Thinking about this some more. I used two "hill climbing" heuristics. For large problems you might need to start with several different random solutions to see if you achieve different end solutions. In practise, these heuristics converged on the one and only solution every time for the "small" (100 city) problem I tested.
First, in your path, remove one city, and try plugging it in the route at another spot. This is n^2.
Second, in the path, take two edges, AB and CD. See if the path with AC and BD is shorter. (This reverses the route between B and C... AD, BC would cut the route in two seperate pieces) This is n^2 as well.
With these two heuristics, you can hill-climb, and achieve a reasonable solution in something like n^3 time. (the "repeat until no more changes" is probably linear in n. I'm guessing here....).
If for the "cost" for each edge you have to call the route planner, this algorithm can provide the route planner with a "max cost". In each case, when a new route cost needs to be calculated, a max can be given, above which a route is not a candidate. This means that the cost table will be filled with a lot of "> xx seconds to travel from here to there", instead of firm "YY seconds"...
How many paths did you end up testing? I really don't remember.
Thinking about it some more, some heuristics that I used depend on the problem being planar. This is an assumption that goes out the door once you have the route planner give the "cost" of each of the n^2 combinations of places.
Some people "give up" the moment a problem has been proven NP-complete/intractible.
In practise, someone running a couple of errands or delivering 10 or 20 club-newsletters, the n^2 part as well as the 2^n TSP problem is solvable in reasonable time on a modern computer.
FedEx and DHL have a more complex scheduling problem. They have thousands of packages, which need distributing over delivery trucks, and then each delivery truck needs to solve a quite large version of the TSP problem. With some heuristics, this can be done in reasonable time. (i.e. before the trucks need to leave...)
About 20 years ago, I had an IBM PC compatible computer, which ran at a whopping 10MHz, instead of the standard 4.77 MHz.
Innocent as I was, I thought the traveling salesman problem could easily be solved by a computer. I wrote the code, and gave it a "standard" traveling salesman problem from a book. 60 CPU-seconds later, I got the optimal answer. It was a 100 city (or was it 133?) problem.
The book said that a pruning brute force algorithm had been run on a CRAY supercomputer. It had come to the same solution that my PC had.
The problem with scaling to 100 errands to run is that you need the full 100 x 100 distance table (cost matrix), and this would be "time in seconds between each of the stops", as issued by a route planner. Calculating 10 thousand routes is something that would still require quite some CPU time, whereas the resulting traveling salesman problem could then be solved on a modern computer in a matter of seconds. Still, with some smart pruning, the number of full-paths that need calculating can probably be significantly reduced.
Interesting problem! (TNT and FedEx probably have several PHDs working on this as we speak....)
From the printed page, I can tell wether it's TeX or Word. Apparently storing the TeX source would be a good idea. The rendering would then be a whole lot better.
I live at 5 degrees east. Thus, I know that because I'm at GMT+1, the sun will be exactly in the south at 12:40 PM. If we change to the "leap hour" strategy, I'll have to remember what the offset is now, and that offset will change all the time...
The register descriptions are one thing. A little wording around that to explain what is what and how it works is another thing that changes things from very tedious to workable.
In a processor manual you see that the 0xaa opcode has the "ADD" mnemonic, and does a = a + , and sets the flags. This AMD manual has the level: "0xaa is called ADD", and nothing more. With a bit of imagination and some previous knowledge how things normally work you can probably figure it out, but the exact operation still has to be reverse engineered.
Lets hope that hte "More to come" is serious. (or that the developers immediately know what's going on if they just know the mnemonics for the registers....)
In the past, when people thought you'd done something, and you said you didn't do it, you often got a chance to PROVE you didn't do it. When that comes to "I was in bed sleeping, how can i prove my alibi?" innocent people start to get thrown in jail it becomes nasty. So nowadays, the consensus is that you are innocent until PROVEN guilty. This proves much harder, and sometimes the bad guys walk, because the DA can't prove beyond reasonable doubt that some bad guy was guilty.
With this in mind, the burdon of proof is put on the manufacturer, with the intention that this will make it easier for consumers to claim warranty. Otherwise the manufacturers could easily claim "you broke it, your fault", the burden of proof "But I didn't break it, this defect was in your product all along" would then fall on the consumer. Without good lawyers, and the ability to hire "independant techincal experts", this would likely fail.
If indeed the consumer breaks things and the manufacturer manages to prove beyond reasonable doubt that the consumer did indeed break the product, then it is only fair that the consumer ends up with having to pay the bill for repairs.
Indeed there remains a big difference between being right, and getting your right (i hope that that's English...:-).
I used to get my computer stuff at a shop, where they would just give me new stuff out of their stock, and would send my broken stuff to the manufacturer. Whenever that came back they just restocked the returned product.
When they no longer stocked that product. I was out of luck: they would send my stuff to be replaced, and if that took 6 weeks, (which is obviously too long for stuff that drops 50% in price in such periods), bad luck for me. I have the right to "my money back" (revert the sale), if they lie to me, or if this would be considered "too long" for normal people. Very difficult to actually get your money back or a quicker turnaround.
Same to you. In this case, the consumer that started the thread is European. He should know his rights before proceding with this. Thanks for your clarification that no, they don't have to show the video. I was exagerating a bit. In practise it seems "pretty darn difficult" to prove abuse by the consumer.
There is an error on page 175. It says: "others - reserved" in the if (D1GRPH_DEPTH = 0x2) section. All options are defined, so there are no others that can be reserved.
The manufacturer's warranties do not cover broken plastics or hinges. There is just one problem with that.
Here in Europe, the government has found that too many shops hide behind excuses like this.
So, first of all, you have a contract with the shop who sold you the item, not with some manufacturer in China. If the shop sends it back to the manufacturer, fine. But if the manufacturer thinks it's not covered, the shop needs to cover warranty issues anyway. Not that they don't often try to hide bihind this, but legally, they shouldn't.
Secondly, you may expect items you buy to have a "nominal lifetime". For things like fridges you should expect it to last at least 10 years. If it breaks after 8, you share the repair costs with the shop 80/20. After a while though, you have to PROVE it was a hidden defect all along, and not something caused by normal use.
In the first year however, warranty says that it IS a hidden defect, unless the shop proves otherwise. So when they come up with a video showing you dropping the laptop, you're out of luck.
Oh... By the way, I boot Knoppix on my new laptop,
gzip/dev/hda | ssh someplace dd of=hda-vista.gz
so that in case of this kind of trouble, I can repeat it with "hda-linux.gz" as the destination, restore the vista partition, and even claim I started working with it yesterday....
To be fair to Skype you have to admit that 85% of the world's computers turning off at the same time is not an event a normal person would predict nor could such an event be tested in advance. Excuse me? It happens every XXth tuesday of the month!
Maybe a long time ago when you were designing the system you wouldn't expect it. But from usage logs, you'd see sudden spikes on certain tuesdays and figure out what it's from. Normal people do not wait for a "massive failure" to occur before upgrading the server to have more capacity or improving the algorithm to require less of the servers.
The Ethernet port is for laptops that don't have wireless, or for people who simply prefer an Ethernet connection over WiFi, which could potentially become congested in an aircraft if in-flight internet usage becomes popular.'"
In my home the internet link is about 8 mbps, while my wireless is 54mbps. The same will hold in an airplane. Actually I doubt they will be able to get 8mpbs for the whole plane.
So congestion will not happen, unless there is an inflight content-server that becomes popular, or people do peer-to-peer on the plane.
Bandwidth on wired ethernet is not unlimited either, and will similarly get congested at a slightly (100 mbps instead of 54) or significant higher bandwidth.
Ok, then I'll just publish the number prior to it, the number after it and ask good ol' Seseame-Street style "what's missing?" Fine. Then legally your riddle becomes a copyright cirumvention device. The DMCA prohibits distribution of such devices.....
They sit on the networks at scan for what are likely copyright violating files (after that they should have to download the files to prove that their copyright has actually been violated, though as far as I know no one has argued this point in court).
In a recent Dutch court ruling, the judge came up with this by him/herself. The argument was that as the RIAA was also trying to mess with P2P networks by anouncing honeypot/dummy files, they couldn't be sure that the file offered wasn't a dummy/honeypot file.
Yes. And because the spammers take random FROM addresses from their spam-list, I now get the autoreplies from YOUR spambot. You are shoving YOUR spam problem into MY face.
For whatever it's worth, if this seems to be in response to some spam, I'll comply, declare that I'm human, and respond. If it seems to be in response to something I sent (reply to something you might have asked on a mailing list perhaps), I'll gladly assume you don't want to see my reply. I took the trouble of providing you with an answer to your question. I'm not going to jump through more hoops for you.
Actually, it's very easy to test this model without "going forward". You simply train it on previous wars except for the last 20. Then you see how it predicts the last 20 wars. Ah, yes. That mistake has been made before. So guys doing this research do this, find it accurate to 10% (i.e. it gets it wrong 90% of the time). What do you think they will do? Publish: "We made a model, and it didn't work"? I don't think so. They will tune their methods until it gets it right 80% of the time, and then publish it.
So while the model didn't get trained on the last 20 years of data, it dit get meta-trained.
The problem with this is that the model was "trained" on the same historical data on which it is eventually tested. This doesn't prove anything.
As an example, a defence contractor once built a system that would recognize wether or not a tank was in a picture. First the system was trained on half the "with tanks" and half the "without tanks" pictures. Next the system got a good percentage correct on the second half of the pictures. It turns out the "with tanks" pictures had been taken on a sunny day, and those without on a cloudy day. So the system was actually telling "sunny" or "cloudy".
In this case, it could very well be that her system predicts the outcome of the war, based on the weather in tokyo 6 weeks before the start of the war. This example was chosen so that you, not an expert in this field, immediately can dismiss this as a nonsense predictor. But as the model gets more complicated, and you feed it lots of parameters that might seem relevant, even the experts will no longer be able to see the value of such complicated predictions. At some point you just have to "trust the computer".
Aerodynamics: Yes. We understand the underlying principles, we've verfied the predictions made by the models in real life, and found that it matches very good.
In this case: No. Before I trust such a model, it would need to be verified (as is, no modifications allowed!) against say at least 20 wars that haven't started yet. If it preditcs the outcome of those correctly, the model has merit.
.. power goes out...... less than twice per year. Doesn't seem very realistic to me.
Hmmmmm..... You mean you live in a country where the power goes out more than once every three years or so on average? Where is that? Remind me to avoid that country in the future.
Power outages have become more of a problem since the market was privatized. but that doesn't mean we get more than one outage every few years or so.
Having read a few software patents lately, most of them are too obvious. This means that given an operating system problem, an average engineer at Microsoft will come up with a very similar solution as an average engineer working on Linux.
If the Linux kernel holding that idea was first, we have prior art. That's easy. If Microsoft filed for the patent earlier, we're in trouble. If indeed it is the obvious way to do things, it will be hard to debunk ("It is really obvious to a reasonable engineer, your honor"), and it will be hard to work around.
I once ran into a firewall that would block outgoing requests to the internet, from say "internet explorer", but allow an application to open a port and then allow people on the internet to connect to that service: "You opened the port, so it must be OK".
Hiring external testers can be a huge expense if done right, And a huge pain in the ass if done wrong.
I developped a project where an external "security expert" was hired. His report consisted of the nmap output against the product. So the system was supposed to do something with web pages. He lists port 80 as open (and thus a vulnerability). Management wants the list cleared out. That's the way it goes.
Contrast is a quality measure for LCD screens. Because of the way LCDs (or DLPs) work, there is always some leakage of light, even when a pixel is completely off. If the amount of light that leaks through is only 1/2000th of what comes through when a pixel is white, that's pretty good.
For LED technology, sending about "0" current through a led, or to put "0" voltage over it, is fairly easy to achieve electronics-wise. This gives about "0" light(*), meaning a contrast ratio of a million or a billion or better is easily achieved.
(*) Sarcasm: there is going to be exactly zero light.
History After getting used to two monitors at work, I installed two at home. Then I wanted a video-projector, so I added a PCI card for the beamer. Then the computer got moved to a different place in the house, and my old 17" monitor stood next to it. So I hooked it up to the third video output of my computer. So now I have three screens! my experience I really love the new setup. The small things that are need a glance every now and then, like network load monitors, or the music player I can put away on the third screen. This frees up the two main screens for real work. For example, while designing a PCB, I can have the schematics on one screen, the layout I'm working on on the second, and man pages or documentation of the chips I'm working with on the third. Really I recommend you get a third monitor....
Wouldn't speed be the most important factor when designing airplanes? Yes. Except that when you pass the sound barrier (or come too close) aerodynamic effects cause fuel costs to skyrocket (pardon the pun). So, Mach.89 is close to the best you can achieve.
What kinda damage would this make if you crash it into a building? I'd think: "Total destruction".
History shows that if you crash a big plane into a skyscraper, the building is destroyed. If you crash a big plane into a large, horizontally layed out building, you make a small dent.
You don't need an A380 to fully destroy large buildings like the twin towers. You will make a slightly larger dent in a large building like the pentagon.
Slashdot Mirror
Thinking about this some more. I used two "hill climbing" heuristics. For large problems you might need to start with several different random solutions to see if you achieve different end solutions. In practise, these heuristics converged on the one and only solution every time for the "small" (100 city) problem I tested.
First, in your path, remove one city, and try plugging it in the route at another spot. This is n^2.
Second, in the path, take two edges, AB and CD. See if the path with AC and BD is shorter. (This reverses the route between B and C... AD, BC would cut the route in two seperate pieces) This is n^2 as well.
With these two heuristics, you can hill-climb, and achieve a reasonable solution in something like n^3 time. (the "repeat until no more changes" is probably linear in n. I'm guessing here....).
If for the "cost" for each edge you have to call the route planner, this algorithm can provide the route planner with a "max cost". In each case, when a new route cost needs to be calculated, a max can be given, above which a route is not a candidate. This means that the cost table will be filled with a lot of "> xx seconds to travel from here to there", instead of firm "YY seconds"...
How many paths did you end up testing?
I really don't remember.
Thinking about it some more, some heuristics that I used depend on the problem being planar. This is an assumption that goes out the door once you have the route planner give the "cost" of each of the n^2 combinations of places.
Some people "give up" the moment a problem has been proven NP-complete/intractible.
In practise, someone running a couple of errands or delivering 10 or 20 club-newsletters, the n^2 part as well as the 2^n TSP problem is solvable in reasonable time on a modern computer.
FedEx and DHL have a more complex scheduling problem. They have thousands of packages, which need distributing over delivery trucks, and then each delivery truck needs to solve a quite large version of the TSP problem. With some heuristics, this can be done in reasonable time. (i.e. before the trucks need to leave...)
About 20 years ago, I had an IBM PC compatible computer, which ran at a whopping 10MHz, instead of the standard 4.77 MHz.
Innocent as I was, I thought the traveling salesman problem could easily be solved by a computer. I wrote the code, and gave it a "standard" traveling salesman problem from a book. 60 CPU-seconds later, I got the optimal answer. It was a 100 city (or was it 133?) problem.
The book said that a pruning brute force algorithm had been run on a CRAY supercomputer. It had come to the same solution that my PC had.
The problem with scaling to 100 errands to run is that you need the full 100 x 100 distance table (cost matrix), and this would be "time in seconds between each of the stops", as issued by a route planner. Calculating 10 thousand routes is something that would still require quite some CPU time, whereas the resulting traveling salesman problem could then be solved on a modern computer in a matter of seconds. Still, with some smart pruning, the number of full-paths that need calculating can probably be significantly reduced.
Interesting problem! (TNT and FedEx probably have several PHDs working on this as we speak....)
From the printed page, I can tell wether it's TeX or Word. Apparently storing the TeX source would be a good idea. The rendering would then be a whole lot better.
I live at 5 degrees east. Thus, I know that because I'm at GMT+1, the sun will be exactly in the south at 12:40 PM. If we change to the "leap hour" strategy, I'll have to remember what the offset is now, and that offset will change all the time...
I've taken a quick look at one of the manuals.
The register descriptions are one thing. A little wording around that to explain what is what and how it works is another thing that changes things from very tedious to workable.
In a processor manual you see that the 0xaa opcode has the "ADD" mnemonic, and does a = a + , and sets the flags. This AMD manual has the level: "0xaa is called ADD", and nothing more. With a bit of imagination and some previous knowledge how things normally work you can probably figure it out, but the exact operation still has to be reverse engineered.
Lets hope that hte "More to come" is serious. (or that the developers immediately know what's going on if they just know the mnemonics for the registers....)
In the past, when people thought you'd done something, and you said you didn't do it, you often got a chance to PROVE you didn't do it. When that comes to "I was in bed sleeping, how can i prove my alibi?" innocent people start to get thrown in jail it becomes nasty. So nowadays, the consensus is that you are innocent until PROVEN guilty. This proves much harder, and sometimes the bad guys walk, because the DA can't prove beyond reasonable doubt that some bad guy was guilty.
:-).
With this in mind, the burdon of proof is put on the manufacturer, with the intention that this will make it easier for consumers to claim warranty. Otherwise the manufacturers could easily claim "you broke it, your fault", the burden of proof "But I didn't break it, this defect was in your product all along" would then fall on the consumer. Without good lawyers, and the ability to hire "independant techincal experts", this would likely fail.
If indeed the consumer breaks things and the manufacturer manages to prove beyond reasonable doubt that the consumer did indeed break the product, then it is only fair that the consumer ends up with having to pay the bill for repairs.
Indeed there remains a big difference between being right, and getting your right (i hope that that's English...
I used to get my computer stuff at a shop, where they would just give me new stuff out of their stock, and would send my broken stuff to the manufacturer. Whenever that came back they just restocked the returned product.
When they no longer stocked that product. I was out of luck: they would send my stuff to be replaced, and if that took 6 weeks, (which is obviously too long for stuff that drops 50% in price in such periods), bad luck for me. I have the right to "my money back" (revert the sale), if they lie to me, or if this would be considered "too long" for normal people. Very difficult to actually get your money back or a quicker turnaround.
Relax, get some coffee or tea or whatever,
Same to you. In this case, the consumer that started the thread is European. He should know his rights before proceding with this. Thanks for your clarification that no, they don't have to show the video. I was exagerating a bit. In practise it seems "pretty darn difficult" to prove abuse by the consumer.
There is an error on page 175. It says: "others - reserved" in the if (D1GRPH_DEPTH = 0x2) section. All options are defined, so there are no others that can be reserved.
The manufacturer's warranties do not cover broken plastics or hinges.
/dev/hda | ssh someplace dd of=hda-vista.gz
There is just one problem with that.
Here in Europe, the government has found that too many shops hide behind excuses like this.
So, first of all, you have a contract with the shop who sold you the item, not with some manufacturer in China. If the shop sends it back to the manufacturer, fine. But if the manufacturer thinks it's not covered, the shop needs to cover warranty issues anyway. Not that they don't often try to hide bihind this, but legally, they shouldn't.
Secondly, you may expect items you buy to have a "nominal lifetime". For things like fridges you should expect it to last at least 10 years. If it breaks after 8, you share the repair costs with the shop 80/20. After a while though, you have to PROVE it was a hidden defect all along, and not something caused by normal use.
In the first year however, warranty says that it IS a hidden defect, unless the shop proves otherwise. So when they come up with a video showing you dropping the laptop, you're out of luck.
Oh... By the way, I boot Knoppix on my new laptop,
gzip
so that in case of this kind of trouble, I can repeat it with "hda-linux.gz" as the destination, restore the vista partition, and even claim I started working with it yesterday....
To be fair to Skype you have to admit that 85% of the world's computers turning off at the same time is not an event a normal person would predict nor could such an event be tested in advance.
Excuse me? It happens every XXth tuesday of the month!
Maybe a long time ago when you were designing the system you wouldn't expect it. But from usage logs, you'd see sudden spikes on certain tuesdays and figure out what it's from. Normal people do not wait for a "massive failure" to occur before upgrading the server to have more capacity or improving the algorithm to require less of the servers.
The Ethernet port is for laptops that don't have wireless, or for people who simply prefer an Ethernet connection over WiFi, which could potentially become congested in an aircraft if in-flight internet usage becomes popular.'"
In my home the internet link is about 8 mbps, while my wireless is 54mbps. The same will hold in an airplane. Actually I doubt they will be able to get 8mpbs for the whole plane.
So congestion will not happen, unless there is an inflight content-server that becomes popular, or people do peer-to-peer on the plane.
Bandwidth on wired ethernet is not unlimited either, and will similarly get congested at a slightly (100 mbps instead of 54) or significant higher bandwidth.
Ok, then I'll just publish the number prior to it, the number after it and ask good ol' Seseame-Street style "what's missing?"
Fine. Then legally your riddle becomes a copyright cirumvention device. The DMCA prohibits distribution of such devices.....
Now what were you trying to accomplish?
They sit on the networks at scan for what are likely copyright violating files (after that they should have to download the files to prove that their copyright has actually been violated, though as far as I know no one has argued this point in court).
/dummy files, they couldn't be sure that the file offered wasn't a dummy/honeypot file.
In a recent Dutch court ruling, the judge came up with this by him/herself. The argument was that as the RIAA was also trying to mess with P2P networks by anouncing honeypot
Yes. And because the spammers take random FROM addresses from their spam-list, I now get the autoreplies from YOUR spambot. You are shoving YOUR spam problem into MY face.
For whatever it's worth, if this seems to be in response to some spam, I'll comply, declare that I'm human, and respond. If it seems to be in response to something I sent (reply to something you might have asked on a mailing list perhaps), I'll gladly assume you don't want to see my reply. I took the trouble of providing you with an answer to your question. I'm not going to jump through more hoops for you.
So while the model didn't get trained on the last 20 years of data, it dit get meta-trained.
The problem with this is that the model was "trained" on the same historical data on which it is eventually tested. This doesn't prove anything.
As an example, a defence contractor once built a system that would recognize wether or not a tank was in a picture. First the system was trained on half the "with tanks" and half the "without tanks" pictures. Next the system got a good percentage correct on the second half of the pictures. It turns out the "with tanks" pictures had been taken on a sunny day, and those without on a cloudy day. So the system was actually telling "sunny" or "cloudy".
In this case, it could very well be that her system predicts the outcome of the war, based on the weather in tokyo 6 weeks before the start of the war. This example was chosen so that you, not an expert in this field, immediately can dismiss this as a nonsense predictor. But as the model gets more complicated, and you feed it lots of parameters that might seem relevant, even the experts will no longer be able to see the value of such complicated predictions. At some point you just have to "trust the computer".
Aerodynamics: Yes. We understand the underlying principles, we've verfied the predictions made by the models in real life, and found that it matches very good.
In this case: No. Before I trust such a model, it would need to be verified (as is, no modifications allowed!) against say at least 20 wars that haven't started yet. If it preditcs the outcome of those correctly, the model has merit.
I'm not going to wait around (I hope).
.. power goes out ...... less than twice per year. Doesn't seem very realistic to me.
Hmmmmm..... You mean you live in a country where the power goes out more than once every three years or so on average? Where is that? Remind me to avoid that country in the future.
Power outages have become more of a problem since the market was privatized. but that doesn't mean we get more than one outage every few years or so.
Having read a few software patents lately, most of them are too obvious. This means that given an operating system problem, an average engineer at Microsoft will come up with a very similar solution as an average engineer working on Linux.
If the Linux kernel holding that idea was first, we have prior art. That's easy. If Microsoft filed for the patent earlier, we're in trouble. If indeed it is the obvious way to do things, it will be hard to debunk ("It is really obvious to a reasonable engineer, your honor"), and it will be hard to work around.
to supress those "annoying messages"
I once ran into a firewall that would block outgoing requests to the internet, from say "internet explorer", but allow an application to open a port and then allow people on the internet to connect to that service: "You opened the port, so it must be OK".
Hiring external testers can be a huge expense if done right,
And a huge pain in the ass if done wrong.
I developped a project where an external "security expert" was hired. His report consisted of the nmap output against the product. So the system was supposed to do something with web pages. He lists port 80 as open (and thus a vulnerability). Management wants the list cleared out. That's the way it goes.
Contrast is a quality measure for LCD screens. Because of the way LCDs (or DLPs) work, there is always some leakage of light, even when a pixel is completely off. If the amount of light that leaks through is only 1/2000th of what comes through when a pixel is white, that's pretty good.
For LED technology, sending about "0" current through a led, or to put "0" voltage over it, is fairly easy to achieve electronics-wise. This gives about "0" light(*), meaning a contrast ratio of a million or a billion or better is easily achieved.
(*) Sarcasm: there is going to be exactly zero light.
History
After getting used to two monitors at work, I installed two at home. Then I wanted a video-projector, so I added a PCI card for the beamer. Then the computer got moved to a different place in the house, and my old 17" monitor stood next to it. So I hooked it up to the third video output of my computer. So now I have three screens!
my experience
I really love the new setup. The small things that are need a glance every now and then, like network load monitors, or the music player I can put away on the third screen. This frees up the two main screens for real work. For example, while designing a PCB, I can have the schematics on one screen, the layout I'm working on on the second, and man pages or documentation of the chips I'm working with on the third. Really I recommend you get a third monitor....
Ehh. You're right! In 2004, the canadian transport safety board adjusted the average person weight by about 50 pounds!
Wouldn't speed be the most important factor when designing airplanes? .89 is close to the best you can achieve.
Yes. Except that when you pass the sound barrier (or come too close) aerodynamic effects cause fuel costs to skyrocket (pardon the pun). So, Mach
What kinda damage would this make if you crash it into a building?
I'd think: "Total destruction".
History shows that if you crash a big plane into a skyscraper, the building is destroyed. If you crash a big plane into a large, horizontally layed out building, you make a small dent.
You don't need an A380 to fully destroy large buildings like the twin towers. You will make a slightly larger dent in a large building like the pentagon.