The traditional key in ignition device is mixing two distinct functions: authentication and mode selection.
The link between the two and the mechanics of the failed device contributed to the problem. Wireless electronic proximity isn't necessary.
What is necessary is a simple reliable device for authentication, a key which is turned to allow other functions to operate. For instance, keys used for industrial or military controls for this purpose. They look like ordinary keys, not the large vehicle keys now common.
Then the selection of ignition mode is entirely separate, and turning to "off" or "accessory" which disengages hydraulics and safety systems while the vehicle is in motion should be disallowed.
| The only reason they're saying something now is because the well publicized Facebook acquisition has made enough noise to catch the attention of the C-level goons that run the place.
who think that because moneybags Facebook is writing large cheques to expensive M&A lawyers and bankers to get the deal to clear quickly that they can be paid off to go away.
It's a win-big/lose-little situation for Zenimax if they already have some lawyers on payroll.
In truth: Mann's and other's reconstructions have been done again and again with different techniques and the answer comes out almost the same. There was no intentional deception or major technical fault in Mann's original work or that of similar predecessors and successors.
Remember that outgoing heat flux is in fourth power of absolute temperature, so ignoring atmospheres (black body) Venus would be emitting 43 times as much heat and so would have to be that much closer.
You can't separate the pressure from the temperature and the actual heat flux and hypothetically imagine a '1 atm' pressure on Venus. With a similar atmosphere as Earth you'd have roughly a surface temperature T so that (T/287)^4 = 1.911, difference in solar insolation.
The message from Germany is that if you replace nuclear power with coal then more CO2 will be emitted. Well, of course. What climate action advocates favor using more coal? None.
If greenhouse gases emissions were actually taxed, then they wouldn't do that.
Of course there are unscientific 'environmentalists' whose emotional reactions to nuclear power (less safe and clean than solar, more safe and clean than coal) and unwillingness to look at quantitative facts lead them to bad outcomes. Just as climate deniers do.
| well, when deniers talk the science, the believers go into "burn the heretics at the stake" mode.
When deniers lie about the science, the believers in science go into "you are an ignorant idiot and ought to STFU and everybody should ignore your drivel" mode, as they ought to, because the future of technological civilization may be at risk from aggravated ignorance.
The actual physics of Anthropogenic Global Warming (of which anthropogenic CO2 is one but not an exclusive component, and no scare quotes needed as it is fact) is based upon the infrared emissivity of gases and their actual dynamics and concentration in the atmosphere.
This physics is lab validated and confirmed by in-situ objective measurements.
Analogies made to the lay public are imprecise, but the underlying science never was.
| So, what did you expect the SEC Chairperson to say.
"We are disturbed by the allegations, but do not yet have sufficient information to make a firm determiniation. We will use all of our investigatory powers, including subpoena, civil and criminal penalties, with cooperation from all other regulatory agencies and the FBI, to get to the heart of this matter and ensure that our markets are safe and honest."
Given that taxes already depend on holding period, it's not a nightmare.
Firms sure as heck do keep super accurate records of the entire transaction stream because it's their job to do so to pay and get paid! Financial market accounts systems are highly accurate and track every transaction!
A large firm doesn't have unlimited ability to dodge taxes, especially taxes that can be checked against the accounting systems.
It doesn't eliminate the information asymmetry, it makes that asymmetry unprofitable to exploit.
The problem is even easier to solve with even less impact---tax orders canceled within a second.
| The scenarios described in Flash Boys book are not "frontrunning", they're more accurately termed speed arbitrage---they know there is a price discrepancy between different venues and take advantage of it before anyone even notices the price discrepancy.
That's called front running.
Imagine:
In the old days, a broker gets an order from a client. He says loudly to the room, Texas Employees wants to buy IBM in large amounts. He chortles and being rather portly slowly ambles over to the phone to talk to his broker. It so happens that is brokerage has an arrangement so that some young sprightly man is invited to listen in and has living quarters in the office. His job is to listen to the fat man and run over and call up his own trader before the fat guy gets to his. His bosses pay the fat guy's boss.
Is that front-running? Youbetcha!
That's what's happening now with the minor exception that the fat guy goes to 20 different phones connecting to different traders slowly and the kid is stationed at phone #1.
| the amount of money pulled out of the market is less than was pulled out by the market makers under the original manual system when there was trading taking place on an exchange floor.
And how much would be pulled out by modern market makers with computers who don't front run? Automated market making without cheating is a good idea.
| You are crazy. Being charged a fraction of a penny more per share is the most you could claim the effect is. The only problem with that argument is it ignores the liquidity and spread reduction produced just by having HFT in the market.
The HFT's job is to avoid providing liquidity and reducing the spread---that's how they make money. Providing liquidity means trading on the opposite direction of a large investor's trade. The firms which mint money from front-running are trading in the same direction as the investor's trade. They sacrifice a tiny amount in the opposite to trigger their trading, and their apologists call this triviality as 'providing liquidity'.
That can't possibly be 'making a market'---and making a market doesn't require super low-latency connections faster than the exchanges own connections, but front-running does.
Look at the evidence of what the HFT's invest their money in---is there any legitimate need for them for market making?
Real market making involves risk by taking on inventory when other people don't want it and then slowly distributing it back out---it is statistical arbitrage of volume vs time.
| If the government controls how you can make money, those with money will seek to control the government. I'm not suggesting a solution, I don't have one.
A democratic republic with an independent, free press and a fearless judiciary and investigative service with subpoena and prosecution power is a good start.
| But understanding that "more regulation" will likely do just as much harm as it does good is important to this debate.
No, that's a completely unwarranted assertion. Stiff regulation against counterfeiting currency is successful so that 99.9% of domestic US dollar notes are legitimate. There's a long history of all sorts of cheating which is significantly precluded by regulation and the threat of punishment---why are there 'compliance' officers and legal in banks and brokerages?
| Are they worse than the market makers that preceded them?
Yes. Market makers had to make a market and not pretend to do so or lie about how much they were making a market.
| Is you pension fund suffering because of increased liquidity and reduced spreads?
Yes, because the liquidity is illusionary in institutional amounts as are the spreads.
| Do you seriously think a major player in the market like a pension fund doesn't use these algorithms themselves?
No matter what algorithm you use if your opponent front runs you by technical interception of the communication channel you can never win.
| Do you invest through a fund that doesn't react to changes in the market and update their practices accordingly?
Suppose one team always intercepted the other team's play calls and not vice versa. Obviously the team being intercepted could attempt to mitigate the problem but it's clearly unfair and unfair to blame the victim.
| And your 401k is managed by people so naive as to allow that?
As if saying "I don't allow that" will make it not happen? Do you understand what is happening?
There is no one central 'place' for advertising orders, there are many exchanges and trading zones operating in distinct geographical areas with differing systems.
An advertisement to buy or sell or and order is propagated on the "white circuit", the normal routes the exchanges talk to one another. The HFT's have lower-latency connections and they detect the activity on X and before the investor's signal is propagated to Y, they do their own faster route and move prices in their favor and contrary to the investor's.
The investor doesn't have any control over this! There is no mitigation---as the exchanges make MONEY off the HFT's and share in the proceeds of ripping off the investors one microcent at a time. The exchanges' best customers are HFT's and they lick their boots.
The HFT's also lie. In other words, they fake being a market maker by pretending to offer liquidity---but when there's an actual order that they don't want to fill they detect it and through their lower-latency communication pull it away a millisecond before the order hits going through the normal investor's communication channel.
It's like going to your supermarket. And in the time interval when the cashier picks up your item to scan it there is a secondary laser system which detects the movement, reads the bar code, and raises the price for the actual scan that you have to pay.
| They don't themselves adopt similar technologies and strategies to mitigate that?
The technologies are extremely expensive and often infeasible to use. Why shoudl I have to pay my investment manager extra to mitigate something against avoidable costs?
In order pay additional costs to be passed down to mitigate the effect of a parasite
I felt your pain, but the practical state of C++ is much better today.
A great application which I use is computational code with modern C++ template libraries for the problem domain. Fortran 2003 is also highly competitive (and in some cases superior) here.
What is "Big Data" analytics? Computations at the limit of capacity. Constant factors really do matter---more efficient computations in space and time mean you can ingest more data and that gives better models. Markov Chain Monte Carlo and other stochastically iterative/converging machine learning algorithms are not trivially parallelizable, and doing so induces a very large overhead in programming complexity.
| By introducing "auto" for types, C++ has admitted that it is too complex for human beings to understand.
No, it's admitting that working with libraries and systems without 'auto' require mental effort by humans that shouldn't be necessary to exert.
| The "auto" type declaration means that only the compiler can figure out what to put for a new type. By using it, the programmer is saying "I have no idea what to put here, so I implicitly trust the compiler to do it for me."
Indeed, and there's nothing wrong with that.
Like f(g(something))
The compiler figured out that the static type constraint of the return value of g() is compatible with the argument of f(). Why is that OK (which has been fine since neolithic ages), but
auto z = g(something) f(z)
not OK?
Simple type inference is decades overdue, and it reflects an actual intent of the programmer: "declare a variable which can hold the return type of some function". The point being that with development, if that function's return type changes (or is implicit through complex templtatizations), with C++11 type inference (auto and decltype), the program continues to compile and work cleanly. Without 'auto' and 'decltype' used appropriately the programmer can't express his actual intent.
I use C++11 now and think it's a good improvement. BTW, the Eigen matrix library for C++ is fantastic. I'm a computational programmer at times and I've waited many years for something this good and easy.
Because there are engineering designs for pressurized water reactors which work and decades of experience making them, and molten-salt cycle reactors intrinsically dissolve large amounts of high-level waste in a liquid in normal operation---(water soluble too sometimes)---and make every nuclear plant also a horrifyingly nasty radioactive reprocessing plant.
I'm for fission (not because it's great but because coal and global warming are much worse), but I like my megacuries encased in zirconium, and very solid.
They're terrible engineering designs for some major reasons.
You have an exceptionally radioactive LIQUID in normal operation. Every nuclear plant is a reprocessing plant. Historically reprocessing, meaning chemical engineering of complex radioactive caustic liquids was always the nastiest, ugliest and messiest part.
I do trust locals to run a solid-state reactor if there's a powerful safety regulator permanently on site whose paycheck does not depend on profitability, e.g. somebody who came from the nuclear Navy---where they have a no mistakes philosophy.
But not having every reactor be a chemical reprocessing plant. There Will Be Leaks and screwups.
The small PWRs solve the major safety problem by having a larger surface area to volume ratio, so they cool down better on their own by physics, before the decay heat builds up to melt the structures in the core.
Slashdot Mirror
The traditional key in ignition device is mixing two distinct functions: authentication and mode selection.
The link between the two and the mechanics of the failed device contributed to the problem. Wireless electronic proximity isn't necessary.
What is necessary is a simple reliable device for authentication, a key which is turned to allow other functions to operate. For instance, keys used for industrial or military controls for this purpose. They look like ordinary keys, not the large vehicle keys now common.
Then the selection of ignition mode is entirely separate, and turning to "off" or "accessory" which disengages hydraulics and safety systems while the vehicle is in motion should be disallowed.
| The only reason they're saying something now is because the well publicized Facebook acquisition has made enough noise to catch the attention of the C-level goons that run the place.
who think that because moneybags Facebook is writing large cheques to expensive M&A lawyers and bankers to get the deal to clear quickly that they can be paid off to go away.
It's a win-big/lose-little situation for Zenimax if they already have some lawyers on payroll.
I think my point was made by that very reply.
In truth: Mann's and other's reconstructions have been done again and again with different techniques and the answer comes out almost the same. There was no intentional deception or major technical fault in Mann's original work or that of similar predecessors and successors.
Venus average surface temperature is 735 Kelvin.
Earth surface temperature is about 287 Kelvin.
Remember that outgoing heat flux is in fourth power of absolute temperature, so ignoring atmospheres (black body) Venus would be emitting 43 times as much heat and so would have to be that much closer.
You can't separate the pressure from the temperature and the actual heat flux and hypothetically imagine a '1 atm' pressure on Venus. With a similar atmosphere as Earth you'd have roughly a surface temperature T so that (T/287)^4 = 1.911, difference in solar insolation.
When logic and reason and decades of actual science fail to work against people who are impervious, then what?
The message from Germany is that if you replace nuclear power with coal then more CO2 will be emitted. Well, of course. What climate action advocates favor using more coal? None.
If greenhouse gases emissions were actually taxed, then they wouldn't do that.
Of course there are unscientific 'environmentalists' whose emotional reactions to nuclear power (less safe and clean than solar, more safe and clean than coal) and unwillingness to look at quantitative facts lead them to bad outcomes. Just as climate deniers do.
| well, when deniers talk the science, the believers go into "burn the heretics at the stake" mode.
When deniers lie about the science, the believers in science go into "you are an ignorant idiot and ought to STFU and everybody should ignore your drivel" mode, as they ought to, because the future of technological civilization may be at risk from aggravated ignorance.
The actual physics of Anthropogenic Global Warming (of which anthropogenic CO2 is one but not an exclusive component, and no scare quotes needed as it is fact) is based upon the infrared emissivity of gases and their actual dynamics and concentration in the atmosphere.
This physics is lab validated and confirmed by in-situ objective measurements.
Analogies made to the lay public are imprecise, but the underlying science never was.
HFT's get the heck out of the market in a plunge.
You may have been filled by a statistical arbitrage trading system, which is fine.
| So, what did you expect the SEC Chairperson to say.
"We are disturbed by the allegations, but do not yet have sufficient information to make a firm determiniation. We will use all of our investigatory powers, including subpoena, civil and criminal penalties, with cooperation from all other regulatory agencies and the FBI, to get to the heart of this matter and ensure that our markets are safe and honest."
Given that taxes already depend on holding period, it's not a nightmare.
Firms sure as heck do keep super accurate records of the entire transaction stream because it's their job to do so to pay and get paid! Financial market accounts systems are highly accurate and track every transaction!
A large firm doesn't have unlimited ability to dodge taxes, especially taxes that can be checked against the accounting systems.
It doesn't eliminate the information asymmetry, it makes that asymmetry unprofitable to exploit.
The problem is even easier to solve with even less impact---tax orders canceled within a second.
| The scenarios described in Flash Boys book are not "frontrunning", they're more accurately termed speed arbitrage---they know there is a price discrepancy between different venues and take advantage of it before anyone even notices the price discrepancy.
That's called front running.
Imagine:
In the old days, a broker gets an order from a client. He says loudly to the room, Texas Employees wants to buy IBM in large amounts. He chortles and being rather portly slowly ambles over to the phone to talk to his broker. It so happens that is brokerage has an arrangement so that some young sprightly man is invited to listen in and has living quarters in the office. His job is to listen to the fat man and run over and call up his own trader before the fat guy gets to his. His bosses pay the fat guy's boss.
Is that front-running? Youbetcha!
That's what's happening now with the minor exception that the fat guy goes to 20 different phones connecting to different traders slowly and the kid is stationed at phone #1.
| the amount of money pulled out of the market is less than was pulled out by the market makers under the original manual system when there was trading taking place on an exchange floor.
And how much would be pulled out by modern market makers with computers who don't front run? Automated market making without cheating is a good idea.
| You are crazy. Being charged a fraction of a penny more per share is the most you could claim the effect is. The only problem with that argument is it ignores the liquidity and spread reduction produced just by having HFT in the market.
The HFT's job is to avoid providing liquidity and reducing the spread---that's how they make money. Providing liquidity means trading on the opposite direction of a large investor's trade. The firms which mint money from front-running are trading in the same direction as the investor's trade. They sacrifice a tiny amount in the opposite to trigger their trading, and their apologists call this triviality as 'providing liquidity'.
That can't possibly be 'making a market'---and making a market doesn't require super low-latency connections faster than the exchanges own connections, but front-running does.
Look at the evidence of what the HFT's invest their money in---is there any legitimate need for them for market making?
Real market making involves risk by taking on inventory when other people don't want it and then slowly distributing it back out---it is statistical arbitrage of volume vs time.
As have every successful democratic government yielding a good quality of life to its citizens.
| If the government controls how you can make money, those with money will seek to control the government. I'm not suggesting a solution, I don't have one.
A democratic republic with an independent, free press and a fearless judiciary and investigative service with subpoena and prosecution power is a good start.
| But understanding that "more regulation" will likely do just as much harm as it does good is important to this debate.
No, that's a completely unwarranted assertion. Stiff regulation against counterfeiting currency is successful so that 99.9% of domestic US dollar notes are legitimate. There's a long history of all sorts of cheating which is significantly precluded by regulation and the threat of punishment---why are there 'compliance' officers and legal in banks and brokerages?
| there's really no data to verify whether markets are rigged or not in the way described in the Flash Boys book.
Sure there is! Stop thinking like a nerd.
How did they do it when it was people with telephones who were illegally front-running?
Seize evidence with subpoena power, threaten the peons with criminal prosecution and get perps to talk under oath.
Exchanges give special technical access to HFT's to do the deeds---because HFT's kick them back money.
You see, a perfectly "free" lassiez-faire market, where bribing the referees is called business.
| Are they worse than the market makers that preceded them?
Yes. Market makers had to make a market and not pretend to do so or lie about how much they were making a market.
| Is you pension fund suffering because of increased liquidity and reduced spreads?
Yes, because the liquidity is illusionary in institutional amounts as are the spreads.
| Do you seriously think a major player in the market like a pension fund doesn't use these algorithms themselves?
No matter what algorithm you use if your opponent front runs you by technical interception of the communication channel you can never win.
| Do you invest through a fund that doesn't react to changes in the market and update their practices accordingly?
Suppose one team always intercepted the other team's play calls and not vice versa. Obviously the team being intercepted could attempt to mitigate the problem but it's clearly unfair and unfair to blame the victim.
| And your 401k is managed by people so naive as to allow that?
As if saying "I don't allow that" will make it not happen? Do you understand what is happening?
There is no one central 'place' for advertising orders, there are many exchanges and trading zones operating in distinct geographical areas with differing systems.
An advertisement to buy or sell or and order is propagated on the "white circuit", the normal routes the exchanges talk to one another. The HFT's have lower-latency connections and they detect the activity on X and before the investor's signal is propagated to Y, they do their own faster route and move prices in their favor and contrary to the investor's.
The investor doesn't have any control over this! There is no mitigation---as the exchanges make MONEY off the HFT's and share in the proceeds of ripping off the investors one microcent at a time. The exchanges' best customers are HFT's and they lick their boots.
The HFT's also lie. In other words, they fake being a market maker by pretending to offer liquidity---but when there's an actual order that they don't want to fill they detect it and through their lower-latency communication pull it away a millisecond before the order hits going through the normal investor's communication channel.
It's like going to your supermarket. And in the time interval when the cashier picks up your item to scan it there is a secondary laser system which detects the movement, reads the bar code, and raises the price for the actual scan that you have to pay.
| They don't themselves adopt similar technologies and strategies to mitigate that?
The technologies are extremely expensive and often infeasible to use. Why shoudl I have to pay my investment manager extra to mitigate something against avoidable costs?
In order pay additional costs to be passed down to mitigate the effect of a parasite
I felt your pain, but the practical state of C++ is much better today.
A great application which I use is computational code with modern C++ template libraries for the problem domain. Fortran 2003 is also highly competitive (and in some cases superior) here.
What is "Big Data" analytics? Computations at the limit of capacity. Constant factors really do matter---more efficient computations in space and time mean you can ingest more data and that gives better models. Markov Chain Monte Carlo and other stochastically iterative/converging machine learning algorithms are not trivially parallelizable, and doing so induces a very large overhead in programming complexity.
| By introducing "auto" for types, C++ has admitted that it is too complex for human beings to understand.
No, it's admitting that working with libraries and systems without 'auto' require mental effort by humans that shouldn't be necessary to exert.
| The "auto" type declaration means that only the compiler can figure out what to put for a new type. By using it, the programmer is saying "I have no idea what to put here, so I implicitly trust the compiler to do it for me."
Indeed, and there's nothing wrong with that.
Like f(g(something))
The compiler figured out that the static type constraint of the return value of g() is compatible with the argument of f(). Why is that OK (which has been fine since neolithic ages), but
auto z = g(something)
f(z)
not OK?
Simple type inference is decades overdue, and it reflects an actual intent of the programmer: "declare a variable which can hold the return type of some function". The point being that with development, if that function's return type changes (or is implicit through complex templtatizations), with C++11 type inference (auto and decltype), the program continues to compile and work cleanly. Without 'auto' and 'decltype' used appropriately the programmer can't express his actual intent.
I use C++11 now and think it's a good improvement. BTW, the Eigen matrix library for C++ is fantastic. I'm a computational programmer at times and I've waited many years for something this good and easy.
Because there are engineering designs for pressurized water reactors which work and decades of experience making them, and molten-salt cycle reactors intrinsically dissolve large amounts of high-level waste in a liquid in normal operation---(water soluble too sometimes)---and make every nuclear plant also a horrifyingly nasty radioactive reprocessing plant.
I'm for fission (not because it's great but because coal and global warming are much worse), but I like my megacuries encased in zirconium, and very solid.
They're terrible engineering designs for some major reasons.
You have an exceptionally radioactive LIQUID in normal operation. Every nuclear plant is a reprocessing plant. Historically reprocessing, meaning chemical engineering of complex radioactive caustic liquids was always the nastiest, ugliest and messiest part.
I do trust locals to run a solid-state reactor if there's a powerful safety regulator permanently on site whose paycheck does not depend on profitability, e.g. somebody who came from the nuclear Navy---where they have a no mistakes philosophy.
But not having every reactor be a chemical reprocessing plant. There Will Be Leaks and screwups.
The small PWRs solve the major safety problem by having a larger surface area to volume ratio, so they cool down better on their own by physics, before the decay heat builds up to melt the structures in the core.
Precisely.
There are certain companies whose core competency is engineering around the government procurement process.
Then there's SpaceX, whose core competency is rocket engineering.