This earlier slashdot post still seems like the best theory, nothing seems to contradict it
http://slashdot.org/comments.p...

I always feel good when I hold my Yellow fluke. Somehow I associate that color with a symbol of quality. SO as stupid as copyrighting Yellow for multimeters sounds when you say it out loud, I can see that this is actually something of immense value to the brand in this case. If you are not a EE then you probably don't understand this sensation.

I suspect they honestly did transfer the keys to a new wallet at some point. Then some time later they "stole" them by deleting those keys from the new wallet (keeping a backup in their hideout) but forgot there might be other copies of the key around when they deleted the keys in the new wallet. The person that stole them might not even realized there was an old wallet. That is how would anyone know the past wallets a key might have resided in. there's no history that tells you how many copies of a key exist. So if you think you have the only one, you might try to steal it.

Consider, What does it mean for a bitcoin to be lost or to be found or to exist. To be "lost" it means no one no longer knows the bitcoin's key. Yet you can also have more than one copy of the key. For example MagicTux might have "stolen" the coin (that is, transfered a coin to a key, deleted that key from the mtGox data base ("oopsie") but secretly kept a copy of the key somewhere else. But then suppose that not all copies of the key were deleted. Both the "found" key and the one lurking in MagicTux's hideout are the same valid key. Either one can spend the coin.

Thus there is not one copy of the coin to be found. there could be many.

If you give someone control of your wallet, so that they know your keys, you can never get back that control. They can always keep copies of it and have the authrority to spend. THe only way to recover control is to make a new key and transfer the coins to that. Thus these exchanges that manage your coins are scary.

I wish there was just some voluntary static typing in python. By this I don't mean run time voluntary type checking. that makes it slower. No I mean a pre-run time filter that optimizes the .pyc to the extent it can.

Here's a link:

http://www.gizmodo.com.au/2013...

Does Google not recall the NSA post it note showing that they intercept the post-SSL server to server commuincations within the googleshpere? NSA doesn't care about HTTPS to google as long as that back channel is still there.

We begin with Goodfellows argument for a fire which, by the way, was also raised by another anaylyst. The we demolish Slates counter argument.

1) There's an electrical fire, all the breakers are tripped (removing the data transponders and maybe the communications). http://www.airtrafficmanagemen... http://www.wired.com/autopia/2...

The Malaysian primary radar inferred a flight path with the turns at VAMPI and GIVAL after the Lankawi International airport overflight:
supposed: flight path :http://skyvector.com/?ll=10.332212843477643,95.11743164439306&chart=304&zoom=8&plan=F.WM.IGARI:F.WM.VAMPI:F.WM.GIVAL:F.VO.IGREX

So Slate asks how do we account for the red herring turns at VAMPI and GIVAL?

3) Coincidentally, after the incapacitated MH370 overshoots the airport, at that very moment UTC March 7 18:00, another 777 flown by Singapore Airlines (Flight SIA 68) crosses MH370s flight path.

http://www.flightradar24.com/2...

4) MH370 is low since its trying to land and so the Malaysian Royal Airforce primary radar is having some trouble following it. The primary radar initially sees one 777 (MH370) then after losing it confuses this with SIA 68, which is the only 777 they can now see in the air at the same GPS coordinate.

5) SIA68 then executes two planned waypoint turns (GIVAL and IGREX), so we get the red herring that a skilled pilot was in control of the flight just before SIA 68, not MH370, goes off the end of the Malaysian radar

We add one more flourish to explain why the Indonesians also missed the (tiny) overflight of one of their archpeligo, a point Slate did not raise.

6) The pilots are incapacitated as MH370 continues on the same line, skims low over the tip of indonesia and flys out into open ocean. As it happens at 18:05 UTC Flight UAE343 (as well as one other flight before it) , also a 777, is also flying over the tip of Indonesia at that same moment so again a potential for misattributed distant radar returns.

http://www.flightradar24.com/2...

Finally tie it into a bow to answer slates last objection:

7) if you extend that line out it will eventually intersect the supposed last ping satellite transmission radius somewhere far off the west coast of Australia, perhaps vaguely near the Coco islands. I can't be too precise because the maps are not draw with correct spherical geometry.

8) since Goodfellow's claim a new set of facts has come out that aid it further. It has been now revealed that the Lankawi overflight path was entered into the computer prior to the "goodnight all is well" message from the co-pilot to the tower. Some people saw that premeditaion as suspicious. However It has also been revealed that extremely conscientious pilots do this routinely. they program the nearest escape path into their flight computers and keep it updated as they travel from way point to way point. they don't hit the execute button. It's just there already to go if things go south and no matter who is flying the plane at that moment. Goodfellow also said the first thing he saw was a pilot who already knew what he was going to do in an emergency and didn't have to think about it. So rather than being suspicious it explains a lot.

Goodfellow also noted that while there is some uncertainty about the strange climb and dives inferred from the (altitude-unreliable) radar data, that these are consistent with a huge smoky fire: climb to 40,000 feet in a desperate move to starve it of oxygen. Then dive at a ridiculous rate to try to blow it out or at least get close to ground for a ditch in the ocean.

the theory is that by the time they got close to Lankawi the pilots would have used up all of the chemical smoke filters they carry for smoke protection (~ 15 minutes maybe) and cockpit would have been overhwhelmed by smoke. If they were not dead (most likely) then about all they might have managed was restoring the autopilot at that point.

While you jest about Hot Wallets I think this really is the issue with Bit Coin. Specifically the incidents with bit coin lately have revealed first of all how robust the underlying system is, while also revealing how vulnerable the commerce mechanisms that wrap it are. It make you realize what you are paying Visa it's percentage for. Not the transaction mechanism, which bit coin shows is cheap, but the transactional security. If I send my bitcoins to someone either for payment or for "safekeeping" (cause I don't trust my own security from getting hacked) or for an "exchange", and I am cheated then the money is gone, flown across borders, and not even the laws are sure it has value let alone worth the cost of enforcement. With Visa I can be ripped off to, but Visa is pretty darn good about trying to keep the customer happy (better than paypal! ). That is bit coin is a great mechinaism but there's no defined trust wrapper for it. But there could be. If some place like Amazon or Alibaba were to escrow the bit coin transactions they could become the enforcer of trust. And of course they would get a fee.

So my point I guess is that while we tout bit coin as being a cheap transaction that doesn't need The Man, if it's going to become big then The Man will have to be involved. But the cool thing is that if that happens you can still do your back alley transaction without the Man's help-- your choice. Moreover, there can be lots of different trust mechanisms, not just one Visa alliance.

Not exactly, that's linear thinking again. Consider how a frequency like 23.123456Kz is represented when the sampling rate is not an integral multiple. The ear can hear that frequency as a pure tone. That is, in principle there is enough information there to infer that it is a pure tone. But the digitized version doesn't play that tone, it plays a mixture of tones with different phases and amplitudes (the convolution of Sin(x)/x with a pure frequency, and only frequencies that happen to be at integral multiples of the sampling rate. The ear can infer it's not a pure tone.

Beyond some sampling rate the ear won't be able to do that trick. But I don't know what that rate is. It might even be less than 20Khz or it might be more.

The nyquist theorem applies when the frequency basis set that generated the time series of equally spaced intervals, was drawn from equally spaced frequencies separated by 1/T. But it doesn't apply if the basis set includes off latice frequencies.

Now a person thinking linearly will say to resolve those off lattice frequencies is equivalent to hearing ultrasonics. It isn't. What it means is to reproduce those off lattice frequency generated time series with on-lattice points you would need to have more frequency components.

Y'know personally, I doubt there's anything I myself could hear at higher sampling rates. I'm just being a pest because people keep insisting that the nyquist theorem applies and therefore there's no possibility that there isn't something being lost by the sampling at constant intervals. That's not true but it also doesn't mean Mr. Young is right either.

Okay.

1) you have two frequencies at 23Khz and 25khz. Say they mix vith a quadratic non-linearity then you get a frequency at 2Khz

2) this is not simply aliasing

3) The case of 2 frequencies is degenerate but if I have multiple simultaeous frequencies mixing down below 20Khz then I can reconstruct the original spectrum under commonly true assumptions (such as sparsity).

There can also be non-linearities in time and phase as well. A trivial example is that a loud sound a moment before can block hearing a quiet sound a moment later.

Finally imagine how the ear actually works. It's not just a comb spectrum analyzer. It's actually sensitive at frequencies in between discrete values. Moreover in order for sound to reach the folicles the furthest in it has to pass by the folicals at the front, Therefore there is a great deal of mixing in time and frequency that can occur from this design.

I'm not saying this matters or that 192 K solves it if it does matter. I'm just saying the linear analyses assume their own conclusion and parade out Nyquists theorem as though it applied.

oops ill chosen values. use 16 samples per second for 2 seconds instead.

Add to the list of things you don't understand: Nyquists theorem.

Educational exercise 1: create 128 random Y values uniformly spaced along the X axis. Delete every other one and fourier transform it. Now pad this out to 2x the frequency and inverse fourier transform it. You get 128 Y values with the original spacing. But they are not the same points you started with.

Educational exercise 2: Create a sine wave at 3.14159 oscilations per second. Sample this at 16 samples per second for 1 second. throw away the second half so you have 8 samples. Fourier transform this. First observe there is no peak at 3.1459 hz because there's no discrete frequency there. Now inverse FT this. You dont get the original 16 points.

I don't think you understand non-linearity and how it can mix down a high frequency signal into an audio range. Another thing you dont understand is what happens when a frequencys period is not an integral multiple of the sample rate. In short are smokeless

The greatest time in that articel is spent claiming that 192Khz is overkill because everything above 20Khz is unhearable. He shows how a square looking waveform has all the right spectral components in the 20Khz range and so therefore it it is not missing anything. This is fourier and nyquist type argument that assumes linearity.

as you put it F( a+b) = F(a) + F(b). When this is true then it's as he said. But if F(a+b) != F(a) + F(b) then you need more than 20Khz to describe the spectrum.

I'm not saying 192Khz is the right thing. I'm just say the entire argument in the article is assuming linearity to draw the conclusions that the 0-20Khz spectrum contains all the information you can hear.

In fact we already know that ears are not linear. This is in fact how some compression algorithms function. They know that as it gets loud that you can't hear quieter frequencies as efficiently so they are removed. This is an example that actually works in the opposite direction-- that there's less information needed. However it supports the notion that describing everything by spectral analysis is wrong when things are linear.

You said, well it's just a change of basis. Sort of. How tightly you want to sample has th be determined first. This is what actually sets the bases that the analysis is going to be changing between. A given point spacing in time for a given lenght of time forces the interval over which the fourier transform exists. Conversely if you insist that the highes frequency is 20K (or 40K for nyquist) then you have fixed the time interval of the sampling. You are then blind to any point in the intervals between which is where the non-linear effects could, conceivebly, hide.