Not really. Docking has been done automatically, manually by an onboard pilot, and remotely. Russian spacecraft since 1985 have used the Kurs system (which they now have to buy from Ukraine, at a somewhat inflated price). That's a full-auto, straight-on approach system, and has a good track record. The US used onboard pilot control for final shuttle docking. The Dragon spacecraft was remotely controlled into a close position to the ISS, then grabbed with the robot arm.
Docking can be complicated. Until the craft are very close and have a very low relative speed, the orbital mechanics dominate the problem. So the maneuvering craft has to be aligned in the direction for firing its maneuvering engine for orbit adjustment.
Early docking (Gemini, Apollo, etc.) was done with the target spacecraft aligned with its own orbit, so that the maneuvering craft didn't have to reorient to dock.
Docking with something big, like the ISS, requires attitude changes and final maneuvering just before docking. Here's a video of a Soyuz docking with the ISS, using one of the less favorable docking locations because two other craft were docked already.
There is absolutely no way some guy firing rounds into any kind of grass caused a fire. I want proof.
See the Salt Lake City Tribune article. The people who started the fire by shooting tried to put it out, then called 911 to report the fire. But by then it was too late.
They may not be subject to criminal charges, but they still have civil liability. They'll be sued for millions of dollars.
It's a slightly different analysis of an idea that's been proposed before. The genera idea is to divert annoying asteroids slightly by firing lasers at them from reasonably close range to boil off some surface. This is a long, slow process, taking years. It has to be done from a great enough distance that the stuff being boiled off doesn't mess up the optics, but not so far off that enough energy can't be delivered to a narrow spot to boil rock. The paper is just an analysis of the size and number of spacecraft required, assuming a solar power system driving a solid state laser. (Why not just focus the sun? The mirror has to be too close to the asteroid to get a tight enough beam, and then it gets hit by the rock being boiled off.)
As a weapon system, it's not very useful. It's too expensive and vulnerable for an anti-satellite weapon. The beam might be able to deliver enough energy through the atmosphere to set fires, but the spacecraft would have to be put in low earth orbit to do it.
Robots are starting to work in unstructured situations. I was there at the moment when this was recognized - the 2005 DARPA Grand Challenge at the California Motor Speedway in Fontana, CA. That's when everything changed.
The 2004 Grand Challenge, remember, was a pathetic joke. No vehicle got further than 7 miles, and that was CMU's. The CMU approach at the time wasn't even really autonomous. Entrants got the route on a CD an hour or so before the start. CMU had imagery of the whole area and tried to plan obstacle avoidance manually just before the start, using a huge team of people in a semitrailer full of workstations. Didn't work; the DoD people in charge had moved some obstacles during the night. And that was the best result. One vehicle came out of the gate, turned hard, and ran back into the starting gate. One flipped over. The big Oskosh entry demolished a SUV parked as an obstacle to be avoided. The whole thing was embarrassing.
DARPA was very displeased with the performance by the universities that had long been receiving DARPA funding for robotics. It was quietly made clear to some major CS departments that their performance had to improve or funding would be cut off. That's why entire CS departments were suddenly devoted to the DARPA Grand Challenge in 2005.
In 2005, things were completely different. Everybody who got that far had already been through an elimination, and every vehicle at the 2005 challenge was better than any of the 2004 entries. There was considerable press coverage, and at first, the press treated it as a joke. But suddenly there were over 20 vehicles running around autonomously, and they weren't crashing into stuff. When multiple vehicles finished the course, it was viewed as a triumph.
Finally, the state of the art had reached the point that money and determination would get problems solved. That wasn't true in the 1980s. NASA threw over $100 million at the Flight Telerobotic Servicer project, and got nothing that worked.
Now check out the DARPA Humanoid Challenge. (There's much dreck about this on blogs and in the popular press. Read the DARPA announcement instead.) They have an approach that's likely to work, and demand simulated demos (in their simulator) in 9 months, with demos on real hardware in 18 months. I personally think they'll get something able to do most of the mobility tasks and some of the manipulation tasks in that time. Useful humanoid robots will be a lot closer in two years.
Price will still be a problem. But not an unsolveable one. These things could be brought down to the price of an SUV, if not lower, through production economies alone. The parts count is probably lower than that for an SUV.
Yes, it's slow. The code is in Python and it's still experimental. I've heard from the Willow Garage people that they've speeded up towel folding 50x since the 2010 demo. Once you can do it at all, it can be done faster and cheaper.
Back in 1947, the present rules of presidential succession were set up. The present line of succession has 18 people. That ought to be enough.
When this is a real worry, a few of those people should be in a bunker. During presidential inaugurations and presidential speeches to Congress, that's actually done; at least one person in the line of succession is in a safe place far away. It's usually someone far down the list, but in 2001 Dick Cheney (VP) was sent to the "undisclosed location", and in 2003, Ashcroft (AG) got bunker duty. In 2005, 2006, and 2007, the president pro tem of the Senate went to the bunker. In 2009, Holder (the AG) got the duty. Since then, after most of a decade with no significant terrorist attacks, it's back to the low-rankers.
In terms of actual threat, nobody in the US presidential line of succession has ever been assassinated.
This is a problem for which a solution was implemented long ago, back when a major war looked like a likely possibility.
This might be useful for detecting people carrying concealed guns. It's known that when people wearing a big dense object step up or down (a curb is sufficient) there's motion that can be noticed. Some cops are trained for this. Now it can be automated.
I've heard this at Dolby's screening room in SF.
It looks like a modest auditorium. It's really a money-is-no-object demo facility. Before a talk on another subject, the Dolby guys couldn't resist showing off. They had a video game with many directional sound outputs hooked into the room's systems, and you could hear the players moving around in the space, behind and above the audience when appropriate. You really can hear somebody sneaking up on you in-game from a platform above you.
It's an experience to hear many-channel sound in a facility like that, but few (if any) commercial theaters are that good acoustically. Unless the room acoustics are very, very good, all those channels won't help much.
We've been generating that list for years. It's based on PhishTank data, updated every 3 hours, and uses Open Directory to decide if a site is "major". 46 domains are on the list today. 9 have been on the list since 2011 or earlier. One has been on the list since 2010 - Google.
Google is the last free hosting service unable to clean up their phishing problem. MSN, Yahoo, and various free hosting services have been successful at aggressively cleaning up phishing problems, and haven't been on this list, other than briefly, for years.
For years, Google didn't realize that Google Spreadsheets could be used to host phishing sites. They finally caught on, and there's now a "report abuse" button on spreadsheets. Most, but not all, of the spreadsheet-hosted phishing sites have been taken down.
If anybody from Google is reading this, go over to your abuse department and apply a clue stick. It should embarrass someone that Google is the most clueless free hosting provider in the world about phishing.
As I point out occasionally, 3D printing isn't all that great. TechShop has both a little MakerBot and a commercial stereolithography machine. I've never seen anything come out of the MakerBot but simple decorative plastic objects, and mostly it sits idle. The commercial machine makes good quality objects, but the process is so slow that the expense is high.
There's this mindset that additive machining is some magic process comparable to a SF matter duplicator. It's not. It's far slower and much more expensive than mass production. Too many people today have never been on a factory floor.
If you want to see something that looks like matter duplication, check out a high speed injection molding line or a small parts stamping plant.
You mean studios will finally be able to pan at a reasonable speed without it looking jittery and fucking terrible?
24 fps is terrible and you should feel bad for propping up a dying standard.
Cameron ("Titanic", "Avatar") is pushing hard for 48FPS. His work has pans over exquisitely detailed backgrounds. That looks awful at 24FPS unless the pans are kept very slow. Or worse, some blur is applied, which is a common solution in action movies and is why many action backgrounds are out of focus. (Well, sometimes the backgrounds are out of focus to hide flaws background art and set construction.)
Action movies are going to look much crisper at higher frame rates.
It's a terrible idea for a company to do this. A company can block access to sites via HTTPS on their own network if they wish. Breaking the encryption and snooping, though, creates liability for the company. There is an expectation of privacy associated with an encrypted connection.
If an employee's legitimate online banking activity (for example, making sure their pay was deposited) results in a security breach, the employer would be liable. The employer may be guilty of an HIPAA violation. If they snoop on an employee's communication with their union or a Government agency, they may violate other laws.
There's been discussion on the Mozilla security list over whether Firefox should raise alarms if it detects a wildcard cert. The consensus seems to be "yes, it should". Mozilla policy is moving towards kicking CA out of the root list if they issue wildcard certs, and adding technical measures to prevent them from working.
Although it's been proposed many times, nobody has ever put up a small, reusable manned spacecraft. The USAF had the DynaSoar program in the 1960s, but that was cancelled. Virgin Galactic is making noises about a small orbital spaceplane. Nothing like that has ever flown, but there's no fundamental obstacle.
The near future of earth orbit space may be Space-X's Falcon Heavy for freight, something from Virgin Galactic for humans, and robotic vehicles for military tasks.
Except that any such taps are instantly detectable, at which point communication stops. Thus, at most 1 bit of information leaks out to an eavesdropper.
That's a property of quantum communication, where observation affects the result. Not this.
It's going to be tough. There's too much in that book to cram into a movie, and most of it contributes to the main plot. What to cut?
Probably most of the virtual reality. VR was more promising in 1992 than it is now. It's been way overdone in movies. Show Hiro in gloves and goggles gear in his storage space, and others briefly in similar gear when appropriate, but spend little screen time on VR.
Use Juanita Marquez, Hiro's ex-girlfriend and linguist/mythologist , as the designated explainer for the psycho-religious stuff. Somebody has to do that job.
If they're lucky, they might be able to get Chloe Grace Moretz ("Hit Girl") as Y.T. That's the toughest casting decision. Any of the usual big hunks can play Raven. A number of older actors could play Uncle Enzo. Ng is a CG character. No idea who should play Hiro.
As someone pointed out, this was on Slashdot 7 years ago. Here's the referenced paper.
The idea is simple. At both ends of the wire, random data modulated with content is being emitted. At any point on the wire, you see the sum of two random sources. But each end knows their own random data, and can subtract it out.
To break the system, you need two taps on the wire, some distance apart. Now you get to see the sums of the signals from each end, but with different time shifts between them due to propagation delay. With that data, you can separate out what's coming from each end. This allows recovering the original signals.
"No new encryption system is worth looking at unless it comes from someone who has already broken a very hard one." - Friedman.
You know, we had the same argument with RISC versus CISC architecture. And we know who lost that one. Badly.
The real question there is, do you want to go superscalar? Sequential RISC CPUs are simpler than sequential CISC CPUs, but once you have pipelines and multiple execution units, there's so much added complexity and transistor count that the difference disappears. If you're willing to have a slow RISC CPU, the transistor count can be quite low. Down at the bottom, where there's on-chip memory and no cache, as with Atmel ATMega CPUs, the transistor count is really low.
ARM started down there, but it's been built up to a serious level of computing power, with multiple cores, layers of caches, MMUs, and all the stuff of a desktop CPU. Once you have all that stuff, the instruction decoder part is a tiny fraction of the transistor count.
(Of course, x86 superscalar machines require the retirement unit from hell to manage all the hard cases, like self-modifying code. I still remember going to the talk where the Intel guy in charge of the Pentium Pro program explained how they did that. 3000 engineers on the design team at peak. One of the people at that talk was the guy who designed the Intel 8051 mostly by himself.)
âoeAt an academic level, it's a very different background,â says Bobby Schnabel, Dean of the School of Informatics at Indiana University and chair of the ACMâ(TM)s Education Policy Committee. "The calculus and differential equations that underlie engineering are not what underlies computer science. It's really discrete mathematics."
That was true a few decades ago. Today, though, all that discrete math isn't as useful. Today, you need calculus and Bayesian statistics for machine learning. You need differential equations and computational geometry for game development and robotics. Number theory, mathematical logic, graph theory, and automata theory just aren't that important any more. Most of what's needed from those fields is now embodied in well-known algorithms.
I got all the classic discrete math training, but over the years, I've had to use far more number-crunching math.
Get your company's lawyers involved in negotiating with "cloud" providers. Make sure all the things that the "cloud" provider can screw up result in substantial financial penalties. Lawyers are paid to prepare for contingencies like that. If a "cloud" provider won't agree to enforceable service agreements, price out business interruption insurance coverage for the cloud provider's failure. Now you have a backup plan and costing for it.
Von Neumann was much more influential than Turing. Not only did von Neumann do brilliant work in multiple areas of mathematics, he invented modern computer architecture. Babbage's design was more like a Jacquard loom card reader coupled to a calculator. Turing's theoretical machine had to roll a long tape back and forth, and the cryptographic machines were essentially hard-wired or plugboard-programmed. Those machines are closer in concept to Hollerith/IBM tabulators of the 1920s to 1950s.
Von Neumann got computer architecture right. He saw that the right answer was RAM, with programs and data in the same memory: The device requires a considerable memory. While it appeared
that various parts of this memory have to perform functions which differ somewhat in their nature
and considerably in their purpose, it is nevertheless tempting to treat the entire memory as one
organ, and to have its parts even as interchangeable as possible for the various functions enumerated
above."
He also figured out that 1) everything inside the machine should be binary, not decimal, 2) memory sizes should be a power of two, 3) about 2^18 bits of RAM were needed to get any useful work done, 4) delay-line memory would work in the short term, but "iconoscope" memory (see Williams tube), which is random access, would be better, and 5) what a reasonable instruction set should look like.
Slashdot Mirror
Not really. Docking has been done automatically, manually by an onboard pilot, and remotely. Russian spacecraft since 1985 have used the Kurs system (which they now have to buy from Ukraine, at a somewhat inflated price). That's a full-auto, straight-on approach system, and has a good track record. The US used onboard pilot control for final shuttle docking. The Dragon spacecraft was remotely controlled into a close position to the ISS, then grabbed with the robot arm.
Docking can be complicated. Until the craft are very close and have a very low relative speed, the orbital mechanics dominate the problem. So the maneuvering craft has to be aligned in the direction for firing its maneuvering engine for orbit adjustment. Early docking (Gemini, Apollo, etc.) was done with the target spacecraft aligned with its own orbit, so that the maneuvering craft didn't have to reorient to dock.
Docking with something big, like the ISS, requires attitude changes and final maneuvering just before docking. Here's a video of a Soyuz docking with the ISS, using one of the less favorable docking locations because two other craft were docked already.
This recent Chinese dock was aligned with the orbit.
There is absolutely no way some guy firing rounds into any kind of grass caused a fire. I want proof.
See the Salt Lake City Tribune article. The people who started the fire by shooting tried to put it out, then called 911 to report the fire. But by then it was too late.
They may not be subject to criminal charges, but they still have civil liability. They'll be sued for millions of dollars.
It's a slightly different analysis of an idea that's been proposed before. The genera idea is to divert annoying asteroids slightly by firing lasers at them from reasonably close range to boil off some surface. This is a long, slow process, taking years. It has to be done from a great enough distance that the stuff being boiled off doesn't mess up the optics, but not so far off that enough energy can't be delivered to a narrow spot to boil rock. The paper is just an analysis of the size and number of spacecraft required, assuming a solar power system driving a solid state laser. (Why not just focus the sun? The mirror has to be too close to the asteroid to get a tight enough beam, and then it gets hit by the rock being boiled off.)
As a weapon system, it's not very useful. It's too expensive and vulnerable for an anti-satellite weapon. The beam might be able to deliver enough energy through the atmosphere to set fires, but the spacecraft would have to be put in low earth orbit to do it.
Robots are starting to work in unstructured situations. I was there at the moment when this was recognized - the 2005 DARPA Grand Challenge at the California Motor Speedway in Fontana, CA. That's when everything changed.
The 2004 Grand Challenge, remember, was a pathetic joke. No vehicle got further than 7 miles, and that was CMU's. The CMU approach at the time wasn't even really autonomous. Entrants got the route on a CD an hour or so before the start. CMU had imagery of the whole area and tried to plan obstacle avoidance manually just before the start, using a huge team of people in a semitrailer full of workstations. Didn't work; the DoD people in charge had moved some obstacles during the night. And that was the best result. One vehicle came out of the gate, turned hard, and ran back into the starting gate. One flipped over. The big Oskosh entry demolished a SUV parked as an obstacle to be avoided. The whole thing was embarrassing.
DARPA was very displeased with the performance by the universities that had long been receiving DARPA funding for robotics. It was quietly made clear to some major CS departments that their performance had to improve or funding would be cut off. That's why entire CS departments were suddenly devoted to the DARPA Grand Challenge in 2005.
In 2005, things were completely different. Everybody who got that far had already been through an elimination, and every vehicle at the 2005 challenge was better than any of the 2004 entries. There was considerable press coverage, and at first, the press treated it as a joke. But suddenly there were over 20 vehicles running around autonomously, and they weren't crashing into stuff. When multiple vehicles finished the course, it was viewed as a triumph.
Finally, the state of the art had reached the point that money and determination would get problems solved. That wasn't true in the 1980s. NASA threw over $100 million at the Flight Telerobotic Servicer project, and got nothing that worked.
Now check out the DARPA Humanoid Challenge. (There's much dreck about this on blogs and in the popular press. Read the DARPA announcement instead.) They have an approach that's likely to work, and demand simulated demos (in their simulator) in 9 months, with demos on real hardware in 18 months. I personally think they'll get something able to do most of the mobility tasks and some of the manipulation tasks in that time. Useful humanoid robots will be a lot closer in two years.
Price will still be a problem. But not an unsolveable one. These things could be brought down to the price of an SUV, if not lower, through production economies alone. The parts count is probably lower than that for an SUV.
Where the fuck is my picks-up-my-clothes-washes-them-and-dries-them-and-folds-them-and-puts-them-away robot?
Here it is, sorting and folding socks.
Yes, it's slow. The code is in Python and it's still experimental. I've heard from the Willow Garage people that they've speeded up towel folding 50x since the 2010 demo. Once you can do it at all, it can be done faster and cheaper.
Back in 1947, the present rules of presidential succession were set up. The present line of succession has 18 people. That ought to be enough.
When this is a real worry, a few of those people should be in a bunker. During presidential inaugurations and presidential speeches to Congress, that's actually done; at least one person in the line of succession is in a safe place far away. It's usually someone far down the list, but in 2001 Dick Cheney (VP) was sent to the "undisclosed location", and in 2003, Ashcroft (AG) got bunker duty. In 2005, 2006, and 2007, the president pro tem of the Senate went to the bunker. In 2009, Holder (the AG) got the duty. Since then, after most of a decade with no significant terrorist attacks, it's back to the low-rankers.
In terms of actual threat, nobody in the US presidential line of succession has ever been assassinated.
This is a problem for which a solution was implemented long ago, back when a major war looked like a likely possibility.
Remember Flash animation cartoons?
"The God and Devil Show", from Mondo Media.
This might be useful for detecting people carrying concealed guns. It's known that when people wearing a big dense object step up or down (a curb is sufficient) there's motion that can be noticed. Some cops are trained for this. Now it can be automated.
I've heard this at Dolby's screening room in SF. It looks like a modest auditorium. It's really a money-is-no-object demo facility. Before a talk on another subject, the Dolby guys couldn't resist showing off. They had a video game with many directional sound outputs hooked into the room's systems, and you could hear the players moving around in the space, behind and above the audience when appropriate. You really can hear somebody sneaking up on you in-game from a platform above you.
It's an experience to hear many-channel sound in a facility like that, but few (if any) commercial theaters are that good acoustically. Unless the room acoustics are very, very good, all those channels won't help much.
Computers don't have independent agency.
Neither do employees.
Here's our current list of major domains being exploited by active phishing scams. Notice who's at the top of the list. Google.
We've been generating that list for years. It's based on PhishTank data, updated every 3 hours, and uses Open Directory to decide if a site is "major". 46 domains are on the list today. 9 have been on the list since 2011 or earlier. One has been on the list since 2010 - Google. Google is the last free hosting service unable to clean up their phishing problem. MSN, Yahoo, and various free hosting services have been successful at aggressively cleaning up phishing problems, and haven't been on this list, other than briefly, for years.
Here's the oldest phishing attack hosted by Google, up since 2010: "Free Habbo Coins. Email your username and password to..."
For years, Google didn't realize that Google Spreadsheets could be used to host phishing sites. They finally caught on, and there's now a "report abuse" button on spreadsheets. Most, but not all, of the spreadsheet-hosted phishing sites have been taken down.
If anybody from Google is reading this, go over to your abuse department and apply a clue stick. It should embarrass someone that Google is the most clueless free hosting provider in the world about phishing.
As I point out occasionally, 3D printing isn't all that great. TechShop has both a little MakerBot and a commercial stereolithography machine. I've never seen anything come out of the MakerBot but simple decorative plastic objects, and mostly it sits idle. The commercial machine makes good quality objects, but the process is so slow that the expense is high.
There's this mindset that additive machining is some magic process comparable to a SF matter duplicator. It's not. It's far slower and much more expensive than mass production. Too many people today have never been on a factory floor. If you want to see something that looks like matter duplication, check out a high speed injection molding line or a small parts stamping plant.
You mean studios will finally be able to pan at a reasonable speed without it looking jittery and fucking terrible?
24 fps is terrible and you should feel bad for propping up a dying standard.
Cameron ("Titanic", "Avatar") is pushing hard for 48FPS. His work has pans over exquisitely detailed backgrounds. That looks awful at 24FPS unless the pans are kept very slow. Or worse, some blur is applied, which is a common solution in action movies and is why many action backgrounds are out of focus. (Well, sometimes the backgrounds are out of focus to hide flaws background art and set construction.)
Action movies are going to look much crisper at higher frame rates.
The concept of "privacy" as a paid, centralized service leaves something to be desired.
It's a terrible idea for a company to do this. A company can block access to sites via HTTPS on their own network if they wish. Breaking the encryption and snooping, though, creates liability for the company. There is an expectation of privacy associated with an encrypted connection. If an employee's legitimate online banking activity (for example, making sure their pay was deposited) results in a security breach, the employer would be liable. The employer may be guilty of an HIPAA violation. If they snoop on an employee's communication with their union or a Government agency, they may violate other laws.
There's been discussion on the Mozilla security list over whether Firefox should raise alarms if it detects a wildcard cert. The consensus seems to be "yes, it should". Mozilla policy is moving towards kicking CA out of the root list if they issue wildcard certs, and adding technical measures to prevent them from working.
Although it's been proposed many times, nobody has ever put up a small, reusable manned spacecraft. The USAF had the DynaSoar program in the 1960s, but that was cancelled. Virgin Galactic is making noises about a small orbital spaceplane. Nothing like that has ever flown, but there's no fundamental obstacle.
The near future of earth orbit space may be Space-X's Falcon Heavy for freight, something from Virgin Galactic for humans, and robotic vehicles for military tasks.
Except that any such taps are instantly detectable, at which point communication stops. Thus, at most 1 bit of information leaks out to an eavesdropper.
That's a property of quantum communication, where observation affects the result. Not this.
It's going to be tough. There's too much in that book to cram into a movie, and most of it contributes to the main plot. What to cut?
Probably most of the virtual reality. VR was more promising in 1992 than it is now. It's been way overdone in movies. Show Hiro in gloves and goggles gear in his storage space, and others briefly in similar gear when appropriate, but spend little screen time on VR.
Use Juanita Marquez, Hiro's ex-girlfriend and linguist/mythologist , as the designated explainer for the psycho-religious stuff. Somebody has to do that job.
If they're lucky, they might be able to get Chloe Grace Moretz ("Hit Girl") as Y.T. That's the toughest casting decision. Any of the usual big hunks can play Raven. A number of older actors could play Uncle Enzo. Ng is a CG character. No idea who should play Hiro.
As someone pointed out, this was on Slashdot 7 years ago. Here's the referenced paper.
The idea is simple. At both ends of the wire, random data modulated with content is being emitted. At any point on the wire, you see the sum of two random sources. But each end knows their own random data, and can subtract it out.
To break the system, you need two taps on the wire, some distance apart. Now you get to see the sums of the signals from each end, but with different time shifts between them due to propagation delay. With that data, you can separate out what's coming from each end. This allows recovering the original signals.
"No new encryption system is worth looking at unless it comes from someone who has already broken a very hard one." - Friedman.
The girl writes a good food blog, with good pictures. She has a bright future, and will probably do much better in life than the people annoying her.
You know, we had the same argument with RISC versus CISC architecture. And we know who lost that one. Badly.
The real question there is, do you want to go superscalar? Sequential RISC CPUs are simpler than sequential CISC CPUs, but once you have pipelines and multiple execution units, there's so much added complexity and transistor count that the difference disappears. If you're willing to have a slow RISC CPU, the transistor count can be quite low. Down at the bottom, where there's on-chip memory and no cache, as with Atmel ATMega CPUs, the transistor count is really low.
ARM started down there, but it's been built up to a serious level of computing power, with multiple cores, layers of caches, MMUs, and all the stuff of a desktop CPU. Once you have all that stuff, the instruction decoder part is a tiny fraction of the transistor count.
(Of course, x86 superscalar machines require the retirement unit from hell to manage all the hard cases, like self-modifying code. I still remember going to the talk where the Intel guy in charge of the Pentium Pro program explained how they did that. 3000 engineers on the design team at peak. One of the people at that talk was the guy who designed the Intel 8051 mostly by himself.)
âoeAt an academic level, it's a very different background,â says Bobby Schnabel, Dean of the School of Informatics at Indiana University and chair of the ACMâ(TM)s Education Policy Committee. "The calculus and differential equations that underlie engineering are not what underlies computer science. It's really discrete mathematics."
That was true a few decades ago. Today, though, all that discrete math isn't as useful. Today, you need calculus and Bayesian statistics for machine learning. You need differential equations and computational geometry for game development and robotics. Number theory, mathematical logic, graph theory, and automata theory just aren't that important any more. Most of what's needed from those fields is now embodied in well-known algorithms.
I got all the classic discrete math training, but over the years, I've had to use far more number-crunching math.
Get your company's lawyers involved in negotiating with "cloud" providers. Make sure all the things that the "cloud" provider can screw up result in substantial financial penalties. Lawyers are paid to prepare for contingencies like that. If a "cloud" provider won't agree to enforceable service agreements, price out business interruption insurance coverage for the cloud provider's failure. Now you have a backup plan and costing for it.
Von Neumann was much more influential than Turing. Not only did von Neumann do brilliant work in multiple areas of mathematics, he invented modern computer architecture. Babbage's design was more like a Jacquard loom card reader coupled to a calculator. Turing's theoretical machine had to roll a long tape back and forth, and the cryptographic machines were essentially hard-wired or plugboard-programmed. Those machines are closer in concept to Hollerith/IBM tabulators of the 1920s to 1950s.
Von Neumann got computer architecture right. He saw that the right answer was RAM, with programs and data in the same memory: The device requires a considerable memory. While it appeared that various parts of this memory have to perform functions which differ somewhat in their nature and considerably in their purpose, it is nevertheless tempting to treat the entire memory as one organ, and to have its parts even as interchangeable as possible for the various functions enumerated above."
He also figured out that 1) everything inside the machine should be binary, not decimal, 2) memory sizes should be a power of two, 3) about 2^18 bits of RAM were needed to get any useful work done, 4) delay-line memory would work in the short term, but "iconoscope" memory (see Williams tube), which is random access, would be better, and 5) what a reasonable instruction set should look like.
They must have had a slow day at TED and needed a talking head.