They could have copied concepts from https/SSL, and things would be significantly better.
Maybe I have high standards, but in my opinion the WiFi bunch started with crap (WEP what a joke), and have continued screwing up since e.g. with WPA PSK, once you know the PSK, you can decrypt ALL communications that used that PSK. So it's a waste of time for FamousCafeBrand and friends to get customers to use WPA PSK. Current popular implementations of WPA Enterprise are too hard - too hard for Mr Latte Drinker and too hard for Low Wage Cashier to explain to Mr Latte Drinker.
With the current dismal state of affairs, FamouseCafeBrand has to leave their WiFi access wide open.
What happens then if you see ten open APs called "FamousCafeBrand"? Are they all legit? Or is there MITM going on? You can't tell.
In contrast if things were done right, when connecting, people might get a pop up saying "The AP cert for FamousCafeBrand.wifi does not appear to be signed by a trusted CA" or "The cert does not match", and they get asked "Do you still want to connect?".
People could still click through, but at least you would have done about as much as you can and should.
I would actually be happy if you can prove me wrong - show me a WiFi set up using current available technology that would work easily and securely for the Mr Latte Drinkers, or Aunt Mays of this world.
To spend your money for your own good (and the greater good) when you don't want to do it yourself;).
Of course, most governments do a mix of doing evil, wasting the money and spending it for the greater good. The ratios determine how good the government is.
It's not well thought out. Otherwise you wouldn't have this issue in the first place.
If I want to share my WiFi it isn't easy to make it known of my wishes and my terms and conditions - after all, though I share it, I might say I log access, (mac addresses, urls etc) just in case someone does something illegal, so that if the cops come, I could throw them that bone to chew on, instead of them chewing on me.
If it were well thought out, it would be easy to have secure encrypted _anonymous_ connections: 1) no need for people to enter a password to get encryption 2) people cannot see each other's traffic - snooping is possible in some encryption modes, for example if everyone knows the WEP key, they can figure out each other's traffic, so you'd need some WPA mode, but these require username and passwords, you could give everyone the same username and password, but there's no standard for Windows, Linux, Mac to try "anonymous" usernames and passwords ala anonymous ftp.
And also there would be a standard way to get info about a wifi zone, and to prompt the user if the info/T&C changes, say when you computer connects to a different AP.
You only encrypt if the company laptop already does it by default.
Otherwise it's suicide.
If they ever take a look at your laptop while it's powered on, and access the harddrive that's encrypted, and find child porn, how does that make you look?
With the amount of hysteria going about, I think the safest way for you to browse would be at home where you have more control over your computer.
One Childish N00b (780549): "wonder if she ever noticed that 'the internet' preferred brunettes?" susano_otter (123650): "Let me guess: Your mom is a brunette..." blofeld9999 (1010357): "She is. I know this because I also prefer brunettes" One Childish N00b (780549): "Dad is that you?".
How about stop going around being so trigger happy about sacking and prosecuting people for such "crimes"?
How about actually following the money trail? Are the malware authors and people putting those images up really doing such stuff for free? Someone must be paying for those ads, the creation of child porn sites etc.
There are more serious crimes than possession of some image file, especially an image file that is likely to be downloaded by malware.
Lastly, Linux isn't going to help. The real problem is mass hysteria - lots of people suddenly turning their brains off when they hear a trigger phrase. Sure child porn is bad, but if you really want to fix it, follow the money to the bitter end. Not go around starting stupid witch hunts. The way they do things, I figure it's just a tool for cynical manipulation of a mindless populace.
Obviously the dept didn't value that man's reputation very much ("We stand by our decision" even after the court cleared him) Nor did they appear to factor all that in - the litigation costs. It sounds even more negligent given they passed that guy a non "clean" laptop in the first place.
Anyway, often the problem is the downtime it takes to reimage the machine - esp if it's an old laptop and nobody has an "up to date" and pristine image.
AFAIK normally nobody cares.
Except in this case. I guess someone cared enough to start a witch hunt and this poor chap got the brunt of it.
Someone screams "child porn" and suddenly it's like a mass shark frenzy with blood in the water.
Writing less lines of code = bust your ego? No wonder you like Java.
I've long suspected that Java makes programmers feel like they've done a lot of work at the end of the day (and so deserve their pay), because Java just is a lot more work:).
I'm a perl guy. I try to use stuff in CPAN all the time, and so I end up writing less code.
It's not so much to do with my ego (I'm not one of those - "everything in a single line of perl" coders). Basically I'm not a great coder, so it's best for me to reuse code from better programmers. The less code I write the fewer bugs I make, and are responsible for.
That said I've had to work around or fix bugs in some CPAN libs. But they tend to have fewer bugs - after all many have been used and tested by others years before I even started writing my programs.
Everyone gets a say in it. And part of that "say" is saying someone else is misusing it.
Sometimes I think it's just gay. Damn straight gay it is.
That said, it helps to keep trying to maintain enough standard meanings otherwise we might as well be speaking completely different languages, and never get anything done (Tower of Babel).
Software Engineering: Someone types "make all" and goes for coffee and posts on slashdot.
Believe me this is a fundamental difference, this is the truth. Many people don't get it even when I tell them the truth.
With Civil Engineering a design could cost 10 million and take 1 year, but the actual building could cost 1000 million to build and take 5 years, so the total cost is 110 million and 3 years. Thus if during the design phase the designers say we need two months and 1 million bucks extra to get things right, it is likely that management will say OK.
With Software Engineering a design could cost 1 million and take 1 year, and the actual cost of building it could be 50 cents and take 1 hour. So the design phase takes the bulk of the cost and time. Thus if during the design phase the designers say they need extra time and 100K extra, it's likely that normal management will be very very unhappy.
With Software Engineering the first draft blueprints and plastic/clay models compile and "kind of" run, and Management typically sells those as version 1.0 and 2.0 for the same price as the real product would be (after all it costs about the same to make a first draft as it does to make a final version, if not more).
With Civil Engineering - good luck selling the plastic models to the customer for the same price.
As for software design principles-
1) Figure out what must be done, and what Marketing et all might want later on 2) Figure out what data fields/structures are needed or DB schema is required that will be able to handle 1) and more. Software is often about processing data. If you have no proper place for the data, it makes things a lot harder (and crappier). 3) Create good protocols that can be extended and built on. 4) Insist on stuff being robust and secure - it's not paranoia - high chance of weird data appearing in your module's inputs.
Once you have a huge system, trying to shove an incompatible protocol or schema onto it is not going to be easy. Look at the move of IPv4 to IPv6 - it sure isn't going at a quick pace.
As I've said before- He may be an asshole, but he is an asshole with taste. So it's funny that some recent stupid article was talking about emulating Steve Job's style of management. You can emulate assholeness but how'd you emulate taste:).
People know he is right when he yells at them that the curve of the product is wrong, or that that's not insanely great and they can do better. Of course when they finally get it right, Steve Jobs tells them it's insanely great.
I suspect the workers are in something like one of those abusive relationships.
And the reality distortion field works on salaries too;).
Steve Wozniak was fairly tolerant of Jobs, so perhaps Apple looks for workers similar to Woz - money not a big deal, can put up with Jobs etc.
I have news for you, we're already living in a world where that "baby" is already _dead_. So if we throw it out with the bathwater, I don't see how things would get worse.
1) There are so many examples of cases where people/companies/organisations/countries kept secrets, but complex stuff was still reverse engineered or reinvented independently. 2) People are using patents to hold monopolies for very long periods (as technology changes slightly, they patent a variation and so on), and for anticompetitive tactics.
The fact is most (ignoring the perpetual motion junk) patents are keeping competitors out far longer than it would take to reverse engineer or reinvent them if they were kept secret.
So where's that benefit you cite?
Give me an example of something currently patented that when kept secret could not be reverse engineered within 10 years once it gets to market (national security stuff doesn't get patented, or is effectively kept secret anyway once the Gov finds out).
It currently takes very many years to get a patent (in various countries too). So shortening the terms isn't going to be useful - you might as well throw the system away.
That's actually why I think patents aren't very useful.
If someone is really innovative even 30 years of monopoly isn't enough to help them - since most people won't get it.
But 30 years of monopoly would be terrible for > 99.99% of the approved patents (which are mostly pretty obvious - e.g. once you encounter the problem, the solution is easily found by anyone competent in the field).
The real innovators are so many steps ahead - they'll think of various problem, then the solutions, and then the problems with the solutions, and then the solutions for those problems, and so on, till they are decades ahead of everyone else.
As for those who say you should actually implement stuff to be able to claim a patent, I give the example of Douglas Engelbart and his team - they actually implemented a lot of stuff, and most people didn't get it till many decades later.
So to me I don't really think there should be patents on inventions - nowadays > 99.99% of them are just trivial junk that clutter up everything and get in the way of real progress. As is they are a net minus to the world. Giving 20 year monopolies to such "innovators" is a travesty, and allowing them to make a minor change and thus extend the monopoly for even longer is crazy - how does that encourage innovation?
If you want to reward innovators, I'd say we should have Prizes for Innovation that are awarded years after - much like the Nobel Prizes. After 10 or 20 years we should be able to tell whether something is really innovative and important.
Perhaps the application fees could go to a fund used to award the prizes and for administrative costs. Money could also come from other sponsors.
Well if he can play his desired game NOW in its full "maximum quality" glory, and he typically spends USD400 a night on entertainment, then it could actually help him save money.
Basically he spends USD400, plays computer games for a few nights, and actually ends up with more money than he would otherwise (I actually know someone who did save some money in a similar way). In contrast if it were USD4000 for a vid card, the calculation could be different - he could get bored of the various games and go back to spending more.
The other way it could be an investment is if he's a professional gamer. Higher fps is better for many games.
If the clients want to start watching at arbitrary times rather than wait for the "next slot to come along" you'll probably need to cache "near" the clients anyway.
The benefit from multicast isn't that great in that scenario.
Multicast is good if you have many clients downloading the same thing at the same time.
That's not quite what people are doing with Youtube.
RTGs don't provide that much power for the weight (and generate quite a lot of heat). Petrol has about 34 megajoules per litre. 50 litre tank = 1.7 gigajoules.
From the wiki, a 39kg RTG can generate a max of 390W (electricity) and 7.2kW(heat). 390W of electricity after 8 hours will give you 11 Megajoules - the amount of energy in 330ml of petrol (think size of soda can), and that assumes 100% efficiency.
Now if you can also convert all of that 7.2 kilowatts of heat to energy - making a total of about 7.6kW (assuming 100%) then you'll have the equivalent of 6.4 litres of petrol after 8 hours (which is about 1/8th of a full tank).
And that 39kg does not include the weight and size of your supercapacitor bank.
How far can you go on 330ml of petrol? Or 6.4 litres of petrol? Before you need to stop for 8 hours?
If you can get 50% efficiency and somehow not use too much space and weight, then perhaps it's not so bad. On most days I can leave my car stationary for 20 hours and not travel very far.
Uh the main benefit of akamai is that an ISP can for example put youtube mirrors on their own network and not have thousands of their customers streaming youtube through their skinny connections to other ISPs.
It's not so much to do with RTT as it is to do with making more efficient use of bandwidth via caching.
"You do realize that private businesses have never caused a war (on their own at least) that resulted in deaths of thousands and in worst case millions of persons."
Yeah, they just get governments to start wars for them. Look up United Fruit Company.
Governments that stay in power maintain a monopoly on violence.
If there weren't pesky (or powerful enough) governments around, what do you really think companies like that would do? Look up the East India Company.
Many large companies would be really happy if there were no powers higher than them that could say "Hey our citizens don't like that Anglo Persian Oil Company is taking more than 80% of net profits from our country's oil".
Once there's more than a few hundred people you cannot escape from a government. Whether it's a democratically elected one, or or one where a Company is the Government, or a Dictatorship you'll end up with a government one way or another.
"Trains should be 3 feet high and passengers loaded like peas in a pod. "
I doubt that's a good idea, it's unlikely that you'd manage to convince people to put up with that AND the near requirement for passengers to wear diapers or something similar.
If a trip takes more than a few hours and there are many passengers on the train the probability of someone needing to go to the loo is going to be very high.
"Tall" trains can already go 250-300kph. How much faster will such midget trains be after sacrificing the ability for passengers to easily walk to the toilet or elsewhere while the train is moving?
Slashdot Mirror
Thought out better? No I don't think so at all.
They could have copied concepts from https/SSL, and things would be significantly better.
Maybe I have high standards, but in my opinion the WiFi bunch started with crap (WEP what a joke), and have continued screwing up since e.g. with WPA PSK, once you know the PSK, you can decrypt ALL communications that used that PSK. So it's a waste of time for FamousCafeBrand and friends to get customers to use WPA PSK. Current popular implementations of WPA Enterprise are too hard - too hard for Mr Latte Drinker and too hard for Low Wage Cashier to explain to Mr Latte Drinker.
With the current dismal state of affairs, FamouseCafeBrand has to leave their WiFi access wide open.
What happens then if you see ten open APs called "FamousCafeBrand"? Are they all legit? Or is there MITM going on? You can't tell.
In contrast if things were done right, when connecting, people might get a pop up saying "The AP cert for FamousCafeBrand.wifi does not appear to be signed by a trusted CA" or "The cert does not match", and they get asked "Do you still want to connect?".
People could still click through, but at least you would have done about as much as you can and should.
I would actually be happy if you can prove me wrong - show me a WiFi set up using current available technology that would work easily and securely for the Mr Latte Drinkers, or Aunt Mays of this world.
Volcanos relocate population centers.
As far as I know, the New Zealand North Island is full of volcanos. There are many people living near Mount Fuji, so on and so forth.
Most people just don't think it will happen to them.
Same reason why lots of people in California are living in earthquake zones even though they know a Big One will eventually hit California.
And that's what Governments and leaders are for.
;).
To spend your money for your own good (and the greater good) when you don't want to do it yourself
Of course, most governments do a mix of doing evil, wasting the money and spending it for the greater good. The ratios determine how good the government is.
It's not well thought out. Otherwise you wouldn't have this issue in the first place.
If I want to share my WiFi it isn't easy to make it known of my wishes and my terms and conditions - after all, though I share it, I might say I log access, (mac addresses, urls etc) just in case someone does something illegal, so that if the cops come, I could throw them that bone to chew on, instead of them chewing on me.
If it were well thought out, it would be easy to have secure encrypted _anonymous_ connections:
1) no need for people to enter a password to get encryption
2) people cannot see each other's traffic - snooping is possible in some encryption modes, for example if everyone knows the WEP key, they can figure out each other's traffic, so you'd need some WPA mode, but these require username and passwords, you could give everyone the same username and password, but there's no standard for Windows, Linux, Mac to try "anonymous" usernames and passwords ala anonymous ftp.
And also there would be a standard way to get info about a wifi zone, and to prompt the user if the info/T&C changes, say when you computer connects to a different AP.
So the tech still needs a fair bit of work.
You only encrypt if the company laptop already does it by default.
Otherwise it's suicide.
If they ever take a look at your laptop while it's powered on, and access the harddrive that's encrypted, and find child porn, how does that make you look?
With the amount of hysteria going about, I think the safest way for you to browse would be at home where you have more control over your computer.
One Childish N00b (780549): "wonder if she ever noticed that 'the internet' preferred brunettes?"
susano_otter (123650): "Let me guess: Your mom is a brunette..."
blofeld9999 (1010357): "She is. I know this because I also prefer brunettes"
One Childish N00b (780549): "Dad is that you?".
How about stop going around being so trigger happy about sacking and prosecuting people for such "crimes"?
How about actually following the money trail? Are the malware authors and people putting those images up really doing such stuff for free? Someone must be paying for those ads, the creation of child porn sites etc.
There are more serious crimes than possession of some image file, especially an image file that is likely to be downloaded by malware.
Lastly, Linux isn't going to help. The real problem is mass hysteria - lots of people suddenly turning their brains off when they hear a trigger phrase. Sure child porn is bad, but if you really want to fix it, follow the money to the bitter end. Not go around starting stupid witch hunts. The way they do things, I figure it's just a tool for cynical manipulation of a mindless populace.
Obviously the dept didn't value that man's reputation very much ("We stand by our decision" even after the court cleared him)
Nor did they appear to factor all that in - the litigation costs.
It sounds even more negligent given they passed that guy a non "clean" laptop in the first place.
Anyway, often the problem is the downtime it takes to reimage the machine - esp if it's an old laptop and nobody has an "up to date" and pristine image.
AFAIK normally nobody cares.
Except in this case. I guess someone cared enough to start a witch hunt and this poor chap got the brunt of it.
Someone screams "child porn" and suddenly it's like a mass shark frenzy with blood in the water.
Writing less lines of code = bust your ego? No wonder you like Java.
:).
I've long suspected that Java makes programmers feel like they've done a lot of work at the end of the day (and so deserve their pay), because Java just is a lot more work
I'm a perl guy. I try to use stuff in CPAN all the time, and so I end up writing less code.
It's not so much to do with my ego (I'm not one of those - "everything in a single line of perl" coders). Basically I'm not a great coder, so it's best for me to reuse code from better programmers. The less code I write the fewer bugs I make, and are responsible for.
That said I've had to work around or fix bugs in some CPAN libs. But they tend to have fewer bugs - after all many have been used and tested by others years before I even started writing my programs.
That's part of changing the English language.
Everyone gets a say in it. And part of that "say" is saying someone else is misusing it.
Sometimes I think it's just gay. Damn straight gay it is.
That said, it helps to keep trying to maintain enough standard meanings otherwise we might as well be speaking completely different languages, and never get anything done (Tower of Babel).
Disclaimer: I am not a qualified Software Engineer or Civil Engineer.
There are bigger differences between software engineering and civil engineering that most people don't get.
When the time comes to actually build the design:
Civil Engineering:
Lots of workers, machines, vehicles, cranes start doing stuff.
Software Engineering:
Someone types "make all" and goes for coffee and posts on slashdot.
Believe me this is a fundamental difference, this is the truth. Many people don't get it even when I tell them the truth.
With Civil Engineering a design could cost 10 million and take 1 year, but the actual building could cost 1000 million to build and take 5 years, so the total cost is 110 million and 3 years. Thus if during the design phase the designers say we need two months and 1 million bucks extra to get things right, it is likely that management will say OK.
With Software Engineering a design could cost 1 million and take 1 year, and the actual cost of building it could be 50 cents and take 1 hour. So the design phase takes the bulk of the cost and time. Thus if during the design phase the designers say they need extra time and 100K extra, it's likely that normal management will be very very unhappy.
With Software Engineering the first draft blueprints and plastic/clay models compile and "kind of" run, and Management typically sells those as version 1.0 and 2.0 for the same price as the real product would be (after all it costs about the same to make a first draft as it does to make a final version, if not more).
With Civil Engineering - good luck selling the plastic models to the customer for the same price.
As for software design principles-
1) Figure out what must be done, and what Marketing et all might want later on
2) Figure out what data fields/structures are needed or DB schema is required that will be able to handle 1) and more. Software is often about processing data. If you have no proper place for the data, it makes things a lot harder (and crappier).
3) Create good protocols that can be extended and built on.
4) Insist on stuff being robust and secure - it's not paranoia - high chance of weird data appearing in your module's inputs.
Once you have a huge system, trying to shove an incompatible protocol or schema onto it is not going to be easy. Look at the move of IPv4 to IPv6 - it sure isn't going at a quick pace.
As I've said before- He may be an asshole, but he is an asshole with taste. So it's funny that some recent stupid article was talking about emulating Steve Job's style of management. You can emulate assholeness but how'd you emulate taste :).
;).
People know he is right when he yells at them that the curve of the product is wrong, or that that's not insanely great and they can do better. Of course when they finally get it right, Steve Jobs tells them it's insanely great.
I suspect the workers are in something like one of those abusive relationships.
And the reality distortion field works on salaries too
Steve Wozniak was fairly tolerant of Jobs, so perhaps Apple looks for workers similar to Woz - money not a big deal, can put up with Jobs etc.
I have news for you, we're already living in a world where that "baby" is already _dead_. So if we throw it out with the bathwater, I don't see how things would get worse.
1) There are so many examples of cases where people/companies/organisations/countries kept secrets, but complex stuff was still reverse engineered or reinvented independently.
2) People are using patents to hold monopolies for very long periods (as technology changes slightly, they patent a variation and so on), and for anticompetitive tactics.
The fact is most (ignoring the perpetual motion junk) patents are keeping competitors out far longer than it would take to reverse engineer or reinvent them if they were kept secret.
So where's that benefit you cite?
Give me an example of something currently patented that when kept secret could not be reverse engineered within 10 years once it gets to market (national security stuff doesn't get patented, or is effectively kept secret anyway once the Gov finds out).
It currently takes very many years to get a patent (in various countries too). So shortening the terms isn't going to be useful - you might as well throw the system away.
That's actually why I think patents aren't very useful.
If someone is really innovative even 30 years of monopoly isn't enough to help them - since most people won't get it.
But 30 years of monopoly would be terrible for > 99.99% of the approved patents (which are mostly pretty obvious - e.g. once you encounter the problem, the solution is easily found by anyone competent in the field).
The real innovators are so many steps ahead - they'll think of various problem, then the solutions, and then the problems with the solutions, and then the solutions for those problems, and so on, till they are decades ahead of everyone else.
As for those who say you should actually implement stuff to be able to claim a patent, I give the example of Douglas Engelbart and his team - they actually implemented a lot of stuff, and most people didn't get it till many decades later.
So to me I don't really think there should be patents on inventions - nowadays > 99.99% of them are just trivial junk that clutter up everything and get in the way of real progress. As is they are a net minus to the world. Giving 20 year monopolies to such "innovators" is a travesty, and allowing them to make a minor change and thus extend the monopoly for even longer is crazy - how does that encourage innovation?
If you want to reward innovators, I'd say we should have Prizes for Innovation that are awarded years after - much like the Nobel Prizes. After 10 or 20 years we should be able to tell whether something is really innovative and important.
Perhaps the application fees could go to a fund used to award the prizes and for administrative costs. Money could also come from other sponsors.
Math is not leading (and I don't think that xkcd strip claims it).
In fact in this case, Rhie found the limit first but just couldn't prove it.
As for proving stuff, Math can help prove some stuff, but not other stuff (it can't normally be used to prove what you had for lunch yesterday).
Well if he can play his desired game NOW in its full "maximum quality" glory, and he typically spends USD400 a night on entertainment, then it could actually help him save money.
Basically he spends USD400, plays computer games for a few nights, and actually ends up with more money than he would otherwise (I actually know someone who did save some money in a similar way). In contrast if it were USD4000 for a vid card, the calculation could be different - he could get bored of the various games and go back to spending more.
The other way it could be an investment is if he's a professional gamer. Higher fps is better for many games.
If the clients want to start watching at arbitrary times rather than wait for the "next slot to come along" you'll probably need to cache "near" the clients anyway.
The benefit from multicast isn't that great in that scenario.
Multicast is good if you have many clients downloading the same thing at the same time.
That's not quite what people are doing with Youtube.
How do ACKs work with multicast?
Or are you talking about streams with no acks?
Nuclear Cat crawling out of the bag? The US seems busy throwing the Nuclear Cat everywhere.
http://www.guardian.co.uk/environment/2006/jan/05/energy.g2
http://www.boston.com/news/nation/washington/articles/2008/03/26/warhead_fuses_mistakenly_sent_to_taiwan/
There's plenty of info on making nukes already in the "wild". So if you give people more details, I'm sure it helps a lot.
RTGs don't provide that much power for the weight (and generate quite a lot of heat). Petrol has about 34 megajoules per litre. 50 litre tank = 1.7 gigajoules.
From the wiki, a 39kg RTG can generate a max of 390W (electricity) and 7.2kW(heat). 390W of electricity after 8 hours will give you 11 Megajoules - the amount of energy in 330ml of petrol (think size of soda can), and that assumes 100% efficiency.
Now if you can also convert all of that 7.2 kilowatts of heat to energy - making a total of about 7.6kW (assuming 100%) then you'll have the equivalent of 6.4 litres of petrol after 8 hours (which is about 1/8th of a full tank).
And that 39kg does not include the weight and size of your supercapacitor bank.
How far can you go on 330ml of petrol? Or 6.4 litres of petrol? Before you need to stop for 8 hours?
If you can get 50% efficiency and somehow not use too much space and weight, then perhaps it's not so bad. On most days I can leave my car stationary for 20 hours and not travel very far.
Uh the main benefit of akamai is that an ISP can for example put youtube mirrors on their own network and not have thousands of their customers streaming youtube through their skinny connections to other ISPs.
It's not so much to do with RTT as it is to do with making more efficient use of bandwidth via caching.
Comb filtering?
;).
I think it's more to do with reality filtering
"You do realize that private businesses have never caused a war (on their own at least) that resulted in deaths of thousands and in worst case millions of persons."
Yeah, they just get governments to start wars for them. Look up United Fruit Company.
Governments that stay in power maintain a monopoly on violence.
If there weren't pesky (or powerful enough) governments around, what do you really think companies like that would do? Look up the East India Company.
Many large companies would be really happy if there were no powers higher than them that could say "Hey our citizens don't like that Anglo Persian Oil Company is taking more than 80% of net profits from our country's oil".
Once there's more than a few hundred people you cannot escape from a government. Whether it's a democratically elected one, or or one where a Company is the Government, or a Dictatorship you'll end up with a government one way or another.
"Trains should be 3 feet high and passengers loaded like peas in a pod. "
I doubt that's a good idea, it's unlikely that you'd manage to convince people to put up with that AND the near requirement for passengers to wear diapers or something similar.
If a trip takes more than a few hours and there are many passengers on the train the probability of someone needing to go to the loo is going to be very high.
"Tall" trains can already go 250-300kph. How much faster will such midget trains be after sacrificing the ability for passengers to easily walk to the toilet or elsewhere while the train is moving?
Thing is the obedient/compliant kids are the ones you don't really have to worry about so much.
;).
They're less likely to fight with each other just because "it's my turn", or "he's being a griefer".
But yeah parents should actually brainwash/domesticate/bring up their kids before MTV/Hollywood/McD/etc do it instead.
It's not going to be as easy as obedience training for dogs, since humans didn't quite have the same breeding program most breeds of dogs had