"If a grandma had genitals of a man, she would be a grandpa". While I do think that carrying firearms onto the plane is a bad idea, in a critical situation like this, I'd rather a couple of folks on the plane had firearms or weapons of some sort. At the very least the pilots, and maybe some of the other staff.
But the wisdom of the day was that if you don't interfere, no one gets hurt. Flight 93 was the only one that realized the rules had changed in time to take action.
And they did, even if it was only to drag the hijackers down bodily before they could steer into their target. They could have done no more than that if they'd all been packing Uzis.
Guns have their uses, but one of the last places I'd want to use one is in a tube where dozens of people are all packed into a small, basically linear space surrounded by control and fluid lines. You'd be better off issuing boxcutters to the passengers and crew. Or maybe 3-oz bottles of something nasty.
You don't need guns to take down a bunch of clowns with boxcutters when there's 200 of you and 4 of them.
What was lacking wasn't firearms, it was the realization that the usual way of dealing with hijackers no longer worked. Flight 93 was warned in time to change to a more active response. If they'd known even earlier, they might have even been able to save the pilots.
Nah, it's nothing new. Friend of mine when we were teens had a 70s compact car - a VW Scirocco, or maybe that replaced this one - where you had to pull the engine to change the oil filter. No joke. The guy on the assembly line could install the filter easily because the engine wasn't in the car yet.
I've seen video footage from factories where the engines were installed from the bottom. No auto shop I've ever been in was equipped to do that that I know of. They had to do it the hard way, removing the hood.
I had a dentist whose Jaguar supposedly required the engine to be pulled in order to do a tune-up every 6 months because it was the only way to get at the sparkplugs. Some Cadillacs reportedly had this issue as well.
A Ford I used to own required pulling the dashboard to replace the heater core. Which is bad enough, but the mechanics cracked the dash. since by that time the plastic had become more brittle. The Ford I owned prior to that one could have its heater core replaced by simply undoing 2 wingnuts, removing a cover plate and lifting.
Electronics, however, are a different matter. You often have no more than 2-3 chips doing the heavy work and they're wave-soldered to the board, So lots of luck there. I did have to heave a perfectly good TV set with excellent picture though because apparently a 15-cent low-voltage capacitor managing the vertical height went sour and there were no available schematics.
Stuff like calculators are the worst, though. Bought one in the late 1980's for $5. Fully warranteed. If it failed, send $6.50 for postage and handling to the warranty deport. When the batteries died, a new set ran about $7.
I'm not a social person. Put me in a room full of people and I'll no more spontaneously talk to them than if I was on the moon. I can communicate, but I can be at least as confusing face-to-face as I can via email.
I have a reputation for doing good stuff, but when I work, I work in my own little world and I don't want to be interrupted while I do it. I have a reputation for adapting to user's needs and even suggesting things that the users didn't realize could make their jobs easier, but I need time in between to think things through and to prototype solutions. Not a lot of time - I did Agile years before Agile was "invented", but nevertheless, time. Uninterrupted time.
Shoving me in a cage of twittering birds is going to reduce my productivity, not enhance it.
You can very easily skive off while sitting in an office in an open cubicle next to the boss. Been there, done that.
The boss cannot discern anything more about the productivity/quality of a programmer's work by looming over him/her/it, than he/she/whatever can do when the programmer is on the other side of the planet.
That's because the programmer's quality and productivity have nothing whatsoever to do with what they're physically doing. You could sink the Titanic with stories of people who got their breakthrough insight in the shower, sleeping or even playing "Pong".
Furthernore, no decent manager is going to be scanning every line of code from every employee. There are better things to be doing.
There are ways to determine whether a programmer is doing good work and they don't require physsical oversight. First, by the rate of development of the project and the parts of the project assigned to the programmer and the number of times it needs emergency fixes. That, after all, is what you're REALLY supposed to be paying for, not a chair ornament. It's not overnight, but then, if you're not running a revolving-door shop, things add up over time. You want full chairs, invest in inflatable Bozo dolls. If you want "busy" attach clockwork to make them rock back and forth.
The second way is by listening to ones employees. The programmer's peers are the ones who know the actual code and they'll complain or praise it, whether it's relatively innocuous stuff like the guy I know who never wrote a subroutine if he could cut-and-paste to the person who always delivered overnight but never checked for null pointers.
That's the one thing that pirated works can never offer.
Once a performance is recorded, it's "dead". It will be (barring damage) the same performance the 10,000th time as it was the first.
Granted, there are certain things that just cannot be done live and without studio support. And I wouldn't give them up, but the spontaneity - and even the flubs - of a live performance have magic all their own.
Even in case #1, there is sometimes things that can be done. For example, speculative execution. If you can boil down to a small number of choices as a result of the first operation, then it may make sense to compute both outcomes. Or there may be some other intermediate value that might be needed in only some outcomes. But this requires application-specific knowledge usually to know exactly what is allowable and what the payoff would be. You wouldn't want to create a situation where executing both cases affects a global resource. So you would need a language expressive enough to hint this information to the compiler.
Very good point. A high-level equivalent to the predictive processing at the CPU hardware level!
There are actually 3 kinds of tasks as far as parallelization goes.
1. Totally linear tasks. Each step relies on the output of its predecessor. Thus nothing can begin before its time. Obviously, handing this sort of work off to a parallel system is a waste of time.
2. Simple parallel tasks. This is the case where you can do a lot of trivial operations in parallel. The computer equivalent of a bunch of people using 4-banger calculators. As long as the tasks indivudually take longer to run than they do to schedule and collect, this is an ideal use case for massively parallel array processors.
3. Complex parallel tasks. This is simply case #2, but armed with HP advanced function calculators. The individual processors would be more than just basic gate arrays and thus able to perform complex math functions in parallel. Not as cheap to scale up, but better than waiting out linear time.
Of course, this is for the ideal world. Real-life heavy-computing scenarios may have components for 1, 2, or all 3 of the above.
It's the coldest hour of the end-of-December day and the current outside temperature here is 19C. The interior temperature is about 25 because the heat from yesterday's sun is still lingering in the building. Heat hasn't kicked on in about 2 weeks.
Spend enough time in that sort of environment and your internal thermosensors reset. Working at a company office desk, I spent a great deal of time contemplating setting fire to to whatever I could burn to keep warm. The temperature? 72 degreees Fahrenheit. Good enough if you're an executive wearing a 3-piece suit and wandering around annoying the grunts, but Bob Cratchitt frigid if you're just sitting there pushing paper and bits.
I can tolerate cooler these days, though. Used to be my summer A/C setting was 83F (27C). Now I prefer it a degree or 2 colder.
Or were you trying to suggest that the Indians north of the Rio Grande also had huge cities? If so, then, to put it politely, you were mistaken....
Depends on your definition of 'huge' but 100,000 people seems reasonably large.
Precisely. Besides, IIRC, at the time, Mexico City was significantly more populous than Paris. And had much better sanitation. But I'm not sure if I ever read it in a textbook.
As an American, I wonder... how do you do Christmas in warm weather?
At the beach!
Seriously, I live in a sub-tropical climate in the USA and for years one of the biggest conundrums we had was what to do with Xmas stockings when most houses didn't have fireplaces!
All that does is demonstrate the inadequacy of the English language as it presently exists and applies to the issues of the day.
If I have been granted a right to restrict copying of a work of mine except to those who I wish to receive a copy (whether for money, other considerations or even gratis), that's not theft, it's true. It's infringement and something of value HAS been lost - my remuneration or permission. Stolen, if you prefer to allow for the possibity that something that was destroyed in potentio - payment - is "theft".
But it's a poor representation to call it theft, because people are looking at the copy, not the intangible right as the thing that was stolen. So on the one hand you have the "pirates" "justifying" themselves because it's "not theft" and on the other, you have the fat cats saying "you STOLE our song/movie/whatever". And I just ran out of quote-marks, I think.
So maybe we just need to invent a new word that can be used to make the distinction clearer. I don't agree with a lot of the current copyright and patent laws, but at least let's not muddy the waters with false meanings when discussing them.
That's why I hate it. The "mirror" can throttle, but that's just a drop in the bucket. Only if all the reflection mirrors are throttling can it help, and the larger the mirror, the larger the number of apparently legitimate requests would be, so it's harder to make them good throttles.
Plus, not only the target system is getting blasted. The mirroring systems are getting a pretty heavy load. They can throttle this, but then they risk choking off the legitimate requests, since a legitimate request and a reflection attack request are indistinguishable.
I think the only real cure would be to drop UDP. You can't do a reflection attack from TCP, since you cannot spoof the sender. At least without some major mucking around with the Internet as a whole.
UDP reflection attacks are one case when "economies of scale" work in reverse.
I can throttle such attacks on my DNS servers, since I'm only serving for a few domains and there's not much urgency.
For large ISPs, however, there's going to be a lot more legitimate traffic for a lot more domains and if you bounce a request, you may be turning away the one legitimate customer in the flood.
I hate UDP reflection with a passion. Ordinary attacks are annoying enough, but if I ever got my hands on the people behind this, I could cheerfully do things involving crudely-sharpened objects, live coals and molten plastic. The CIA could take notes and learn new techniques.
There's no reason to shop at a retailer unless you're desperate and need something now
Sometimes it's actually nice to fondle the merchandise. There's only so much you can get out of an online catalog, especially since the pictures are usually pretty poor.
I was beginning to believe the Radio Shack mantra, until I wandered into my neighborhood store the other day. They had an impressive array of sensors and kits for stand-alone, Rasberry Pi, Basic Stamp and Arduino, as well as the aforementioned systems themselves. Also the LittleBits stuff, including the Korg synth.
Plus essential cables, connectors and adapters, various useful batteries for UPS's, alarm systems and Roomba.
Oh yeah. And cell phones. And a handful of TVs.
And a whole rack full of soldering irons. And the parts cases.
So they're not quite as useless as they've been made out to be.
As for price, I'm not so impoverished that getting the Low Price Always is the overriding criteria for my purchasing decisions.
"Coding" had nothing to do with theaters dropping the movie. What made them drop the movie was a terrorist threat that spoked a bunch of bean counters.
And it wouldn't surprise me to learn that the bean counters had also been the ones that nixed proper security procedures within Sony.
Because IT Doesn't Matter. What matters is getting the Low Price Always.
Slashdot Mirror
"If a grandma had genitals of a man, she would be a grandpa". While I do think that carrying firearms onto the plane is a bad idea, in a critical situation like this, I'd rather a couple of folks on the plane had firearms or weapons of some sort. At the very least the pilots, and maybe some of the other staff.
But the wisdom of the day was that if you don't interfere, no one gets hurt. Flight 93 was the only one that realized the rules had changed in time to take action.
And they did, even if it was only to drag the hijackers down bodily before they could steer into their target. They could have done no more than that if they'd all been packing Uzis.
Guns have their uses, but one of the last places I'd want to use one is in a tube where dozens of people are all packed into a small, basically linear space surrounded by control and fluid lines. You'd be better off issuing boxcutters to the passengers and crew. Or maybe 3-oz bottles of something nasty.
You don't need guns to take down a bunch of clowns with boxcutters when there's 200 of you and 4 of them.
What was lacking wasn't firearms, it was the realization that the usual way of dealing with hijackers no longer worked. Flight 93 was warned in time to change to a more active response. If they'd known even earlier, they might have even been able to save the pilots.
Last century's motto: To Boldly Go
This century's motto: Always the Low Price. Always.
Nah, it's nothing new. Friend of mine when we were teens had a 70s compact car - a VW Scirocco, or maybe that replaced this one - where you had to pull the engine to change the oil filter. No joke. The guy on the assembly line could install the filter easily because the engine wasn't in the car yet.
I've seen video footage from factories where the engines were installed from the bottom. No auto shop I've ever been in was equipped to do that that I know of. They had to do it the hard way, removing the hood.
I had a dentist whose Jaguar supposedly required the engine to be pulled in order to do a tune-up every 6 months because it was the only way to get at the sparkplugs. Some Cadillacs reportedly had this issue as well.
A Ford I used to own required pulling the dashboard to replace the heater core. Which is bad enough, but the mechanics cracked the dash. since by that time the plastic had become more brittle. The Ford I owned prior to that one could have its heater core replaced by simply undoing 2 wingnuts, removing a cover plate and lifting.
Electronics, however, are a different matter. You often have no more than 2-3 chips doing the heavy work and they're wave-soldered to the board, So lots of luck there. I did have to heave a perfectly good TV set with excellent picture though because apparently a 15-cent low-voltage capacitor managing the vertical height went sour and there were no available schematics.
Stuff like calculators are the worst, though. Bought one in the late 1980's for $5. Fully warranteed. If it failed, send $6.50 for postage and handling to the warranty deport. When the batteries died, a new set ran about $7.
Companies WebEx all the time these days. Business travel is an expense many attempt to keep to a bare minumum.
I could care less if I can physically see the people I'm talking to. I can do pretty much anything necessary via Skype, IM, and email.
Why not open a car factory in Bora Bora?
Or Spring Hill, TN?
What you DON'T want is me.
I'm not a social person. Put me in a room full of people and I'll no more spontaneously talk to them than if I was on the moon. I can communicate, but I can be at least as confusing face-to-face as I can via email.
I have a reputation for doing good stuff, but when I work, I work in my own little world and I don't want to be interrupted while I do it. I have a reputation for adapting to user's needs and even suggesting things that the users didn't realize could make their jobs easier, but I need time in between to think things through and to prototype solutions. Not a lot of time - I did Agile years before Agile was "invented", but nevertheless, time. Uninterrupted time.
Shoving me in a cage of twittering birds is going to reduce my productivity, not enhance it.
You can very easily skive off while sitting in an office in an open cubicle next to the boss. Been there, done that.
The boss cannot discern anything more about the productivity/quality of a programmer's work by looming over him/her/it, than he/she/whatever can do when the programmer is on the other side of the planet.
That's because the programmer's quality and productivity have nothing whatsoever to do with what they're physically doing. You could sink the Titanic with stories of people who got their breakthrough insight in the shower, sleeping or even playing "Pong".
Furthernore, no decent manager is going to be scanning every line of code from every employee. There are better things to be doing.
There are ways to determine whether a programmer is doing good work and they don't require physsical oversight. First, by the rate of development of the project and the parts of the project assigned to the programmer and the number of times it needs emergency fixes. That, after all, is what you're REALLY supposed to be paying for, not a chair ornament. It's not overnight, but then, if you're not running a revolving-door shop, things add up over time. You want full chairs, invest in inflatable Bozo dolls. If you want "busy" attach clockwork to make them rock back and forth.
The second way is by listening to ones employees. The programmer's peers are the ones who know the actual code and they'll complain or praise it, whether it's relatively innocuous stuff like the guy I know who never wrote a subroutine if he could cut-and-paste to the person who always delivered overnight but never checked for null pointers.
Free as in frei. Kostenlos bier für alles!
So, can Hinglish speakers comprehend those who speak Bombay Welsh?
That's the one thing that pirated works can never offer.
Once a performance is recorded, it's "dead". It will be (barring damage) the same performance the 10,000th time as it was the first.
Granted, there are certain things that just cannot be done live and without studio support. And I wouldn't give them up, but the spontaneity - and even the flubs - of a live performance have magic all their own.
Even in case #1, there is sometimes things that can be done. For example, speculative execution. If you can boil down to a small number of choices as a result of the first operation, then it may make sense to compute both outcomes. Or there may be some other intermediate value that might be needed in only some outcomes. But this requires application-specific knowledge usually to know exactly what is allowable and what the payoff would be. You wouldn't want to create a situation where executing both cases affects a global resource. So you would need a language expressive enough to hint this information to the compiler.
Very good point. A high-level equivalent to the predictive processing at the CPU hardware level!
There are actually 3 kinds of tasks as far as parallelization goes.
1. Totally linear tasks. Each step relies on the output of its predecessor. Thus nothing can begin before its time. Obviously, handing this sort of work off to a parallel system is a waste of time.
2. Simple parallel tasks. This is the case where you can do a lot of trivial operations in parallel. The computer equivalent of a bunch of people using 4-banger calculators. As long as the tasks indivudually take longer to run than they do to schedule and collect, this is an ideal use case for massively parallel array processors.
3. Complex parallel tasks. This is simply case #2, but armed with HP advanced function calculators. The individual processors would be more than just basic gate arrays and thus able to perform complex math functions in parallel. Not as cheap to scale up, but better than waiting out linear time.
Of course, this is for the ideal world. Real-life heavy-computing scenarios may have components for 1, 2, or all 3 of the above.
It's the coldest hour of the end-of-December day and the current outside temperature here is 19C. The interior temperature is about 25 because the heat from yesterday's sun is still lingering in the building. Heat hasn't kicked on in about 2 weeks.
Spend enough time in that sort of environment and your internal thermosensors reset. Working at a company office desk, I spent a great deal of time contemplating setting fire to to whatever I could burn to keep warm. The temperature? 72 degreees Fahrenheit. Good enough if you're an executive wearing a 3-piece suit and wandering around annoying the grunts, but Bob Cratchitt frigid if you're just sitting there pushing paper and bits.
I can tolerate cooler these days, though. Used to be my summer A/C setting was 83F (27C). Now I prefer it a degree or 2 colder.
Or were you trying to suggest that the Indians north of the Rio Grande also had huge cities? If so, then, to put it politely, you were mistaken....
Depends on your definition of 'huge' but 100,000 people seems reasonably large.
Precisely. Besides, IIRC, at the time, Mexico City was significantly more populous than Paris. And had much better sanitation. But I'm not sure if I ever read it in a textbook.
Look at this, we have a post on "Chinese censorship" and here is the first reference to Tienanmen square, an event that took place in 1989.
Whoever controls the Past Controls the Present.
As an American, I wonder... how do you do Christmas in warm weather?
At the beach!
Seriously, I live in a sub-tropical climate in the USA and for years one of the biggest conundrums we had was what to do with Xmas stockings when most houses didn't have fireplaces!
Oh, and it's summer here - a balmy 25 degree Celsius (=77 degrees F). And I live three minutes walk from the beach ....
It's winter in Florida, but yesterday's high was 80 degrees F. And nowhere in the state is more than a 2-hour drive to the beach.
Winter isn't what it used to be.
All that does is demonstrate the inadequacy of the English language as it presently exists and applies to the issues of the day.
If I have been granted a right to restrict copying of a work of mine except to those who I wish to receive a copy (whether for money, other considerations or even gratis), that's not theft, it's true. It's infringement and something of value HAS been lost - my remuneration or permission. Stolen, if you prefer to allow for the possibity that something that was destroyed in potentio - payment - is "theft".
But it's a poor representation to call it theft, because people are looking at the copy, not the intangible right as the thing that was stolen. So on the one hand you have the "pirates" "justifying" themselves because it's "not theft" and on the other, you have the fat cats saying "you STOLE our song/movie/whatever". And I just ran out of quote-marks, I think.
So maybe we just need to invent a new word that can be used to make the distinction clearer. I don't agree with a lot of the current copyright and patent laws, but at least let's not muddy the waters with false meanings when discussing them.
That's why I hate it. The "mirror" can throttle, but that's just a drop in the bucket. Only if all the reflection mirrors are throttling can it help, and the larger the mirror, the larger the number of apparently legitimate requests would be, so it's harder to make them good throttles.
Plus, not only the target system is getting blasted. The mirroring systems are getting a pretty heavy load. They can throttle this, but then they risk choking off the legitimate requests, since a legitimate request and a reflection attack request are indistinguishable.
I think the only real cure would be to drop UDP. You can't do a reflection attack from TCP, since you cannot spoof the sender. At least without some major mucking around with the Internet as a whole.
UDP reflection attacks are one case when "economies of scale" work in reverse.
I can throttle such attacks on my DNS servers, since I'm only serving for a few domains and there's not much urgency.
For large ISPs, however, there's going to be a lot more legitimate traffic for a lot more domains and if you bounce a request, you may be turning away the one legitimate customer in the flood.
I hate UDP reflection with a passion. Ordinary attacks are annoying enough, but if I ever got my hands on the people behind this, I could cheerfully do things involving crudely-sharpened objects, live coals and molten plastic. The CIA could take notes and learn new techniques.
There's no reason to shop at a retailer unless you're desperate and need something now
Sometimes it's actually nice to fondle the merchandise. There's only so much you can get out of an online catalog, especially since the pictures are usually pretty poor.
I was beginning to believe the Radio Shack mantra, until I wandered into my neighborhood store the other day. They had an impressive array of sensors and kits for stand-alone, Rasberry Pi, Basic Stamp and Arduino, as well as the aforementioned systems themselves. Also the LittleBits stuff, including the Korg synth.
Plus essential cables, connectors and adapters, various useful batteries for UPS's, alarm systems and Roomba.
Oh yeah. And cell phones. And a handful of TVs.
And a whole rack full of soldering irons. And the parts cases.
So they're not quite as useless as they've been made out to be.
As for price, I'm not so impoverished that getting the Low Price Always is the overriding criteria for my purchasing decisions.
"Coding" had nothing to do with theaters dropping the movie. What made them drop the movie was a terrorist threat that spoked a bunch of bean counters.
And it wouldn't surprise me to learn that the bean counters had also been the ones that nixed proper security procedures within Sony.
Because IT Doesn't Matter. What matters is getting the Low Price Always.
Eureka!
Seriously, tell most people it "displaces 5 times as much water" and they're going to be thinking volume, not weight.