I recently converted my Malibu garden lights to LEDs. The "solar" Malibu lights don't work well at all in a heavily shaded garden, so I have to use the plug-in type. The old 11-watt wedge-based light bulbs consumed a total of almost 500 watts (there were about 40 of them). Times 10 hours a night, that took 5 KWH per day. The new LED wedge-base "bulbs" consume a total of 38 watts. That is better than a 90 percent reduction.
That project is not as "cool" as implementing solar panels, but it makes more economic sense.
The total for the 40 bulbs was just about $150. So the LED bulbs will pay for itself in less than a year. Adding solar panels and batteries could reduce the operating cost further. But eliminating the last 38 watts of usage won't pay for very much capital cost. So, for the time being, the payback just isn't there.
The main point is to consider the operating cost vs the capital cost for each application. Sometimes you can get bigger savings for less capital by doing something other than the obvious, trendy, project.
I have a friend who owns a small restaurant, selling smoothies and sandwiches. He has internet access from the back office, and uses it to communicate with vendors.
He doubled his breakfast and lunch business over the last few months by putting up a wireless router and giving away wifi access. The sign says "with any purchase" but there is no easy way to implement that, so he just leaves it unsecured. Most people buy something anyway.
It costs him almost nothing, and helps to sell food by making the location more welcoming to his customers. It won't take very long for other small food and beverage businesses to catch on.
It's kind of like "air conditioned" businesses used to be. Fifty years ago, air conditioning was unusual. But customers liked it, so the businesses that had it got the customers. Now, every business has it. The only real difference is that wifi is a lot cheaper to provide.
to tell the difference between subterfuge and sheer incompetence.
Is very easy in this case. The white house has billions of budget authority for this stuff. They can hire just about any consultant they want. And they have thousands of very talented people on staff, including a few hundred good support people, and a few hundred good data-center people.
There is no incompetence.
This isn't hard evidence of an "error". An "error" is by definition unintentional. When simple counting is done incorrectly by a computer, for numbers that are small enough to fit in 32 bits, it isn't an "error". The computer is just doing what it was programmed to do. I cannot imagine a programmer being so incompetent as to program an increment instruction incorrectly. That cannot possibly be an "error." It has to be intentional.
The evidence reported by this article is clear, unambiguous, damning evidence of election rigging. There is no other way to interpret it.
Take one reasonable small car. Add a thousand pounds of "safety" equipment. Then add another 800 pounds of "emissions" equipment. While you're at it, throw in an air conditioner, power steering, power seats, a sun roof, and a few other details. Then beef up the structure and suspension enough to carry all that extra weight. Then evaluate the acceleration and braking performance, and beef up the motor and the brakes to accelerate and brake all that extra weight. The bigger motor needs more fuel to go the same distance, so there's more weight in the fuel tank, too. Then your marketing people decide they want 0-to-60 in 6.3 seconds, so you offer an optional larger motor, which requires a stronger transmission, even more structure and suspension, and yet more fuel. Then the optional motor is made standard, along with the automatic transmission. Add a hundred pounds here and a hundred pounds there, and pretty soon you're talking about some serious weight.
The engineers do their best to reduce weight wherever they can. But, when it's all said and done, an average small car today weighs twice as much as similar products from forty or fifty years ago. All those "features" add mass and inertia which has to be accelerated and decelerated.
The problem is that elected officials, election supervisors, elections workers, and prosecutors have an incentive not to rock the boat. They all benefit from the rigged election systems.
For every particular procedure or mechanism that is proposed to ensure honest elections, there is a way to circumvent the protection. Elaborate procedures or processes are not the answer. We need to create some incentive to counterbalance the existing incentive to rig elections. And, we need incentives to detect and prosecute election rigging.
This can be corrected by a simple new law. The major elements of that new law need to be something like:
1. Election-rigging is a felony, punishable by extremely long prison term, without any privileges or possibility of parole;
2. Huge reward for evidence or testimony of election-rigging leading to conviction. If multiple witnesses come forward, the reward is per-person, not shared.
3. Big bonus to any prosecutor who successfully prosecutes any case of election-rigging to conviction.
4. Rewards and bonuses are to be paid from assets seized from the convicted persons. If they had insufficient assets, the state treasury will make up the difference.
If they just dropped the SYN packets, it would probably be just about as effective at defending against SYN-flooding DOS attacks. But, it would be more inconvenient for the legitimate users whose calls also get dropped in the process. By giving the caller an error indication, the network allows those users to retry sooner, and help;s to minimize the collateral damage to innocent bystanders. Imagine a condition where you have two hosts on your ethernet behind a NAT firewall. The kids machine is running virus that does a SYN attack. The parent on the other host has real work to do, and would be seriously inconvenienced by having his/her calls dropped. (Of course, the parent might have a responsibility to supervise the kid, but that is another issue altogether.)
I can't speak to Comcast's business practices, because I haven't done business with them for many years. However, I do expect that their technical people are just doing the best they can within the limits of their existing hardware and network architecture. You can't roll trucks to replace equipment very often, so you have to do the best you can with whatever hardware is in the field. That is never going to be the latest stuff.
We synthetically generated TCP SYN packets at a rate of 100 SYN packets per second using the hping utility... The IP Time to Live (TTL) field for these forged TCP RST packets is consistently set to 255
So, when new connection requests are issued at the rate of 100 per second, the first router is resetting some of those requests.
The application is issuing new connection requests at a prodigious rate. The router determines that this is beyond the capacity for the router, or perhaps beyond some limit imposed on that router by the internal network. Or, perhaps, it is beyond a rate parameter that is used to detect DOS attacks.
When such a limit is exceeded, there are a few reasonable responses for the router to choose from: It can drop random packets; It can drop random SYN packets; it can drop packets from the attacking host; or it can NAK/RST some of those SYN packets. All of those are legitimate router responses. The reset packets are not "forged". They are legitimate responses in the protocol. The primitive operation is called a "provider disconnect indication".
I don't see any problem in the protocol here. And, I don't see any problem in the router behavior. The router is just protecting itself and the network from overload conditions. By selecting to disconnect calls from a host that is using far more resource than other hosts, it is just protecting the other hosts from a DOS attack by that first host.
The title of the summary should be "Local routers defend agaist DOS attack".
The Bushite fascists now control the country. We live in a police state. They can search and seize at will, without restraint. They can take people into custody without warrant or charge, and without any limit as to duration or methods of interrogation. The president even claims the power, now, to declare martial law.
I expect that we will not actually have an election in November. They might hold a mock election, to give the appearance of an election, but without actually counting any votes. But I wouldn't even count on that. I think he will make up an excuse to declare martial law and cancel the elections before then.
The only possible saving grace is that our military people are sworn to preserve and protect the constitution, not the president. And the great majority of them are good, honest people. If the Bushites go too far, they might do something about it. Then we will have a military dictatorship instead of a civilian police state. But, just maybe, they will then organize new elections to reconstitute the republic.
The Republicans who control those two states keep rigging the elections. They have to do it in different ways each year, because the system and procedures are updated after each time they get caught, to prevent them doing it the same say again. After all, that's the only way they can stay in power. This year they changed the primary date, knowing that the date would violate Democratic party rules and cause those delegates not to be seated, and then spun the news to try to blame the Democrats.
This fall, look for problems with the new registration-verification machines, and with the optical scanning machines. The statewide registration databases are new, as are the registration-verification machines, so that area is ripe for fraud. In Florida, the new election law says that the "recount" procedure is to reprint the summary reports from the optical scanning machines, and that paper ballots that have been scanned may NOT be manually inspected or audited.
The right answer to most such questions is, "It depends."
What sort of tools will be useful in your future career?
Or, what sort of knowledge will be interesting to learn?
If you are concerned with serious engineering issues, such as safety and correctness, you might want to learn something about "formal methods". Languages to look at include Z and B. And, of course, there's a little field called "computational logic."
If you are concerned with commercial byte-pushing, you obviously need to be conversant in SQL and an assortment of scripting languages. And, of course, there's a little field called "accounting" that might be useful.
I don't think there is any one right answer. The choice of intellectual tools that you will need really depends on the choice of what kinds of work you want to do.
Many managers sincerely believe that their job is to get the work done for the minimum cost. Most of them don't understand that some people really do produce more or better results. And they don't understand the cost structure of software. So they hire the cheapest people, without regard to qualification, ability, or talent, and whine about the "shortage" of even-cheaper people. That could be seen as pointy-haired stupidity. Or, it could be a smart move by that manager.
A battalion of fresh-out programmers who have no intention of mastering the subject can crank out lots of code in a hurry. Most of it won't work, and so will require lots of testing and rewriting. All that testing and rewriting increases the number of man-hours in the project, which can be used to justify more warm bodies, and more budget, which can in turn be used to justify a higher salary/bonus for the manager. The constant hiring and turnover assures that no one ever develops the expertise to actually get the system working. And the unrealistic ceiling on compensation assures that highly competent and experienced people cannot be hired. So the project never ends, but the manager gets promoted.
A squad of seasoned professional developers, with higher degrees and/or decades of experience, can develop the same system in a fraction of the time. Most of their stuff will work with very little testing and debugging required. So their manager cannot easily justify increasing the number of people, and his successful project looks like a smaller project. After all, it only took ten people time one year, so it can't be a huge thing, can it? That manager then has to come up with other reasons, besides the number employees, to justify his salary/bonus increase. And, he has to find another project for next year.
So the cheapest-people (and lots of them) strategy doesn't work for the company, but it works really well for the manager. At least in the short run, which is all they seem to care about any more.
We used to think that about the earth, too. I would be very surprised if it is actually true.
I expect we will eventually realize that the universe is hyper-spherical, and that time is a vector perpendicular to the surface of that hypersphere. The projection of one time-vector on another time-vector gives the appearance of time dilation, which has been interpreted as expansion. And, the radius of that hypersphere is the zero on the red-shift graph. But that is just a guess.
In the mean time, we can have fun with the big-bangers by asking about the other 95.38% of the mass/energy/whatever that they presently can't account for. The fact that so much is left unaccounted and unexplained tells me the flat-universe theory is, at best, not complete. Any true description of the universe probably should account for just about 100% of the universe.
Does it allow people to watch TV that they didn't pay for? Yes. The TV is offered for free. People who accept the offer can watch it for free.
Does it prevent Verizon and MobiTV from receiving revenue that they should from the streams? No. Verizon and MobiTV could just withdraw the free offer, and implement a different access-controlled method for the same video.
Is it wrong? No. Someone offers free goods. You accept the offer. You have not done anything wrong.
Does MobiTV and Verizon have the right to send a cease and desist letter? Yes. Anyone can write a letter. It means nothing.
Were MobiTV and Verizon stupid to offer this data online for free? Maybe -- It could have been done intentionally. Lots of people put video online, for free.
Were MobiTV and Verizon stupid to continue offering this data online for free, after they decided that they didn't want to? Yes.
It isn't just the cost. You also have to consider the benefits and risks.
The decision to replace an aging software system is a business decision, driven by the numbers. When the (benefit - cost) of the new system exceed the (benefit - cost) of continuing the old system, then it's time to buy the new system. If both are negative, it's time to just shut it down. Part of the cost of the old system is the people you need to keep it running. If it requires unusual skills or high levels of expertise, those might be expensive people. If it requires lots of labor to maintain and operate, then the total payroll might be large even if the individuals are not expensive. Part of the cost of any new system also includes the people you need, both to implement it and to keep it running afterward. Business opportunity and business risk also must be estimated for each alternative. The list of costs and benefits might be extensive, and some of those numbers might be guesses, but ultimately the numbers will make the decision.
The manager who cries about not being able to find good people really just doesn't want to pay for them. There are plenty of highly skilled people in the world. But, the low bidder doesn't generally win the auction.
Exactly how long does he think it will take before someone, somewhere, installs a router between the old Internet and the New Internet?
I would guess it might take slightly longer than a nanosecond. But not by much. Most of the first New Internet routers will be installed in schools, to protect the children. I'm pretty sure that there is at least one evil grad student in one of our schools who is fully capable of configuring a router.
On second thought, the New Internet would probably be connected to the Old Internet before it even boots up for the first time.
This is not because of issues with 5NF but rather because of DBA's who don't understand the math or don't understand the data.
It is more often due to practical data that doesn't quite fit into the nice, original schema. This is especially common in production systems where the data is accumulated over years, and the application programmers are sometimes required to stuff new data into the existing database in whatever way it will fit.
By the way, the quote is usually attributed to Lawrence Peter Berra,
"In theory, theory and practice are the same. In practice, they are not."
But we can't be sure, because Berra also said, "I never said most of the things I said."
Something on that satellite is very important to keep secret, and could survive an uncontrolled reentry.
Shooting down the satellite allows us to keep that thing secret, either by controlling where the secret lands, or by making sure it is destroyed on reentry.
What could the secret be? The optics or sensors might be better than previously disclosed. There might be a new kind of sensor or antennae. There might be some signal processing equipment, encryption equipment, or maybe a transmitter designed to jam their stuff. But most of that sort of equipment would be designed so that it would certainly be destroyed by reentry, just on the off chance that it might come down in a foreign nation. We routinely don't worry about leftover fuel or oxidizer, or a few tons of metal. So it must be something else. It must be something that is intentionally made to survive reentry. That kind of limits the options. We used to use film canisters to reenter imagery, with secret kinds of film. But digital cameras and the desire for rapid imaging made that obsolete many years ago. What are the other kinds of things that would be on a military satellite, and made to survive reentry? Perhaps warheads?
In operating system theory, it is well known that a scheduling algorithm called "Shortest Job First" yields the least total waiting time. The SJF algorithm is usually implemented by giving a "new" job high priority, and then reducing the priority gradually as the job accumulates resource usage. The algorithm was developed in the 1960's to allow time-sharing operating systems to provide rapid keystroke response, while continuing to process large batch jobs in the background.
For communication systems, the same principle applies. The only difference is that the network is sharing a different resource (circuit bandwidth), instead of cpu time. The "new" connection gets high priority, and then that priority is reduced as the number of bytes/packets transferred over that connection increases. This allows rapid response for interactive applications, like browsing or editing, while also allowing the network to process large data transfers in the background. To apply it to datagram traffic, the switch just keeps a priority for each source/destination address-pair in cache, and any pair that is not in the cache is regarded as "new".
This has been pretty much standard practice in packet communication switching for a very long time. There is no surprise here, at least not to those of us who have not been doing communications network programming for a few decades.
I can imagine a whole new category of stage lights.
Imagine panels of pixels, with each pixel being a small, independent spotlight, steerable by software.
Cover the front of the balcony, and the upper lighting balcony, with these panels.
Then the software can steer the required amount of light (number of pixels) of the required color(s),
to any point on the stage, for each independent scene in the show. Floods could be simulated by
hundreds of parallel spots, or some of the individual sources could be wider than others.
Getting hard shadows or lines might be difficult, but I can see lots of advantages.
There would not be a need to point hundreds of individual fixtures during setup on a new stage.
Just mount the panels on the trusses and use the software to align the system to the stage dimensions.
And, instead of a few hundred sources for a few hundred scenes,
you could have millions of sources for thousands of scenes.
Finally, with millions of sources, it isn't a big deal if one of them burns out during the first act.
During rehearsal, the director says "I want a spot on that". In a few seconds, the lighting guy moves a mouse and the spot of light appears on the stage. The director says, "make it more amber", and again it happens in a few seconds.
You might still need some special effect units, but for general scene illumination it could be very nice.
The real problem with the old ebay feedback system is that the buyer cannot really tell the difference between the good sellers and the bad sellers.
The buyer and seller have very different responsibilities. The buyers only responsibilities are to pay for the item and provide a valid shipping address. The seller must represent the item truthfully, provision it, pack it so that it will not get damaged, and then ship promptly after receiving payment.
Most sellers, in my experience, do not leave feedback until after the buyer leaves feedback. Retaliation for non-positive feedback is completely routine. The bad sellers use threats of bad feedback to extort good feedback from buyers. If the seller misrepresents the item, packs it badly, or waits several weeks to ship it, then the seller has earned a poor mark, and the buyer should be able to leave negative feedback without retaliation.
The seller has no risk if he/she just waits for payment before shipping. All of the risk is assumed by the buyer. The buyer is out the money and must wait for the goods to arrive. Hence, the purpose of the reputation system must be to help the buyers distinguish the good sellers from the bad sellers. The old system is just not serving that purpose.
Quite right. The requirement for voting systems is not "absence of evidence of wrongdoing."
What is actually required is "evidence of absence of wrongdoing." And, that evidence of absence must be fully convincing to the great majority of average people who have absolutely no technical understanding beyond basic arithmetic. If "evidence of absence of wrongdoing" is not obvious, we should reasonably infer that wrongdoing (as, election rigging) has taken place.
The only way to produce the required evidence, and to make it simple enough so that everyone can understand it, is: 1. Use simple paper ballots, not invisible bits; 2. Allow everyone see that no one has any opportunity to alter the ballots; and 3. Allow everyone to see that the ballots are counted correctly.
Bloody revolutions happen when enough people feel that the government is wrong that they act in concert to change the government. There will always be some people who feel that the government is wrong. Hence, the only way to prevent revolutions is to convince those people that there are not enough of them to actually force the change.
Slashdot Mirror
I recently converted my Malibu garden lights to LEDs. The "solar" Malibu lights don't work well at all in a heavily shaded garden, so I have to use the plug-in type. The old 11-watt wedge-based light bulbs consumed a total of almost 500 watts (there were about 40 of them). Times 10 hours a night, that took 5 KWH per day. The new LED wedge-base "bulbs" consume a total of 38 watts. That is better than a 90 percent reduction.
That project is not as "cool" as implementing solar panels, but it makes more economic sense. The total for the 40 bulbs was just about $150. So the LED bulbs will pay for itself in less than a year. Adding solar panels and batteries could reduce the operating cost further. But eliminating the last 38 watts of usage won't pay for very much capital cost. So, for the time being, the payback just isn't there.
The main point is to consider the operating cost vs the capital cost for each application. Sometimes you can get bigger savings for less capital by doing something other than the obvious, trendy, project.
I have a friend who owns a small restaurant, selling smoothies and sandwiches. He has internet access from the back office, and uses it to communicate with vendors.
He doubled his breakfast and lunch business over the last few months by putting up a wireless router and giving away wifi access. The sign says "with any purchase" but there is no easy way to implement that, so he just leaves it unsecured. Most people buy something anyway.
It costs him almost nothing, and helps to sell food by making the location more welcoming to his customers. It won't take very long for other small food and beverage businesses to catch on.
It's kind of like "air conditioned" businesses used to be. Fifty years ago, air conditioning was unusual. But customers liked it, so the businesses that had it got the customers. Now, every business has it. The only real difference is that wifi is a lot cheaper to provide.
to tell the difference between subterfuge and sheer incompetence.
Is very easy in this case. The white house has billions of budget authority for this stuff. They can hire just about any consultant they want. And they have thousands of very talented people on staff, including a few hundred good support people, and a few hundred good data-center people. There is no incompetence.
They did exactly what the boss wanted them to do.
This isn't hard evidence of an "error". An "error" is by definition unintentional. When simple counting is done incorrectly by a computer, for numbers that are small enough to fit in 32 bits, it isn't an "error". The computer is just doing what it was programmed to do. I cannot imagine a programmer being so incompetent as to program an increment instruction incorrectly. That cannot possibly be an "error." It has to be intentional.
The evidence reported by this article is clear, unambiguous, damning evidence of election rigging. There is no other way to interpret it.
Take one reasonable small car. Add a thousand pounds of "safety" equipment. Then add another 800 pounds of "emissions" equipment. While you're at it, throw in an air conditioner, power steering, power seats, a sun roof, and a few other details. Then beef up the structure and suspension enough to carry all that extra weight. Then evaluate the acceleration and braking performance, and beef up the motor and the brakes to accelerate and brake all that extra weight. The bigger motor needs more fuel to go the same distance, so there's more weight in the fuel tank, too. Then your marketing people decide they want 0-to-60 in 6.3 seconds, so you offer an optional larger motor, which requires a stronger transmission, even more structure and suspension, and yet more fuel. Then the optional motor is made standard, along with the automatic transmission. Add a hundred pounds here and a hundred pounds there, and pretty soon you're talking about some serious weight.
The engineers do their best to reduce weight wherever they can. But, when it's all said and done, an average small car today weighs twice as much as similar products from forty or fifty years ago. All those "features" add mass and inertia which has to be accelerated and decelerated.
You may now resume saying "Well, this isn't rocket science"
Except that this is rocket science.
The problem is that elected officials, election supervisors, elections workers, and prosecutors have an incentive not to rock the boat. They all benefit from the rigged election systems.
For every particular procedure or mechanism that is proposed to ensure honest elections, there is a way to circumvent the protection. Elaborate procedures or processes are not the answer. We need to create some incentive to counterbalance the existing incentive to rig elections. And, we need incentives to detect and prosecute election rigging.
This can be corrected by a simple new law. The major elements of that new law need to be something like:
1. Election-rigging is a felony, punishable by extremely long prison term, without any privileges or possibility of parole;
2. Huge reward for evidence or testimony of election-rigging leading to conviction. If multiple witnesses come forward, the reward is per-person, not shared.
3. Big bonus to any prosecutor who successfully prosecutes any case of election-rigging to conviction.
4. Rewards and bonuses are to be paid from assets seized from the convicted persons. If they had insufficient assets, the state treasury will make up the difference.
Thank you, eimsand.
If they just dropped the SYN packets, it would probably be just about as effective at defending against SYN-flooding DOS attacks. But, it would be more inconvenient for the legitimate users whose calls also get dropped in the process. By giving the caller an error indication, the network allows those users to retry sooner, and help;s to minimize the collateral damage to innocent bystanders. Imagine a condition where you have two hosts on your ethernet behind a NAT firewall. The kids machine is running virus that does a SYN attack. The parent on the other host has real work to do, and would be seriously inconvenienced by having his/her calls dropped. (Of course, the parent might have a responsibility to supervise the kid, but that is another issue altogether.)
I can't speak to Comcast's business practices, because I haven't done business with them for many years. However, I do expect that their technical people are just doing the best they can within the limits of their existing hardware and network architecture. You can't roll trucks to replace equipment very often, so you have to do the best you can with whatever hardware is in the field. That is never going to be the latest stuff.
Read RFC 4987, "TCP SYN Flooding Attacks and Common Mitigation"
We synthetically generated TCP SYN packets at a rate of 100 SYN packets per second using the hping utility ... The IP Time to Live (TTL) field for these forged TCP RST packets is consistently set to 255
So, when new connection requests are issued at the rate of 100 per second, the first router is resetting some of those requests.
The application is issuing new connection requests at a prodigious rate. The router determines that this is beyond the capacity for the router, or perhaps beyond some limit imposed on that router by the internal network. Or, perhaps, it is beyond a rate parameter that is used to detect DOS attacks.
When such a limit is exceeded, there are a few reasonable responses for the router to choose from: It can drop random packets; It can drop random SYN packets; it can drop packets from the attacking host; or it can NAK/RST some of those SYN packets. All of those are legitimate router responses. The reset packets are not "forged". They are legitimate responses in the protocol. The primitive operation is called a "provider disconnect indication".
I don't see any problem in the protocol here. And, I don't see any problem in the router behavior. The router is just protecting itself and the network from overload conditions. By selecting to disconnect calls from a host that is using far more resource than other hosts, it is just protecting the other hosts from a DOS attack by that first host.
The title of the summary should be "Local routers defend agaist DOS attack".
The coup is already done. The people lost.
The Bushite fascists now control the country. We live in a police state. They can search and seize at will, without restraint. They can take people into custody without warrant or charge, and without any limit as to duration or methods of interrogation. The president even claims the power, now, to declare martial law.
I expect that we will not actually have an election in November. They might hold a mock election, to give the appearance of an election, but without actually counting any votes. But I wouldn't even count on that. I think he will make up an excuse to declare martial law and cancel the elections before then.
The only possible saving grace is that our military people are sworn to preserve and protect the constitution, not the president. And the great majority of them are good, honest people. If the Bushites go too far, they might do something about it. Then we will have a military dictatorship instead of a civilian police state. But, just maybe, they will then organize new elections to reconstitute the republic.
The Republicans who control those two states keep rigging the elections. They have to do it in different ways each year, because the system and procedures are updated after each time they get caught, to prevent them doing it the same say again. After all, that's the only way they can stay in power. This year they changed the primary date, knowing that the date would violate Democratic party rules and cause those delegates not to be seated, and then spun the news to try to blame the Democrats.
This fall, look for problems with the new registration-verification machines, and with the optical scanning machines. The statewide registration databases are new, as are the registration-verification machines, so that area is ripe for fraud. In Florida, the new election law says that the "recount" procedure is to reprint the summary reports from the optical scanning machines, and that paper ballots that have been scanned may NOT be manually inspected or audited.
The right answer to most such questions is, "It depends."
What sort of tools will be useful in your future career? Or, what sort of knowledge will be interesting to learn?
If you are concerned with serious engineering issues, such as safety and correctness, you might want to learn something about "formal methods". Languages to look at include Z and B. And, of course, there's a little field called "computational logic."
If you are concerned with commercial byte-pushing, you obviously need to be conversant in SQL and an assortment of scripting languages. And, of course, there's a little field called "accounting" that might be useful.
I don't think there is any one right answer. The choice of intellectual tools that you will need really depends on the choice of what kinds of work you want to do.
Many managers sincerely believe that their job is to get the work done for the minimum cost. Most of them don't understand that some people really do produce more or better results. And they don't understand the cost structure of software. So they hire the cheapest people, without regard to qualification, ability, or talent, and whine about the "shortage" of even-cheaper people. That could be seen as pointy-haired stupidity. Or, it could be a smart move by that manager.
A battalion of fresh-out programmers who have no intention of mastering the subject can crank out lots of code in a hurry. Most of it won't work, and so will require lots of testing and rewriting. All that testing and rewriting increases the number of man-hours in the project, which can be used to justify more warm bodies, and more budget, which can in turn be used to justify a higher salary/bonus for the manager. The constant hiring and turnover assures that no one ever develops the expertise to actually get the system working. And the unrealistic ceiling on compensation assures that highly competent and experienced people cannot be hired. So the project never ends, but the manager gets promoted.
A squad of seasoned professional developers, with higher degrees and/or decades of experience, can develop the same system in a fraction of the time. Most of their stuff will work with very little testing and debugging required. So their manager cannot easily justify increasing the number of people, and his successful project looks like a smaller project. After all, it only took ten people time one year, so it can't be a huge thing, can it? That manager then has to come up with other reasons, besides the number employees, to justify his salary/bonus increase. And, he has to find another project for next year.
So the cheapest-people (and lots of them) strategy doesn't work for the company, but it works really well for the manager. At least in the short run, which is all they seem to care about any more.
Spacetime is flat to within a 2% error margin.
We used to think that about the earth, too. I would be very surprised if it is actually true.
I expect we will eventually realize that the universe is hyper-spherical, and that time is a vector perpendicular to the surface of that hypersphere. The projection of one time-vector on another time-vector gives the appearance of time dilation, which has been interpreted as expansion. And, the radius of that hypersphere is the zero on the red-shift graph. But that is just a guess.
In the mean time, we can have fun with the big-bangers by asking about the other 95.38% of the mass/energy/whatever that they presently can't account for. The fact that so much is left unaccounted and unexplained tells me the flat-universe theory is, at best, not complete. Any true description of the universe probably should account for just about 100% of the universe.
Is it a hack? No. It's an url.
Does it allow people to watch TV that they didn't pay for? Yes. The TV is offered for free. People who accept the offer can watch it for free.
Does it prevent Verizon and MobiTV from receiving revenue that they should from the streams? No. Verizon and MobiTV could just withdraw the free offer, and implement a different access-controlled method for the same video.
Is it wrong? No. Someone offers free goods. You accept the offer. You have not done anything wrong.
Does MobiTV and Verizon have the right to send a cease and desist letter? Yes. Anyone can write a letter. It means nothing.
Were MobiTV and Verizon stupid to offer this data online for free? Maybe -- It could have been done intentionally. Lots of people put video online, for free.
Were MobiTV and Verizon stupid to continue offering this data online for free, after they decided that they didn't want to? Yes.
It isn't just the cost. You also have to consider the benefits and risks.
The decision to replace an aging software system is a business decision, driven by the numbers. When the (benefit - cost) of the new system exceed the (benefit - cost) of continuing the old system, then it's time to buy the new system. If both are negative, it's time to just shut it down. Part of the cost of the old system is the people you need to keep it running. If it requires unusual skills or high levels of expertise, those might be expensive people. If it requires lots of labor to maintain and operate, then the total payroll might be large even if the individuals are not expensive. Part of the cost of any new system also includes the people you need, both to implement it and to keep it running afterward. Business opportunity and business risk also must be estimated for each alternative. The list of costs and benefits might be extensive, and some of those numbers might be guesses, but ultimately the numbers will make the decision.
The manager who cries about not being able to find good people really just doesn't want to pay for them. There are plenty of highly skilled people in the world. But, the low bidder doesn't generally win the auction.
Exactly how long does he think it will take before someone, somewhere, installs a router between the old Internet and the New Internet?
I would guess it might take slightly longer than a nanosecond. But not by much. Most of the first New Internet routers will be installed in schools, to protect the children. I'm pretty sure that there is at least one evil grad student in one of our schools who is fully capable of configuring a router.
On second thought, the New Internet would probably be connected to the Old Internet before it even boots up for the first time.
This is not because of issues with 5NF but rather because of DBA's who don't understand the math or don't understand the data.
It is more often due to practical data that doesn't quite fit into the nice, original schema. This is especially common in production systems where the data is accumulated over years, and the application programmers are sometimes required to stuff new data into the existing database in whatever way it will fit.
By the way, the quote is usually attributed to Lawrence Peter Berra, "In theory, theory and practice are the same. In practice, they are not." But we can't be sure, because Berra also said, "I never said most of the things I said."
Something on that satellite is very important to keep secret, and could survive an uncontrolled reentry.
Shooting down the satellite allows us to keep that thing secret, either by controlling where the secret lands, or by making sure it is destroyed on reentry.
What could the secret be? The optics or sensors might be better than previously disclosed. There might be a new kind of sensor or antennae. There might be some signal processing equipment, encryption equipment, or maybe a transmitter designed to jam their stuff. But most of that sort of equipment would be designed so that it would certainly be destroyed by reentry, just on the off chance that it might come down in a foreign nation. We routinely don't worry about leftover fuel or oxidizer, or a few tons of metal. So it must be something else. It must be something that is intentionally made to survive reentry. That kind of limits the options. We used to use film canisters to reenter imagery, with secret kinds of film. But digital cameras and the desire for rapid imaging made that obsolete many years ago. What are the other kinds of things that would be on a military satellite, and made to survive reentry? Perhaps warheads?
In operating system theory, it is well known that a scheduling algorithm called "Shortest Job First" yields the least total waiting time. The SJF algorithm is usually implemented by giving a "new" job high priority, and then reducing the priority gradually as the job accumulates resource usage. The algorithm was developed in the 1960's to allow time-sharing operating systems to provide rapid keystroke response, while continuing to process large batch jobs in the background.
For communication systems, the same principle applies. The only difference is that the network is sharing a different resource (circuit bandwidth), instead of cpu time. The "new" connection gets high priority, and then that priority is reduced as the number of bytes/packets transferred over that connection increases. This allows rapid response for interactive applications, like browsing or editing, while also allowing the network to process large data transfers in the background. To apply it to datagram traffic, the switch just keeps a priority for each source/destination address-pair in cache, and any pair that is not in the cache is regarded as "new".
This has been pretty much standard practice in packet communication switching for a very long time. There is no surprise here, at least not to those of us who have not been doing communications network programming for a few decades.
I can imagine a whole new category of stage lights.
Imagine panels of pixels, with each pixel being a small, independent spotlight, steerable by software. Cover the front of the balcony, and the upper lighting balcony, with these panels. Then the software can steer the required amount of light (number of pixels) of the required color(s), to any point on the stage, for each independent scene in the show. Floods could be simulated by hundreds of parallel spots, or some of the individual sources could be wider than others. Getting hard shadows or lines might be difficult, but I can see lots of advantages.
There would not be a need to point hundreds of individual fixtures during setup on a new stage. Just mount the panels on the trusses and use the software to align the system to the stage dimensions. And, instead of a few hundred sources for a few hundred scenes, you could have millions of sources for thousands of scenes. Finally, with millions of sources, it isn't a big deal if one of them burns out during the first act.
During rehearsal, the director says "I want a spot on that". In a few seconds, the lighting guy moves a mouse and the spot of light appears on the stage. The director says, "make it more amber", and again it happens in a few seconds.
You might still need some special effect units, but for general scene illumination it could be very nice.
The real problem with the old ebay feedback system is that the buyer cannot really tell the difference between the good sellers and the bad sellers.
The buyer and seller have very different responsibilities. The buyers only responsibilities are to pay for the item and provide a valid shipping address. The seller must represent the item truthfully, provision it, pack it so that it will not get damaged, and then ship promptly after receiving payment.
Most sellers, in my experience, do not leave feedback until after the buyer leaves feedback. Retaliation for non-positive feedback is completely routine. The bad sellers use threats of bad feedback to extort good feedback from buyers. If the seller misrepresents the item, packs it badly, or waits several weeks to ship it, then the seller has earned a poor mark, and the buyer should be able to leave negative feedback without retaliation.
The seller has no risk if he/she just waits for payment before shipping. All of the risk is assumed by the buyer. The buyer is out the money and must wait for the goods to arrive. Hence, the purpose of the reputation system must be to help the buyers distinguish the good sellers from the bad sellers. The old system is just not serving that purpose.
Quite right. The requirement for voting systems is not "absence of evidence of wrongdoing." What is actually required is "evidence of absence of wrongdoing." And, that evidence of absence must be fully convincing to the great majority of average people who have absolutely no technical understanding beyond basic arithmetic. If "evidence of absence of wrongdoing" is not obvious, we should reasonably infer that wrongdoing (as, election rigging) has taken place.
The only way to produce the required evidence, and to make it simple enough so that everyone can understand it, is: 1. Use simple paper ballots, not invisible bits; 2. Allow everyone see that no one has any opportunity to alter the ballots; and 3. Allow everyone to see that the ballots are counted correctly.
Bloody revolutions happen when enough people feel that the government is wrong that they act in concert to change the government. There will always be some people who feel that the government is wrong. Hence, the only way to prevent revolutions is to convince those people that there are not enough of them to actually force the change.
Try this sequence:
a) Allow this absurdity and insult to rational intelligence that is a Creation Museum die;
b) Purchase the mastodon skull from the bankruptcy auction to preserve an excellent fossil and put it on display for educational value.