear of the car and then the laser tracked to the front.
45mph is 66 feet/sec. 62mph is 91 feet/sec (at least that's for British miles - I think American miles are the same)
So if the car is moving forwards at 66 feet/sec while the gun is tracking forwards on the car at 25 feet per second then you could easily get a reading of 62mph while tracking the car for half a second. It depends on how long the laser has to track the car to get a reading. Just moving from windscreen to front grill over 1/4 second would give the same error.
If the gun is a doppler type radar gun then anything else moving nearby will cause funny echoes to be returned. You will get echoes from each moving object, together with echoes equal to the sum and differences of the speeds. Usually the more complicated echoes will be at such a low level that the gun doesn't detect them. And 17mph difference would imply either overtaking a cyclist or a cyclist going the other way and I wouldn't expect there to be enough reflection from the cyclist to be able to confuse the radar gun like that.
I also don't know where the "expert" got the idea that GPS units calculate speed from position and time. The speed measurement on a handheld GPS uses doppler and is extremely accurate, even when the position lock is poor. At least up to about 45mph, a GPS on the handlebars of a bicycle and a bicycle computer that is counting wheel revolutions and knows the circumference of the wheel agree to within 0.2mph provided to take the time to measure the front wheel circumference carefully. You don't even need to be watching the two, just find a hill, reset the maximum speed on both, cycle down it. Check the maximum speed measured on each device. The GPS is so accurate that now I actually use the GPS to calibrate my front wheel circumference - because it's hard to measure really accurately on your own because you can't easily keep a proper riding weight on the wheel while trying to measure and count wheel revolutions. Now I just do a quick roll the bike along a straight line - which will get you 1% accuracy or better and then, if there's an obvious bias in the readings between gps and bike computer, just adjust the circumference by a few millimetres to compensate.
Oh and contrary to them, I didn't have to hide my smoking habit. Because I didn't smoke. Even though my parents made it perfectly clear that they would not forbid it since they were such bad examples themselves.
It's not the prohibition, it's the disapproval. When you respect your parents, and they respect you, then disapproving of something is far more powerful than trying to prohibit it.
I'll bet that your parents disapproved of themselves smoking, and you knew they would disapprove of you smoking. But they also knew that they couldn't stop you if you were determined to - and you knew they couldn't have stopped you. Respect all around.
Slashdot Mirror
Its fairly simple to keep track when all you're looking at is cash in and cash out.
Now imagine you've got $10 million spread across maybe 1000 different investments. There will be a line for each dividend received, any tax already paid (don't know exactly how the US works but in the UK 10% tax is taken at source before you get the money).
Then you'll have to account for cheques written but not yet presented. It's no good saying "Ah yes, that looks about right" when there's about $10K of cash in your account if there's an unpresented cheque of 100K and fraud of 98K approximately compensating.
I have problems balancing my equity investments with what is reported in statements and I've not got much invested directly. For example, my statements include a "capital account" and an "income account". Money seems to get transferred randomly between them. So I might get a dividend of 100 pounds appearing in the income account, then there is a management fee deducted, and then money is transferred to the capital account. Effectively I have to work through two statements eliminating the common items. And that's just to balance the cash.
There are times, e.g. takeovers where you get cash and shares, when the only way I've been able to make the accounts balance is to look at the numbers on my statement, look at my numbers, and then adjust the commission or tax. For example, back in March Alfred McAlpine was taken over by Carrillion. I got some cash and some Carrillion shares. I know what I should have got for my shares. But the only way I can make the numbers balance is to put in a random number for commission charged by my broker.
I have NEVER managed to get my calculations to match my pension exactly after I've made a payment into it. Say I pay in 1000 pounds and I look up the price of the fund I'm investing in. I cannot make the numbers balance at all. The discrepancy is small but it's there. I've got one fund I don't pay into any more, but the number of units I hold changes monthly. It's only a few pounds per month but it makes checking the accounts impossible.
Tim.
Huh?
char x[9];
printf("%d\n", (int)sizeof x);
will print 9 exactly as required.
There are a handful of cases where arrays do not decay to pointers. This is one of them.
Tim.
Trivially.
N in base N has the representation 10 which has two digits. 2 is prime. QED.
Any base greater than the square root of the number and smaller than or equal to the number will have two digits in its representation.
Tim.
There is some detectable C-14 in some coal (and oil). AFAIAA nobody is absolutely sure where it comes from, there are several possibilities (or a combination of more than one).
http://www.talkorigins.org/faqs/c14.html
But even for the coal with the highest levels of C-14, releasing the CO2 into the atmosphere will dilute the existing C-14 in the atmosphere, not increase its concentration.
Atmospheric testing of nuclear weapons caused a significant spike in C-14 concentrations in the 50s/60s.
Tim.
To me, a website who has let their certificate expire or is too cheap to spend $10 a year to get a real certificate is not a website that I want to be doing business with in the first place.
Presumably you use unencrypted sites. So why would you refuse to use sites that have a self signed certificate? Does the fact that you have some additional protection against some attackers (e.g. someone with a packet sniffer at your isp) but not total protection immediately stop you using the site?
IMO there should be little difference between http:/// and https:/// when using self signed certs. If anything there should be a small warning for the http:/// that the connection is unencrypted.
Tim.
There isn't a coal plant on the planet that could get an operating license as a nuclear plant, given the amount of radioactive carbon they dump into the air.
While coal plants release more radioactivity into the environment[1] than a nuclear plant is allowed to, radioactive carbon is one thing they do NOT release. In fact, one of the reasons we can be sure that the increase in CO2 in the atmosphere is fossil, rather than some natural effect of the biosphere is because it's _less_ radioactive than it would naturally be.
This has also caused problems for carbon dating - the relative reduction in C-14 has caused 20th Century objects to look older than we would expect due to having a lower C-14 activity which we would normally attribute to C-14 decay but is, in fact, due to the dilution from fossil CO2.
[1] Much of the radioactivity ends up in the ash heap but a significant amount goes up the smoke stack, particularly radioactive gases like radon.
Tim.
Put it this way -- would chess be more fun if people tried to knock each other's pieces off the board while their back was turned?
Knock them off? I'm using a thermonuclear warhead and I'm going to ANNIHILATE them.
Tim.
But if you crashed and became brain damaged, you would be very expensive to look after, and I (and a whole bunch of other people) are going to pay for that with our insurance premiums and taxes.
Head injury rate is positively correlated with helmet wearing. Cycle usage is negatively correlated with helmet wearing.
Everywhere where helmets for cyclists have been made compulsory (and enforced), the decline in cycling has been larger than the decline in head injuries, i.e. head injury risk has increased.
http://www.cyclehelmets.org/
Tim.
Bother. Meant to add,
and in BillG's case the money almost certainly won't disapper but will be paid to someone else. I doubt he's going to start turning $100 bills into paper aeroplanes and throwing them into the sea just to see how far he can throw them - but he might give his kids some quarters (I assume they're 1/4 dollar, not 1/4 cent) to try skipping.
Tim.
I'd bet you would.
Lets say BillG has $10b of disposable money (i.e. not stuff he actually uses e.g. home, car etc)
Lets say you've got $1m in the same way (you've probably got a lot less).
BillG blowing $1000 is like you blowing 10 cents (is that a dime?)
Now imagine you're down at the beach skipping stones http://en.wikipedia.org/wiki/Stone_skipping
One of your mates says "hey, anyone got a dime to try?" and you're the only one with one in their pocket. Are you seriously saying that you would never even consider throwing a dime into the sea just to see how well it went?
(And I suspect in practice $1000 to BillG is more like 1c or less to you - indeed, it won't be surprising if, not counting the value of your home, you're actually in debt due to a mortgage.)
Now lets say you give $1000 per year to a charity and BillG gives $10m.
Are you really going to forgo trying to skip the dime and up your donation to $1000.10? Do you think BillG is going to up his donation to $10,001,000 instead of, when he's on holiday with his kids, doing something "silly" that costs $1000 "because it's fun".
Tim.
And that's also the strategy that top human chess players use to beat top computer chess players.
When Kasparov lost there was some surprise that he played an "active" game[1]. This plays into the computers strength because it can easily spot bad moves (by both players) and so it dramatically reduces the search space.
Tim.
[1] he made some bizarre decisions throughout the series of games - my own gut feeling is that he started the series assuming he would win easily and wasn't sufficiently prepared. In particular, his first move on the sixth game might have surprised a human player who would probably have spent more time researching other moves but this "shock" was completely ineffective against Deep Blue, and, indeed resulted in a "book" position for Deep Blue at move 8
http://en.wikipedia.org/wiki/Deep_Blue_-_Kasparov,_1997,_Game_6
Wouldn't feeding everything into a news server and then reading it with a newsreader that supports scoring do this?
I've used this in the past to read mailing lists, gateway the emails onto a local news server and then you (mostly) get proper threading.
Set up the local group as moderated and you can even use the news server to reply (but be careful if you do this as a small bug can result in you echoing every message you receive back to the mailing list which will not make you popular)
Tim.
The data looking random or not has nothing to do with the information capacity of the channel.
Shannons definition of information capacity is simply the maximum amount of information that can be recovered by the receiver on the channel.
A channel where the receiver can't recover data under any circumstances is a 0 capacity channel, that reason could be interfering noise or the fact that the channel doesn't exist.
Which poses a problem with this theory, it basically says 2 channels that don't exist can transmit information, which is intuitively incorrect.
I've only skimmed the paper but that's not what they are saying.
In classical information theory, if you have two channels, each with a capacity of 0 then the combined capacity of the channels is 0. It doesn't matter why each of the channels has a capacity of 0, whether that's noise, broken cable, etc. If you know the capacity of the channels you know the combined capacity.
Once we start transmitting quantum information things are harder. Just because each channel individually has a capacity of 0 doesn't *necessarily* mean that the capacity of the combined channels is also 0.
While this is only of theoretical interest at the moment its the sort of thing that could end up having practical consequences in the future. Someone has a quantum channel that's reaching capacity so they need to commission a new channel. By correctly specifying the new channel relative to the existing capacity they can actually add more new capacity than the new channel is capable of supporting.
This concept isn't completely unknown in the classical world. If you have a factory that needs electricity, you can supply the power on three phases using less conductor area than a single phase supply would need.
Tim.
Exactly.
A "secure" and encrypted connection to a compromised or malicious server is worthless.
Exactly! My accountant needs some documents from me. Rather than email them I have them up on a secure site. If my accountant connects to the wrong site I really don't care, he's not going to find the documents he needs so he's going to give me a call and ask where they are.
Self signed certs are for when you want to do the encryption but you're doing the authentication via other means.
I've used this in the past (although not to my accountant).
At the very worst, a self signed certificate is no worse than a plain HTTP connection.
If we didn't have plain HTTP at all then we would consider sites using self signed certificates as secure (or insecure) as a plain HTTP connection.
Tim.
Again as mentioned before LDPCs are not useful in these situations they are only useful in overcoming erasures within data communications.
Why aren't erasure codes any good here? What errors are we trying to recover from?
Don't we have known bad sectors from the disk?
I don't know if, when you get a bad sector the drive returns nothing or whether the drive can be told to return its best guess so I can see it might depend whether your code word is a sector or something smaller.
Tim.
I wasn't sure from your original post but this is wrong.
If you compress the nuclei enough you get fusion. Fusion, as well as merging the two nuclei, usually also involves transforming some of the protons or neutrons into the other type. But the electrons are uninvolved (other than you might get electrons or positrons created in the fusion)
Neutronium is when you've not only merged the nuclei but you've squashed the electrons down into the nucleus as well. The electrons combine with the protons to give just a mass of neutrons.
Neutrons themselves are unstable with a half life of about 12 minutes. If you reduce the pressure then neutronium will start to decay. In a neutron star the electrons cannot escape the gravity well so the star is stable.
Tim.
Very easily. The US and Europe are different markets. Analytics for pricing have shown time and again that Europeans and Britons are willing to pay more for consumer electronics and for software. Hence, suppliers charge more.
As time goes on and the "global" market homogenizes, this will change. But until then, pricing decisions based upon local markets will continue to create situations like those described in the summary.
There's another complication - if you import something into the UK (as a private citizen), it's IMPOSSIBLE to find out how much tax will be charged at import until after the item has cleared UK customs.
So the first you know is when you've ordered something from the states for $1000 that costs GBP1000 and when the courier arrives, theres a 300GBP tax bill to pay before the courier will give you the goods.
And this is on something that you also ordered two weeks ago, exactly the same paperwork, that got charged 15GBP.
Additionally, the courier will then probably charge you another 10% of the tax to collect the tax on your behalf (minimum 10GBP) non negotiable - you can't pay the taxman directly.
The rules are impenetrable. There's VAT, import duty, VAT on import duty, import duty on VAT, particular taxes on particular goods. The rates depend on what order the taxes are applied. If your bicycle arrives with the pedals separate then it's 1% import duty. If the pedals are attached then it's 20% (I made that bit up but there's definitely something funny with importing bicycles). Whether the import duty is charged on the "value" of the goods (i.e. the price you would have to pay in the UK) or the price you paid in the US.
Of course, the companies that sell in the UK (I assume it's similar in Europe) are perfectly happy with this state of affairs because it discourages parallel imports and allows them to maintain their margins.
And then, of course, we have crazy laws - a shop cannot buy Levi's jeans from wholesalers in Europe, import them into the UK (paying whatever taxes are required) and then resell them without the permission of Levi (US).
http://news.bbc.co.uk/1/hi/business/2163561.stm
Is it any wonder that US companies screw over European customers when the European courts enforce the price discrepancy even when the ONLY difference to getting the same goods cheaper is using the supply chain in a different manner.
It would be like being charged to use a non toll road because your alternative route was the toll road.
Tim.
This doesn't follow at all.
Liouville's constant is not only irrational, it's transcendental. But it only contains the digits 0 and 1.
Tim.
IIRC, using oyster it's 90p for a single bus journey. But your bus travel is capped at 3GBP (about $6)
Likewise, tube travel. In central London it's 1.50GBP per journey but the cap (for zones 1 and 2) is 4.80GBP. Also the bus travel counts towards this. So, if you're staying outside central London as a tourist then get a bus into the centre, travel about the centre by bus and tube and then just get a bus back to your hotel and your travel will be capped at a maximum of 4.80GBP (so long as you always touch in and out on the tube)
You can also buy a travelcard. This will be 50p more than the capped oyster fare but there's no risk of forgetting to touch in or out.
I typically use foot and bicycle in central London (because it's quicker) but provided you don't pay cash fares and use oyster or a travel card, I think the pricing for public transport is actually pretty good. Due to the fixed per journey price there are some extortionate per mile charges:
http://maps.google.co.uk/maps?f=d&hl=en&geocode=11804929658401494669,51.511495,-0.128425&saddr=A400%2FCharing+Cross+Rd+%4051.511495,+-0.128425&daddr=51.512776,-0.124133&mra=dme&mrcr=0&mrsp=1&sz=16&doflg=ptm&sll=51.51065,-0.127115&sspn=0.005863,0.012724&ie=UTF8&z=16
would be about 12GBP/mile cash fare if you took the tube rather than walked it and about 4.50GBP/mile using Oyster but you quickly reach the cap.
Tim.
Even from the mid 80's to the early 21st century it dumbed down. I even still have a few magazines from about 2003 that I haven't bothered to take out of their plastic mailing wrapping.
I recently (few years ago now) let my subscription lapse.
But it's not just SciAm, or American publications. I used to subscribe to Maplin Electronics. It used to have interesting projects and articles that you could learn from. But the projects degenerated into building kits, no information on how they worked, what you might like to try modifying etc, and the articles just became reviews of high street consumer gear.
Presumably, however, this move was more profitable for Maplin, and it's more profitable for SciAm.
Tim.
Solving for t involves solving for the doppler shift even if you don't explicitly export that information.
The receiver generates the C/A chip code (I don't know why GPS calls it chip and not bit) and then finds the best correlation to the received signal in order to calculate the time. I think it's finding the time of the start of the (1023 bit) code down to an accuracy of about 1% of one bit.
I've noticed slow lock even on a bicycle. The worst case is when you start moving before it has spotted any satellites at all. It's not so bad once it's got one. (IIUC, once it's got one satellite, it will have the almanac for the rest within about 30 seconds which makes it easier to find the other satellites). Obviously, if the GPS was last used within about six hours then the previous almanac might be good enough. Modern receivers get a lock much faster than old ones, my Etrex Legend HCx will get a lock when moving as quickly as my Summit will when standing still, but gets a lock when standing still even more quickly.
Tim.
The GPS has to factor out both the effect from the satellite and the effect from your moving. The satellite effect is easy to factor out because the satellite broadcasts its own ephemeris. (It also broadcasts the almanac for the other satellites so the GPS receiver knows where to start looking for the signals from the other satellites) But the GPS has to simultaneously solve for both your position and speed. When your speed is zero there is no "unexpected" factor to be solved for. I expect that probably, if you were driving in a straight line then it wouldn't matter too much, but typically, if you turn the GPS on in a car as you pull out of your drive you're going to be making lots of speed and direction changes while the GPS receiver is trying to get a lock.
I'm actually amazed that the separate satellite signals can be separated out at all. It always strikes me as incredible that you can simultaneously shout up to 32[1] different things all at the same time on the same frequency and the receiver can can detect which shouts are present and which aren't. If I had never seen how it was done I'd have said it was impossible to do.
Tim.
[1] I think this is the upper limit for the GPS C/A PRNG coding but I'm not certain
so you have to do differentials on position for speed, and it gets ugly.
No you don't. It's doppler, exactly the same as the radar gun.
Infact, it's that doppler effect that can make handheld GPS slow to lock on if you're moving. If you're planning to take one in a car (or on a bicycle), let it get a lock before you start moving.
Tim.
If the gun was a laser type - then possibly the laser tracked across the car. Lets say the car was 12 feet long and initially the reflection was from the rear of the car and then the laser tracked to the front.
45mph is 66 feet/sec. 62mph is 91 feet/sec (at least that's for British miles - I think American miles are the same)
So if the car is moving forwards at 66 feet/sec while the gun is tracking forwards on the car at 25 feet per second then you could easily get a reading of 62mph while tracking the car for half a second. It depends on how long the laser has to track the car to get a reading. Just moving from windscreen to front grill over 1/4 second would give the same error.
If the gun is a doppler type radar gun then anything else moving nearby will cause funny echoes to be returned. You will get echoes from each moving object, together with echoes equal to the sum and differences of the speeds. Usually the more complicated echoes will be at such a low level that the gun doesn't detect them. And 17mph difference would imply either overtaking a cyclist or a cyclist going the other way and I wouldn't expect there to be enough reflection from the cyclist to be able to confuse the radar gun like that.
I also don't know where the "expert" got the idea that GPS units calculate speed from position and time. The speed measurement on a handheld GPS uses doppler and is extremely accurate, even when the position lock is poor. At least up to about 45mph, a GPS on the handlebars of a bicycle and a bicycle computer that is counting wheel revolutions and knows the circumference of the wheel agree to within 0.2mph provided to take the time to measure the front wheel circumference carefully. You don't even need to be watching the two, just find a hill, reset the maximum speed on both, cycle down it. Check the maximum speed measured on each device. The GPS is so accurate that now I actually use the GPS to calibrate my front wheel circumference - because it's hard to measure really accurately on your own because you can't easily keep a proper riding weight on the wheel while trying to measure and count wheel revolutions. Now I just do a quick roll the bike along a straight line - which will get you 1% accuracy or better and then, if there's an obvious bias in the readings between gps and bike computer, just adjust the circumference by a few millimetres to compensate.
Tim.
Oh and contrary to them, I didn't have to hide my smoking habit. Because I didn't smoke. Even though my parents made it perfectly clear that they would not forbid it since they were such bad examples themselves.
It's not the prohibition, it's the disapproval. When you respect your parents, and they respect you, then disapproving of something is far more powerful than trying to prohibit it.
I'll bet that your parents disapproved of themselves smoking, and you knew they would disapprove of you smoking. But they also knew that they couldn't stop you if you were determined to - and you knew they couldn't have stopped you. Respect all around.
Tim.