Cash is already phased out. Take $12,000 out of your bank. Oh, that's right. You can't. They don't have that much. You have to call ahead....
That may be true for law abiding citizens, but consider the drug trade. The people who grow opium in Afgahanistan or refine heroin in Columbia want to get paid from the cash sales to users in the U.S. and Europe. That mean a net outflow of funds from industrialized nations to third world countries to pay for the drugs. All international terrorists have to do is pay the suppliers in Afghanistan, Columbia etc. and have their agents collect cash from dealer networks inside the U.S. and Europe, maybe offering a discount relative to other ways of getting cash out. The cash itself does not have to move far at all, certainly not across borders. Given the huge scale of international drug trafficking, only a tiny diversion of its cash flow is required to keep local terrorist cells comfortably in business.
I found that developers who staffed our booth at trade shows often retuned with renewed enthusiasm and ideas for improving our products, based on talking to potential customers and seeing competitive products.
They weren't the same size as telco racks (i.e. 19")?
No they were not. IBM had it's own standards for mechanical packaging. But my recollection is that IBM's requirement that all its products fit through a standard 29" door predated the 360 line and was mandated by their sales department, who never wanted to lose a sale because the product could not get into a building.
The 026 didn't have parentheses either; we had to multipunch them.
That's not the whole story. Prior to EBCDIC, the convention was to use % and lozenge in the commercial 026 character set to represent left and right parentheses and # to represent equals. IBM offered 026 keypunches with those characters replaced on the keyboard and print matrix and compilers understood them as "(", ")" and "=". It was called the scientific character set option. When EBCDIC was introduced, the commercial card codings were retained and new card codes introduced for "(", ")", "=" and many others. But compilers I worked with all recognized the older codings, at least until I stopped using keypunches in the mid 1970s. But if you had a need to enter a proper EBCDIC "(", ")" or "=" on an 026 you would have had to multi-punch, as you say. Was there some software you used that did not recognize the "scientific" 026 encodings?
what you are deliberately leaving out is that OS X has a fraction of the marketshare of windows and that is the main reason.
If smaller marketshare is the main reason OS X has much less malware than Windows, isn't that still a compelling reason to buy a Mac? Let all the cheapskates who want to save a few hundred bucks on their computer deal with the mass insecurity.
Back when Fortran came out, keyboards weren't standardized yet. Also, some systems used variations of 7-bit or even non-ASCII encodings like EBCDIC....
Uhh, not quite. Back when Fortran was introduced in 1957 and when comparison operators were introduced in Fortran IV in 1962, pretty much everyone was entering programs on punched cards, mostly using the IBM 026 keypunch machine which did not have the greater than and less than symbols. In fact a special 026 "scientific character set" was required to show parentheses or equal sign on the keyboard or to print them on the cards. While each computer model tended to have its own internal character code, the punched card code itself was well standardized, even among IBM's competitors. The EBCDIC character set, which does have greater than and less than symbols, was introduced with the IBM System/360 in the mid-1960s, along with the IBM 129 keypunch, but the less expensive (and indestructible) 026 continued to be used by programmers for many years thereafter, especially at academic institutions. There is a good picture of the 026 keyboard on Wikipedia: File:IBM 026 keyboard.mw.jpg
I agree with most of that, but I never had much problem keeping my slide rules lubricated, aligned, and tensioned. Slide rule maintenance was a lot less hassle than keeping a cell phone charged and the software updated. I used a slide rule through college and into my first job, but jumped at buying a TI SR-50 scientific calculator when they first came out in 1973, also paying about $300. But the tactile feel for calculations that the slide rule provided has never left me. I still own mine and have a slide rule app on my iPhone, just for fun. If you've never used one, get one on eBay and give it a try.
Point taken, I should have said zero ongoing energy cost. But the cost of the solar panels should be compared to the cost of the equipment generating power for conventional bitcoin mining and the cost of constructing the transmission lines bringing that power to the data center. And solar is getting to be competitive on a cost per watt-hour at the source.
An interesting architecture might combine a processing unit with a solar panel on a simple stand. Large numbers of these units could could then be placed in a desert area, communicating via a mesh network, with almost no additional infrastructure, not even roads. Coining cryptocurrency and code breaking do not require much interprocessor communication. Energy costs would be zero. There would be no need to locate near a power transmission grid, so the most desolate desert areas might be suitable, minimizing land costs. Robotic dune buggies or drones could be used to dust off the panels periodically. and remove units as needed for repair or hardware upgrade. If batteries get cheap enough, they could be incorporated to allow processing to continue at night, perhaps at a faster clock due to easier cooling. Human staffing might be limited to security and backup maintenance.
Read her stuff. She got it. Here is a quote from a review by Countess Lovelace of an article about the Analytical Engine by an Italian engineer, Menabrea, that neatly sums up the multidisciplinary nature of computer architecture:
"We refer the reader to the ‘Edinburgh Review’ of July 1834, for a very able account of the Difference Engine. The writer of the article we allude to has selected as his prominent matter for exposition, a wholly different view of the subject from that which M. Menabrea has chosen. The former chiefly treats it under its mechanical aspect, entering but slightly into the mathematical principles of which that engine is the representative, but giving, in considerable length, many details of the mechanism and contrivances by means of which it tabulates the various orders of differences. M. Menabrea, on the contrary, exclusively developes the analytical view; taking it for granted that mechanism is able to perform certain processes, but without attempting to explain how; and devoting his whole attention to explanations and illustrations of the manner in which analytical laws can be so arranged and combined as to bring every branch of that vast subject within the grasp of the assumed powers of mechanism. It is obvious that, in the invention of a calculating engine, these two branches of the subject are equally essential fields of investigation, and that on their mutual adjustment, one to the other, must depend all success. They must be made to meet each other, so that the weak points in the powers of either department may be compensated by the strong points in those of the other. They are indissolubly connected, though so different in their intrinsic nature, that perhaps the same mind might not be likely to prove equally profound or successful in both. ”
A human mission to Phobos would make more sense if we placed many more robotic probes on Mars' surface first. The astronauts on Phobos could then control them in near real time, allowing parallel exploration of many locations the planet's surface, rather than a single landing spot.
People like me who don't drink and drive pay higher insurance rates to subsidize those drivers that do. Think about it, the insurance exposure for one drunk going home from a bar in the back seat of an Uber vehicle is far less than the insurance exposure for that same drunk behind the wheel of a car driving home. I'd much rather subsidize the added risk of Uber drivers than the risk the drunks they carry.
Would I make a major enterprise purchase based on a Slashdot discussion? Absolutely not. Would I want to read a Slashdot discussion and maybe follow suggested links and look up all the buzzwords BEFORE talking to vendors or consultants? Absolutely.
Given that the entire corpus of State Department cables classified up to Secret was leaked by Private Manning and that the State Department's unclassified email system was so badly hacked they struggled to get it cleared (if they have) and that other sensitive government systems, like the OPM database of security clearance records, have been completely compromised, there is reason to think Clinton's use of her own server may have provided better protection for sensitive information than official government channels. It could hardly have been worse.
MUMPS started on the 18-bit PDP-7/9/15 series and kept the ponderous PDP-15 alive for years after other uses had switched to minicomputers like the PDP-11.
While standards would be great going forward, it does not follow that existing satellites can't be repaired. Satellites have to be fueled before launch and their makers do not design unique fuel fill ports for each satellite they build. There are undoubtedly a few common sizes and it should not be that difficult to equip a repair station with a set of adaptors or a universal fill device that can work with a large fraction of existing fill ports. Same goes for grappling points. All satellites have fittings that attach them to the launch vehicle. Again there are a limited number of designs and a robot arm could be designed to grab many if not most of them.
The software and internal state of robot A can be passed to robot B in an encrypted file over the Internet, carrying the robot's "Identity" with it. So border controls like a passport are meaningless.
>... Why would you think Swift is appropriate to "replace" C++ or Java?
Java is designed to be interpreted, which makes is less suitable as a system programming language. C and C++ are full of pitfalls for programmers. Arguably the C languages are a major source of the world's increasing computer insecurity. Swift is designed to be compiled and it avoids many of the C family's problems, while working with C family libraries. Someone really good with strong support from a major player has tried to get it right. The Apple community seems to think he got close. Time will tell if any other community agrees.
For any readers not in the programming community, he is of course kidding: there are dozens of programming languages in active use, far too many. The problem is that there is no one modern programming language that has gotten wide adoption. Swift is taking over Apple's rapidly growing software ecosystem, giving Swift momentum. Open sourcing is a good first step toward making Swift a candidate for replacing C++ and Java, but they will need to do more. I'd like to see a Swift development environment for small stand alone microprocessors with a tie-in to Apple's HomeKit. That and some good security tools could make Swift the language of choice for the Internet of Things.
There is no excess capacity on the Amtrak/New Jersey Transit lines. A project is underway (East Side Access) that will connect Grand Central with the JFK Airtrain people mover, but it's late and massively expensive. New York State is proposing a similar people mover to connect Laguardia with the subway and commuter rail. Extending that elevated people mover to link with JFK is a more realistic option since it could use air rights over highways. This could allow JFK to back up Laguardia, for example.
Ancient people threw their trash into holes as well. It is a problem archeologists deal with all the time and one of the reasons sites are excavated and documented very carefully.
I think you are missing the point. When the pullout tabs were phased out in favor of tabs that stay with the can, I remember thinking that a thousand years from now discarded pullout tabs will be a valuable archeological resource. They are distinctive, ubiquitous, and indestructible, and because they were only used during a short time, they would conclusively date any architectural layer they were found in. Maybe modern circuit boards with their date coded components will serve a similar purpose. I wonder what it would take to get current manufacturers to emboss a year code in can tops or in IC dies? Make trash serve history.
Leavitt's discovery is on a par with Galileo's discovery of the moons of Jupiter in changing our view of the universe. It combined a brilliant insight, that all the stars in the Small Magellanic Cloud are at about the same distance from Earth, and a lot of hard work analyzing photographic plates. Her measurement tool is still the main one used to determine cosmic distances from beyond the range of stellar parallax out to nearby galaxies, and is used, in turn, to calibrate Type Ia supernova, the standard candle for probing deep into intergalactic space, and back to the Big Bang.
Another example of women astronomers getting less recognition than they deserve is Jocelyn Bell, who discovered millisecond pulsars and whose thesis advisor won the Nobel Prize for that discovery.
"But that is only a theoretical advantage, not a real one, since the current demand curve for electric power fits the production curve of PV quite well."
That's not true. Peak consumption is around 7 pm, after solar drops off. Visit the California ISO site caiso.com and look at their renewable graphs. Being able to provide power a few hours after sunset is a big win.
A much better source is Cal Iso, which runs the California grid and publishes a graph of demand and sources every day. http://content.caiso.com/green... The peak power use is generally around 7 pm, after solar production has stopped. Wind output various greatly from day to day.
Slashdot Mirror
Cash is already phased out. Take $12,000 out of your bank. Oh, that's right. You can't. They don't have that much. You have to call ahead. ...
That may be true for law abiding citizens, but consider the drug trade. The people who grow opium in Afgahanistan or refine heroin in Columbia want to get paid from the cash sales to users in the U.S. and Europe. That mean a net outflow of funds from industrialized nations to third world countries to pay for the drugs. All international terrorists have to do is pay the suppliers in Afghanistan, Columbia etc. and have their agents collect cash from dealer networks inside the U.S. and Europe, maybe offering a discount relative to other ways of getting cash out. The cash itself does not have to move far at all, certainly not across borders. Given the huge scale of international drug trafficking, only a tiny diversion of its cash flow is required to keep local terrorist cells comfortably in business.
I found that developers who staffed our booth at trade shows often retuned with renewed enthusiasm and ideas for improving our products, based on talking to potential customers and seeing competitive products.
They weren't the same size as telco racks (i.e. 19")?
No they were not. IBM had it's own standards for mechanical packaging. But my recollection is that IBM's requirement that all its products fit through a standard 29" door predated the 360 line and was mandated by their sales department, who never wanted to lose a sale because the product could not get into a building.
The 026 didn't have parentheses either; we had to multipunch them.
That's not the whole story. Prior to EBCDIC, the convention was to use % and lozenge in the commercial 026 character set to represent left and right parentheses and # to represent equals. IBM offered 026 keypunches with those characters replaced on the keyboard and print matrix and compilers understood them as "(", ")" and "=". It was called the scientific character set option. When EBCDIC was introduced, the commercial card codings were retained and new card codes introduced for "(", ")", "=" and many others. But compilers I worked with all recognized the older codings, at least until I stopped using keypunches in the mid 1970s. But if you had a need to enter a proper EBCDIC "(", ")" or "=" on an 026 you would have had to multi-punch, as you say. Was there some software you used that did not recognize the "scientific" 026 encodings?
what you are deliberately leaving out is that OS X has a fraction of the marketshare of windows and that is the main reason.
If smaller marketshare is the main reason OS X has much less malware than Windows, isn't that still a compelling reason to buy a Mac? Let all the cheapskates who want to save a few hundred bucks on their computer deal with the mass insecurity.
Back when Fortran came out, keyboards weren't standardized yet. Also, some systems used variations of 7-bit or even non-ASCII encodings like EBCDIC. ...
Uhh, not quite. Back when Fortran was introduced in 1957 and when comparison operators were introduced in Fortran IV in 1962, pretty much everyone was entering programs on punched cards, mostly using the IBM 026 keypunch machine which did not have the greater than and less than symbols. In fact a special 026 "scientific character set" was required to show parentheses or equal sign on the keyboard or to print them on the cards. While each computer model tended to have its own internal character code, the punched card code itself was well standardized, even among IBM's competitors. The EBCDIC character set, which does have greater than and less than symbols, was introduced with the IBM System/360 in the mid-1960s, along with the IBM 129 keypunch, but the less expensive (and indestructible) 026 continued to be used by programmers for many years thereafter, especially at academic institutions. There is a good picture of the 026 keyboard on Wikipedia: File:IBM 026 keyboard.mw.jpg
I agree with most of that, but I never had much problem keeping my slide rules lubricated, aligned, and tensioned. Slide rule maintenance was a lot less hassle than keeping a cell phone charged and the software updated. I used a slide rule through college and into my first job, but jumped at buying a TI SR-50 scientific calculator when they first came out in 1973, also paying about $300. But the tactile feel for calculations that the slide rule provided has never left me. I still own mine and have a slide rule app on my iPhone, just for fun. If you've never used one, get one on eBay and give it a try.
Point taken, I should have said zero ongoing energy cost. But the cost of the solar panels should be compared to the cost of the equipment generating power for conventional bitcoin mining and the cost of constructing the transmission lines bringing that power to the data center. And solar is getting to be competitive on a cost per watt-hour at the source.
An interesting architecture might combine a processing unit with a solar panel on a simple stand. Large numbers of these units could could then be placed in a desert area, communicating via a mesh network, with almost no additional infrastructure, not even roads. Coining cryptocurrency and code breaking do not require much interprocessor communication. Energy costs would be zero. There would be no need to locate near a power transmission grid, so the most desolate desert areas might be suitable, minimizing land costs. Robotic dune buggies or drones could be used to dust off the panels periodically. and remove units as needed for repair or hardware upgrade. If batteries get cheap enough, they could be incorporated to allow processing to continue at night, perhaps at a faster clock due to easier cooling. Human staffing might be limited to security and backup maintenance.
Read her stuff. She got it. Here is a quote from a review by Countess Lovelace of an article about the Analytical Engine by an Italian engineer, Menabrea, that neatly sums up the multidisciplinary nature of computer architecture: "We refer the reader to the ‘Edinburgh Review’ of July 1834, for a very able account of the Difference Engine. The writer of the article we allude to has selected as his prominent matter for exposition, a wholly different view of the subject from that which M. Menabrea has chosen. The former chiefly treats it under its mechanical aspect, entering but slightly into the mathematical principles of which that engine is the representative, but giving, in considerable length, many details of the mechanism and contrivances by means of which it tabulates the various orders of differences. M. Menabrea, on the contrary, exclusively developes the analytical view; taking it for granted that mechanism is able to perform certain processes, but without attempting to explain how; and devoting his whole attention to explanations and illustrations of the manner in which analytical laws can be so arranged and combined as to bring every branch of that vast subject within the grasp of the assumed powers of mechanism. It is obvious that, in the invention of a calculating engine, these two branches of the subject are equally essential fields of investigation, and that on their mutual adjustment, one to the other, must depend all success. They must be made to meet each other, so that the weak points in the powers of either department may be compensated by the strong points in those of the other. They are indissolubly connected, though so different in their intrinsic nature, that perhaps the same mind might not be likely to prove equally profound or successful in both. ”
A human mission to Phobos would make more sense if we placed many more robotic probes on Mars' surface first. The astronauts on Phobos could then control them in near real time, allowing parallel exploration of many locations the planet's surface, rather than a single landing spot.
People like me who don't drink and drive pay higher insurance rates to subsidize those drivers that do. Think about it, the insurance exposure for one drunk going home from a bar in the back seat of an Uber vehicle is far less than the insurance exposure for that same drunk behind the wheel of a car driving home. I'd much rather subsidize the added risk of Uber drivers than the risk the drunks they carry.
Would I make a major enterprise purchase based on a Slashdot discussion? Absolutely not. Would I want to read a Slashdot discussion and maybe follow suggested links and look up all the buzzwords BEFORE talking to vendors or consultants? Absolutely.
Given that the entire corpus of State Department cables classified up to Secret was leaked by Private Manning and that the State Department's unclassified email system was so badly hacked they struggled to get it cleared (if they have) and that other sensitive government systems, like the OPM database of security clearance records, have been completely compromised, there is reason to think Clinton's use of her own server may have provided better protection for sensitive information than official government channels. It could hardly have been worse.
MUMPS started on the 18-bit PDP-7/9/15 series and kept the ponderous PDP-15 alive for years after other uses had switched to minicomputers like the PDP-11.
While standards would be great going forward, it does not follow that existing satellites can't be repaired. Satellites have to be fueled before launch and their makers do not design unique fuel fill ports for each satellite they build. There are undoubtedly a few common sizes and it should not be that difficult to equip a repair station with a set of adaptors or a universal fill device that can work with a large fraction of existing fill ports. Same goes for grappling points. All satellites have fittings that attach them to the launch vehicle. Again there are a limited number of designs and a robot arm could be designed to grab many if not most of them.
The software and internal state of robot A can be passed to robot B in an encrypted file over the Internet, carrying the robot's "Identity" with it. So border controls like a passport are meaningless.
> ... Why would you think Swift is appropriate to "replace" C++ or Java?
Java is designed to be interpreted, which makes is less suitable as a system programming language. C and C++ are full of pitfalls for programmers. Arguably the C languages are a major source of the world's increasing computer insecurity. Swift is designed to be compiled and it avoids many of the C family's problems, while working with C family libraries. Someone really good with strong support from a major player has tried to get it right. The Apple community seems to think he got close. Time will tell if any other community agrees.
For any readers not in the programming community, he is of course kidding: there are dozens of programming languages in active use, far too many. The problem is that there is no one modern programming language that has gotten wide adoption. Swift is taking over Apple's rapidly growing software ecosystem, giving Swift momentum. Open sourcing is a good first step toward making Swift a candidate for replacing C++ and Java, but they will need to do more. I'd like to see a Swift development environment for small stand alone microprocessors with a tie-in to Apple's HomeKit. That and some good security tools could make Swift the language of choice for the Internet of Things.
There is no excess capacity on the Amtrak/New Jersey Transit lines. A project is underway (East Side Access) that will connect Grand Central with the JFK Airtrain people mover, but it's late and massively expensive. New York State is proposing a similar people mover to connect Laguardia with the subway and commuter rail. Extending that elevated people mover to link with JFK is a more realistic option since it could use air rights over highways. This could allow JFK to back up Laguardia, for example.
Ancient people threw their trash into holes as well. It is a problem archeologists deal with all the time and one of the reasons sites are excavated and documented very carefully.
I think you are missing the point. When the pullout tabs were phased out in favor of tabs that stay with the can, I remember thinking that a thousand years from now discarded pullout tabs will be a valuable archeological resource. They are distinctive, ubiquitous, and indestructible, and because they were only used during a short time, they would conclusively date any architectural layer they were found in. Maybe modern circuit boards with their date coded components will serve a similar purpose. I wonder what it would take to get current manufacturers to emboss a year code in can tops or in IC dies? Make trash serve history.
Leavitt's discovery is on a par with Galileo's discovery of the moons of Jupiter in changing our view of the universe. It combined a brilliant insight, that all the stars in the Small Magellanic Cloud are at about the same distance from Earth, and a lot of hard work analyzing photographic plates. Her measurement tool is still the main one used to determine cosmic distances from beyond the range of stellar parallax out to nearby galaxies, and is used, in turn, to calibrate Type Ia supernova, the standard candle for probing deep into intergalactic space, and back to the Big Bang. Another example of women astronomers getting less recognition than they deserve is Jocelyn Bell, who discovered millisecond pulsars and whose thesis advisor won the Nobel Prize for that discovery.
"But that is only a theoretical advantage, not a real one, since the current demand curve for electric power fits the production curve of PV quite well." That's not true. Peak consumption is around 7 pm, after solar drops off. Visit the California ISO site caiso.com and look at their renewable graphs. Being able to provide power a few hours after sunset is a big win.
A much better source is Cal Iso, which runs the California grid and publishes a graph of demand and sources every day. http://content.caiso.com/green... The peak power use is generally around 7 pm, after solar production has stopped. Wind output various greatly from day to day.