Will that "police" handle override the safety feature that makes it stop when someone's standing in the road? (Well, it's a safety feature when it stops for a toddler -- it's a non-safety feature when it stops for a 14 year old crack addict with a gun...)
Yes, I know that for the typical cost of a mass transit system, you could have bought every regular rider at least one new luxury car. But:
1) How was the cost of building the roads factored in? Not fair to compare the cost of a new railbed to just stuffing more traffic into existing roads... I'm not sure about this, but it would seem like a mile of dual-track railbed built only heavy enough for car-like vehicles should cost less than a mile of four-lane road, and (with central control) would handle more traffic.
2) I'd think the cars for one of these systems would actually be a little simpler than an automobile built for independent operation. Electric motor instead of that horribly complex gasoline engine and transmission. Simpler suspension, because you can count on the rails meeting certain standards for smoothness, and no steering. Lots of electronics, but that's cheap nowadays. So if the cars were built in sufficient quantity, they'd cost the same or less as autos.
Of course, the trouble is that autos are built by the millions, but trams are custom-built. Could you design the trams to use an existing car body, just drop an electric motor under the hood, leave out the steering, and change the wheels?
If you've still got a pasenger train system, you're doing better than the USA. We've got a gov't corporation (Amtrak) running the ghost of the train system, and a number of disconnected local mass transit systems, which are mostly tax-subsidized havens for homeless and criminals. There are exceptions, e.g. when we lived in Virginia, I'd take the family to the Smithsonian on weekends by driving about halfway (approximately to the Beltway), then parking and taking the subway. These subways were clean, felt safe, and were a pretty nice ride on weekends, but the _only_ place it went we were interested in was the Smithsonian-capitol area, and one big underground shopping mall. And the rails didn't go far enough out into the countryside -- if you wanted to catch a subway on workdays at 8am say, the last 10 miles by road would take 2 hours, then all the parking would be gone at the station... The subway took a little pressure off the roads inside the beltway, but really seemed more like a way for the politicians to show off "were doing _something_" than practical transportation to work.
Even the freight railways seem to be in pretty poor shape in the US. Many of the old railroad companies over here are doing pretty well, though -- for 50 years they took the profits from freight rail service, minimizing maintenance of railbed and rolling stock, and bought other businesses not related to the rails at all. I'm not sure if this was just the short-sighted greed that commonly afflicts corporations ("we can look great this quarter if we destroy our main business forever"), or a perception that railroads were going the way of buggy-whip manufacturing so they'd better find another business while it was still possible to milk the rails for money. Anyway, they're milked dry now...
SOMEHOW, we managed to get our families around without SUVs for years. Cars used to be bigger, especially on the inside. The CAFE mileage limits shrunk the outside. The safety standards require thicker sides, and thicker padding on everything. All the pollution stuff in the engine compartment plus the crush zones require more length in the hood. Result: much less room for the passengers. The backseat of a new Buick gives hardly more room than the backseat of a 1968 VW Beatle, and the bug was a sub-compact!
The other thing, of course, is that the baby boomers are getting older. That means wider waists, back problems, bigger rears, arthritic knees, and more fat everywhere. (This of course doesn't apply to me. 8-) I might be 48 and 55 pounds heavier than I was at 20, but I'm just well-padded, and that knee has been bad since high school. It just hurts more when I try to fold my legs into a little car, lately. Cars must be getting smaller...)
Long before the term "SUV" existed, I noticed the same about the drivers of Cadillacs and Lincolns. In those days, these were gigantic luxury sedans, and the average driver of them was either about 90 years old, had never bothered to learn how to manuever that land-barge, or just plain didn't give a damn about anyone else. I rather suspect most of them fell into the last category (a*holes), whether or not they were also senile and ignorant...
So what happened? The EPA and mileage standards forced Cadillac and Lincoln to shrink their cars. Sedans are now too small to contain the a*holes' egos, so they buy SUV's, and drive them just as badly as they drove the big sedans. Maybe worse, because these things really are _trucks_ and ought to be driven somewhat differently than a car.
Not that all SUV drivers are like that, by any means. I often drive a 4x4 pickup, which is about the same size as an SUV, but cheaper and a bit more useful for hauling anything but people. (And I do use the 4wd -- I live on a dirt road which was a foot deep in snow this morning, and turns to deep mud in the spring.) I would have bought an SUV instead if I'd realized how soon I was going to have 5 grandchildren -- the Ford Escort "station wagon" really isn't big enough, and neither is the Dodge Dakota king cab. I think I'm a pretty safe driver in my truck, because I don't forget it _is_ a truck, take it easier on the turns, etc., and I'd be equally safe in an SUV.
The a*holes are not a good reason to have the government hammer all SUV owners. The a*holes would still take over the roads in some other sort of oversized ego-haulers. What's needed in the first place is traffic law enforcement. In the second place, drop the CAFE standards, and let the car companies bring back the big sedans for those who want them.
One more suggestion: how about making traffic tickets proportional to the weight of the vehicle? This won't affect good drivers of any size vehicle, but it will get a little expensive for bad drivers in 8500 pound vehicles. And bad drivers are certainly more dangerous to others when they're in bigger vehicles.
We used to own both beta and VHS players, and I wonder if that box of beta tapes is still in the back of the closet? When we bought the players and the first tapes, everyone knew that there were two competing technologies, and that one or the other would win out in the end, and eventually we'd lose the ability to view those tapes. Actually, most of the betas were bought extremely discounted (under $5), after beta players started disappearing from American stores -- but we were able to watch them for 3 to 5 years until the player broke beyond repair, so it was worth it. Then we could either buy the movie again in VHS (not at full price either, because now it was old), or let it go. I think mostly we just let it go -- hollywood has produced very few movies worth going back and watching again 10 years later. (Casablanca is the only one in DVD that I know of yet. nothing yet. Star Wars also gets heavily played but that's because the children like it and the Episode 4 opening was my first date with my wife. Oh, and it would be too obvious if I just kept rewinding the Carrie Fisher as slave girl scenes in ROTJ. 8-)
You do have a point there, but... The Beta/VHS situation sucked, but at least it was out in the open, and it took a lot longer than the Windows upgrade cycle before the losing technology disappeared. I wonder what percentage of Americans understand that the copy-protected CD's are only going to be good for a limited time. And finally, most people want to see new movies but play the same damned music over and over.
An even better use for the tethered hydraulic suit: firefighting. You power the suit hydraulics with high-pressure water from the pumper truck, in a once-through cycle. That is, you plug the suit into a firehose, the water drives hydraulic cylinders in the suit, then sprays out and helps keep you cool.
The first problem this would solve is that firehose nozzles (called "monitors") can generate more reaction force than a man can handle. Now the hose man is wearing a big, heavy exoskeleton; he clunks into position, then locks the suit and switches the water into the nozzle.
Second of course is the risks run going into burning buildings. With an insulated, armored suit powered by an armored water hose, you can safely walk through any fire that isn't hot enough to endanger the hose. If you don't like the situation around the available door and window entries, you can punch and kick and create a new doorway somewhere safer. The armor gives you some protection against collapsing buildings -- it wouldn't have saved them at the WTC of course, but if a normal house collapses with firemen inside now, they're probably dead, with this they're probably alive. (Not able to dig themselves out because the hose is caught, but alive, and maybe shaking things around enough to make themselves real easy to find.)
And finally, if you have to dig through a collapsed building, you can pick up much bigger pieces and toss them farther. And all the other possible uses of excess strength in rescues, wherever the hoses will reach...
These suits would be rather expensive, with the stainless steel hydraulics and all, but it seems like they would let a fire crew get into a burning building and check for survivors faster, and at reduced risk. And in some cases there would be off-setting cost savings, e.g. if one man in a suit can handle a monitor that needed three men, then you can make some staff cuts.
Portable fusion is extremely unlikely. OTOH, if you are looking at combat applications (like Starship Trooper the book, not the movie), I cab't think of anything else that would give the power needed for more than a few minutes of combat in an exoskeleton. Batteries certainly won't do -- aside from the recharging cycle, the power/weight ratio isn't high enough.
Where it's OK to be tethered (like loading cargo, construction sites, etc.), the power unit could be a big hydraulic pump & pressure tank, connected to the user by hoses. That is, wherever the exoskeleton arm needs a "muscle", you put a cylinder which is activated by letting high-pressure hydraulic fluid in and out. This gives very high power density at the user end, although it's coming from a fixed unit that outweighs him and his suit. Or a centaur unit (exoskeloton arms on the front of a truck, for example) could carry the hydraulic pump unit; this would be good for cargo handling and bomb-loading, but I don't see much use for it in combat.
The best currently-conceivable portable power technology for combat exoskeletons is to burn fuel in cylinders, which are linked in just like hydraulic cylinders. Muscles and hydraulics let you adjust the speed and force even in mid-stroke, while with fuel-cylinders it would seem that your only control is how much fuel is squirted in, when it's lit, and when you vent the exhaust gas and let the cylinder contract. So I'm not sure how much fine control you can manage with that -- but with practice, I think a man could learn to walk and run across country in such a suit, and carry 1,000 pounds of gear or so. Even if the suit is too twitchy to allow aiming a gun with the exoskeleton, you can thus have one guy in the squad carrying artillery for the others to use. But how long would a load of fuel last?
This is how the military works. The poor dumb f*s load the bombs while the techs, pilots, and other valuable people go in the bunker or run errands well out of range... 8-)
Seriously, just because someone isn't a hacker doesn't make him stupid. The USAF trains stupid guys as cooks or personnel clerks, not as bomb loaders. (This is why our records were always f*d up and we'd rather eat at the Navy mess hall if possible...) It's much safer to have someone used to manual labor doing the loading than some geek -- even if the AF managed to put some muscles on the geek. Manual laborers do develop a pretty good instinctive understanding of forces and balance; they can't calculate it, but they do know how far they can lean over while holding a 100 pound bomb. And if the bomb is big enough that manual lifting isn't going to do, then (at present) you've got these same guys driving forklifts or something. It's much safer to have them running a rig that amplifies their muscle power so they can use their experience in hand-loading, than running a fork truck with a half-dozen control levers that do _not_ work intuitively.
More to the point, if you're hired to write a _new_ program to solve a _new_ problem, you aren't going to find an existing program you can copy, and you won't know how to write a new one. Employers are very happy if you can get the job done fast and cheap by finding a solution in existing code, but they hire _programmers_ because they expect to need programs that haven't been written before.
It would seem like this program, or any conceivable cheat detector other than a bug in the student's laptop, would be ineffective on "hello world" programs, because after you eliminate white space, variables, and simple re-arranging, there's only one reasonable algorithm.
So my question is, how in heck can you teach "basketweaving" majors enough programming to where they can write programs that the cheat detector will work on? IIRC, it took half the semester for my CS101 to get a class full of nerds up to the point where we could write programs complex enough to have 20 different variations on the algorithm. (Of course, things might move a little faster now that they don't have to type FORTRAN into punch cards and wait a day for it to be run. But still, business, communications, and PE majors???)
"Forced" isn't the right word -- but it sounds like pre-internet there were very few alternatives for someone who wanted to make a living performing music, except for about five companies all offering pretty much the same contract. This really sounds like a case for anti-trust enforcement, but (a) you knee-jerk libertarians don't believe in that either, and it never seems to be actually effective in cases like this, where several companies just imitate each other's policies. They don't actually collude, but you can't find a different deal anywhere in the market. True, in the long run cartels like this will crack, but people have to find work and buy their necessities _now_, not wait for someone else to enter the market with a different deal.
To take one prime example where America has left us almost choiceless: who did you vote for in 2000, Gore because he planned to take away economic freedom, or Bush because he planned to take away other freedoms? I voted for Harry Browne -- but as far as affecting the leadership of the US goes, that is as effective as not voting at all.
It is true that slavery existed when the original articles of the Constitution were written but nowhere does it actually use the phrase slaves, Indians, or Native Americans. It mentions that other people should be counted as 3/5ths for the purpose of apportioning congressional representatives on the basis of population. This did not condone or condemn slavery.
The constitution also stated that Congress could not ban the importation of slaves until 1808. This looks to me like condoning slavery -- but hoping it would go away in a few generations. The position of slavery in the Constitution is the result of compromises recognizing the impossibility of righting long-standing wrongs all at once. Many southerners such as Jefferson were for ending slavery in principle, but in practice they couldn't see a way to do it in less than 50 years without the whole south going bankrupt. (It wasn't just the south: For instance, New York state passed a law freeing slaves one birth-year at a time. This didn't free Sojourner Truth until the late 1820's, 50 years after the Revolution started. But AFAIK no large property holders went bankrupt because of freeing their slaves...)
So the founding fathers were uncomfortable about slavery, but not sufficiently so as to free the slaves at once and bankrupt the slaveowners. (Tax receipts were nowhere near enough to allow the government to buy them out.) Nor would they risk a rupture with the more adamant of the southern slaveholders by cutting off importation immediately and making it clear that slavery could only continue for a limited time -- a southern secession at that point (or mere failure to ratify the Constitution and give the central government enough tax revenues to maintain an army & navy) could have wound up with _everyone_ being enslaved by King George or some other European monarch...
Note that in 1776, slavery was legal virtually everywhere in the world; England was the first nation to finally ban all slavery around then (there were very few slaves left to be freed), serfdom (white agricultural slave labor) was common in several European countries, and in Russia serfs could still be sold at auction. It was expected that, just as economic and social factors had ended slavery and serfdom of whites in England and other "advanced" European nations a few centuries before, black slavery was going to wither away naturally as the nation grew. And to help hurry this along, the better class of southerners (Lees, Jeffersons, Washingtons) often freed their slaves in their wills (it's one thing to do something when you choose to, and quite another to be forced to by new laws), and the importation of new slaves was stopped as soon as possible while keeping the southern states in the union. With no new imports of slaves, some people freeing their slaves every year, and a rather low reproductive rate among slaves up to that time, they expected to have less slaves every year after 1808 until owning men became rare enough to be a serious faux pas...
The problem is, the invention of the cotton gin in the US turned cotton from an overly costly specialty fabric to one of the cheapest sources of fiber for cloth. This coincided with inventions in textile milling, which greatly reduced the cost of turning the ginned cotton into clothes. Suddenly people of ordinary incomes could buy different clothes for each day of the week. (Previously the poor basically wore one set of clothes until they rotted off, then begged or traded for cast-offs, and even the moderately rich had fewer clothes than a modern welfare queen...) The demand for cotton skyrocketed. Southern landowners could make real money at last, if they could get the brutally hard labor done in the fields -- but most free men would rather take their chances on the frontier than work the cotton fields. So the south imported a lot more slaves than expected before the 1808 deadline, and provident slaveowners set about encouraging their "servants" to reproduce. Yankees were horrified and moralized incessantly, southerners invented all sorts of specious arguments to defend themselves, and so it drifted on until the southern US and Russia were the only places remaining where European Christians were so backwards as to own other men... And the southerners had so convinced themselves they were right, that they were ready to kill over it.
How do you get an "accurate to one millimeter" measurement to an object covered with dust grains, plus pebbles, rocks, boulders, mountains and craters? Doesn't your value change depending on whether you measure to the top of that 1 mm sand particle. or to the rock it's on?
OK, they say they are trying to measure the center to center distance, but they don't get that directly. The real measurement is from a telescope mount on top of a mountain on Earth, to a retro-reflector on the Moon. Do you actually know the height of that mountain at the observatory to 1 mm? And can you correct that height to the day the measurement was taken? (Some sorts of subsoil will shrink and swell depending on water contact, sometimes resulting in the ground rising and falling a few feet annually. I'd think that deep down in a mountain would be rock so it wouldn't do that, but in most cases the whole mountain is either rising or falling by at least millimeters a year, and if there is any soil cover weather changes might change the height by a few millimeters.)
And on the moon, you are measuring to a reflector which is basically laying where the astronauts dropped it 30 years ago. How would the distance from the reflector to the center of the moon be measured? Laser beams & navigational gear in satellites orbiting the moon? What satellites?
Tenure, etc., are definitely supposed to be based primarily on publishing important, well-accepted research (except in the smaller institutions that actually expect some good teaching too). That's the theory anyhow, with the practice never quite coming up to it -- a**kissing has always counted, and handing the Dean a $100K check ought to beat kissing his...
This source code issue doesn't just stand and fall on it's own, it's just a part of the tension between working for corporations that can pay well but don't want too much published and keeping up with the normal requirements of academia. You let them restrict what you can publish, and how can it be properly peer-reviewed? And if your work can't be peer-reviewed, then the U begins to lose the reputation for good research that brought in the top students, top profs, and the grants too. OTOH, there's S^a^t^a^n^ the corporation standing there offering a great big check... It's been 18 years since I've been around a university, so maybe things have got worse.
If the U's can't maintain a good balance, maybe some mandatory requirements _are_ needed. Either the research is 100% privately funded -- including full cost of school staff time and using the labs, offices, and students -- or it's public domain.
Go explain that to your grandmother. You will be able to start a new humor site from her responses... Geeks understand, but Win95 and it's successors sure weren't designed for geeks.
Why didn't they call the button "Menu", because that's what it really is? Because Microsoft never thought beyond the new user's first day!
From the OpenInformatics FAQ:
We are not proposing a specific license; only that whatever license is used provides certain minimal rights to the users of the software:
[1] The Right to View the Source Code: electronic access to the source code, without royalty.
[2] The Right to Redistribution: anyone may redistribute the work.
[3] The Right to Create Derivative Works: anyone may create (and redistribute) a derivative of the work.
[4] The Right of Ownership: the originating author or organization may retain the Copyright for commercial licensing purposes.
This primarily grows from the "open to scrutiny" requirement, although it also goes beyond it a little. [1] places it open to scrutiny, [2] allows wider distribution without burdening the University's bandwidth, [3] allows deeper scrutiny by testing changes in the source code, among other things.
Number [4] on the other hand, allows for researchers, universities, or private sources of partial funding to get the commercial rights. Distribution of source code and the right to run object code are quite separable legally -- although there is a certain practical difficulty in enforcement if a $5,000/seat commercial product can be duplicated by running a free download through gnu c. Just don't call the commercial vendor for support...
So this won't give the University as much immediate financial gain as keeping the source secret might, but it allows some, and a U's two MAIN goals are supposed to be educating students and advancing knowledge; staying solvent is just a means to those ends. Open source (in the generic sense) certainly advances knowledge a lot better than secret source. In the long run, open-sourcing might even speed new developments enough to balance out the reduced payoff per development.
Finally, when the best available starting point for a research project was proprietary code, this arrangement would give the researcher some chance of getting permission to publish the partly-his, partly-theirs code while leaving the ownership with the original proprietor.
I wouldn't go so far as to prohibit researchers starting with proprietary source, but consider this: Profs get their reputations from publishing research results which are thought highly of by knowledgeable colleagues, and tenure, promotions, jobs at better universities, and Nobel prizes all follow. IF most published research comes to include the source code to whatever software was developed to complete the research, then publications which don't include source will look not-so-good. So the profs will have a pretty good reason to avoid using proprietary code given any decent alternative, but the decision is still up to them.
So an illogical user interface is OK because you're used to it? NOT!
In most embedded systems -- things which are made to be used by anyone at all -- push "start" to stop is utterly unacceptable. And I know UML is beginning to be used in designing embedded systems, for the whole machine, not just the software...
[we]write out our documentation in a text editor, and send it to the Rational guy, who puts it in the model. Then we show the pretty diagrams to the customer.
Do the customers ever look at those diagrams and then come back and ask for changes? For instance, suppose a certain design team 8 years ago had shown UML diagrams to an outsider, and explained, following along it: "Now, to shut down, the user moves the mouse to the Start button..." "Wait a minute, you hit START to STOP???" Think they might have changed that?
Out where contracts may actually require programmers to produce software to the customer's satisfaction, one thing like this caught early enough to be easily fixed is worth a hell of a lot of diagrams. Imagine a soda machine where you have to press "SELL" to buy...
a disproportionate amount of the examples and diagrams involve physical systems instead of software systems. It's as though software design is a bit of an afterthought, which is fine, but the book could have been richer had it focused more on this aspect of UML implementation rather than, for instance, how to use the UML to model a soda machine.
Hey, I've got friends who make their living programming the microcontrollers for soda machines, etc. And I'm sure there are many more people doing this sort of programming ("embedded") than hacking OS's, so have some respect.
Anyway, modeling the whole system is what UML is about. Forget that flipper that drops one can, and the code will take your money and say thank you, but you aren't getting any soda.
Slashdot Mirror
Will that "police" handle override the safety feature that makes it stop when someone's standing in the road? (Well, it's a safety feature when it stops for a toddler -- it's a non-safety feature when it stops for a 14 year old crack addict with a gun...)
Most of that complexity is just electronics -- expensive then, especially when reliability was required, cheap now.
Yes, I know that for the typical cost of a mass transit system, you could have bought every regular rider at least one new luxury car. But:
1) How was the cost of building the roads factored in? Not fair to compare the cost of a new railbed to just stuffing more traffic into existing roads... I'm not sure about this, but it would seem like a mile of dual-track railbed built only heavy enough for car-like vehicles should cost less than a mile of four-lane road, and (with central control) would handle more traffic.
2) I'd think the cars for one of these systems would actually be a little simpler than an automobile built for independent operation. Electric motor instead of that horribly complex gasoline engine and transmission. Simpler suspension, because you can count on the rails meeting certain standards for smoothness, and no steering. Lots of electronics, but that's cheap nowadays. So if the cars were built in sufficient quantity, they'd cost the same or less as autos.
Of course, the trouble is that autos are built by the millions, but trams are custom-built. Could you design the trams to use an existing car body, just drop an electric motor under the hood, leave out the steering, and change the wheels?
If you've still got a pasenger train system, you're doing better than the USA. We've got a gov't corporation (Amtrak) running the ghost of the train system, and a number of disconnected local mass transit systems, which are mostly tax-subsidized havens for homeless and criminals. There are exceptions, e.g. when we lived in Virginia, I'd take the family to the Smithsonian on weekends by driving about halfway (approximately to the Beltway), then parking and taking the subway. These subways were clean, felt safe, and were a pretty nice ride on weekends, but the _only_ place it went we were interested in was the Smithsonian-capitol area, and one big underground shopping mall. And the rails didn't go far enough out into the countryside -- if you wanted to catch a subway on workdays at 8am say, the last 10 miles by road would take 2 hours, then all the parking would be gone at the station... The subway took a little pressure off the roads inside the beltway, but really seemed more like a way for the politicians to show off "were doing _something_" than practical transportation to work.
Even the freight railways seem to be in pretty poor shape in the US. Many of the old railroad companies over here are doing pretty well, though -- for 50 years they took the profits from freight rail service, minimizing maintenance of railbed and rolling stock, and bought other businesses not related to the rails at all. I'm not sure if this was just the short-sighted greed that commonly afflicts corporations ("we can look great this quarter if we destroy our main business forever"), or a perception that railroads were going the way of buggy-whip manufacturing so they'd better find another business while it was still possible to milk the rails for money. Anyway, they're milked dry now...
SOMEHOW, we managed to get our families around without SUVs for years. Cars used to be bigger, especially on the inside. The CAFE mileage limits shrunk the outside. The safety standards require thicker sides, and thicker padding on everything. All the pollution stuff in the engine compartment plus the crush zones require more length in the hood. Result: much less room for the passengers. The backseat of a new Buick gives hardly more room than the backseat of a 1968 VW Beatle, and the bug was a sub-compact!
The other thing, of course, is that the baby boomers are getting older. That means wider waists, back problems, bigger rears, arthritic knees, and more fat everywhere. (This of course doesn't apply to me. 8-) I might be 48 and 55 pounds heavier than I was at 20, but I'm just well-padded, and that knee has been bad since high school. It just hurts more when I try to fold my legs into a little car, lately. Cars must be getting smaller...)
#4 and 6 = SUV's seem to attract bad drivers.
Long before the term "SUV" existed, I noticed the same about the drivers of Cadillacs and Lincolns. In those days, these were gigantic luxury sedans, and the average driver of them was either about 90 years old, had never bothered to learn how to manuever that land-barge, or just plain didn't give a damn about anyone else. I rather suspect most of them fell into the last category (a*holes), whether or not they were also senile and ignorant...
So what happened? The EPA and mileage standards forced Cadillac and Lincoln to shrink their cars. Sedans are now too small to contain the a*holes' egos, so they buy SUV's, and drive them just as badly as they drove the big sedans. Maybe worse, because these things really are _trucks_ and ought to be driven somewhat differently than a car.
Not that all SUV drivers are like that, by any means. I often drive a 4x4 pickup, which is about the same size as an SUV, but cheaper and a bit more useful for hauling anything but people. (And I do use the 4wd -- I live on a dirt road which was a foot deep in snow this morning, and turns to deep mud in the spring.) I would have bought an SUV instead if I'd realized how soon I was going to have 5 grandchildren -- the Ford Escort "station wagon" really isn't big enough, and neither is the Dodge Dakota king cab. I think I'm a pretty safe driver in my truck, because I don't forget it _is_ a truck, take it easier on the turns, etc., and I'd be equally safe in an SUV.
The a*holes are not a good reason to have the government hammer all SUV owners. The a*holes would still take over the roads in some other sort of oversized ego-haulers. What's needed in the first place is traffic law enforcement. In the second place, drop the CAFE standards, and let the car companies bring back the big sedans for those who want them.
One more suggestion: how about making traffic tickets proportional to the weight of the vehicle? This won't affect good drivers of any size vehicle, but it will get a little expensive for bad drivers in 8500 pound vehicles. And bad drivers are certainly more dangerous to others when they're in bigger vehicles.
We used to own both beta and VHS players, and I wonder if that box of beta tapes is still in the back of the closet? When we bought the players and the first tapes, everyone knew that there were two competing technologies, and that one or the other would win out in the end, and eventually we'd lose the ability to view those tapes. Actually, most of the betas were bought extremely discounted (under $5), after beta players started disappearing from American stores -- but we were able to watch them for 3 to 5 years until the player broke beyond repair, so it was worth it. Then we could either buy the movie again in VHS (not at full price either, because now it was old), or let it go. I think mostly we just let it go -- hollywood has produced very few movies worth going back and watching again 10 years later. (Casablanca is the only one in DVD that I know of yet. nothing yet. Star Wars also gets heavily played but that's because the children like it and the Episode 4 opening was my first date with my wife. Oh, and it would be too obvious if I just kept rewinding the Carrie Fisher as slave girl scenes in ROTJ. 8-)
You do have a point there, but... The Beta/VHS situation sucked, but at least it was out in the open, and it took a lot longer than the Windows upgrade cycle before the losing technology disappeared. I wonder what percentage of Americans understand that the copy-protected CD's are only going to be good for a limited time. And finally, most people want to see new movies but play the same damned music over and over.
When the first 8-track tapes were first released, there weren't people already working on making the next version of the tape player incompatible.
Planned obsolescence in IP is obscene.
When the first 8-track tapes were first released, there weren't people already working on making the next version of the tape player obsolete.
An even better use for the tethered hydraulic suit: firefighting. You power the suit hydraulics with high-pressure water from the pumper truck, in a once-through cycle. That is, you plug the suit into a firehose, the water drives hydraulic cylinders in the suit, then sprays out and helps keep you cool.
The first problem this would solve is that firehose nozzles (called "monitors") can generate more reaction force than a man can handle. Now the hose man is wearing a big, heavy exoskeleton; he clunks into position, then locks the suit and switches the water into the nozzle.
Second of course is the risks run going into burning buildings. With an insulated, armored suit powered by an armored water hose, you can safely walk through any fire that isn't hot enough to endanger the hose. If you don't like the situation around the available door and window entries, you can punch and kick and create a new doorway somewhere safer. The armor gives you some protection against collapsing buildings -- it wouldn't have saved them at the WTC of course, but if a normal house collapses with firemen inside now, they're probably dead, with this they're probably alive. (Not able to dig themselves out because the hose is caught, but alive, and maybe shaking things around enough to make themselves real easy to find.)
And finally, if you have to dig through a collapsed building, you can pick up much bigger pieces and toss them farther. And all the other possible uses of excess strength in rescues, wherever the hoses will reach...
These suits would be rather expensive, with the stainless steel hydraulics and all, but it seems like they would let a fire crew get into a burning building and check for survivors faster, and at reduced risk. And in some cases there would be off-setting cost savings, e.g. if one man in a suit can handle a monitor that needed three men, then you can make some staff cuts.
Ever wonder why all FBI agents are "special" agents?
Portable fusion is extremely unlikely. OTOH, if you are looking at combat applications (like Starship Trooper the book, not the movie), I cab't think of anything else that would give the power needed for more than a few minutes of combat in an exoskeleton. Batteries certainly won't do -- aside from the recharging cycle, the power/weight ratio isn't high enough.
Where it's OK to be tethered (like loading cargo, construction sites, etc.), the power unit could be a big hydraulic pump & pressure tank, connected to the user by hoses. That is, wherever the exoskeleton arm needs a "muscle", you put a cylinder which is activated by letting high-pressure hydraulic fluid in and out. This gives very high power density at the user end, although it's coming from a fixed unit that outweighs him and his suit. Or a centaur unit (exoskeloton arms on the front of a truck, for example) could carry the hydraulic pump unit; this would be good for cargo handling and bomb-loading, but I don't see much use for it in combat.
The best currently-conceivable portable power technology for combat exoskeletons is to burn fuel in cylinders, which are linked in just like hydraulic cylinders. Muscles and hydraulics let you adjust the speed and force even in mid-stroke, while with fuel-cylinders it would seem that your only control is how much fuel is squirted in, when it's lit, and when you vent the exhaust gas and let the cylinder contract. So I'm not sure how much fine control you can manage with that -- but with practice, I think a man could learn to walk and run across country in such a suit, and carry 1,000 pounds of gear or so. Even if the suit is too twitchy to allow aiming a gun with the exoskeleton, you can thus have one guy in the squad carrying artillery for the others to use. But how long would a load of fuel last?
This is how the military works. The poor dumb f*s load the bombs while the techs, pilots, and other valuable people go in the bunker or run errands well out of range... 8-)
Seriously, just because someone isn't a hacker doesn't make him stupid. The USAF trains stupid guys as cooks or personnel clerks, not as bomb loaders. (This is why our records were always f*d up and we'd rather eat at the Navy mess hall if possible...) It's much safer to have someone used to manual labor doing the loading than some geek -- even if the AF managed to put some muscles on the geek. Manual laborers do develop a pretty good instinctive understanding of forces and balance; they can't calculate it, but they do know how far they can lean over while holding a 100 pound bomb. And if the bomb is big enough that manual lifting isn't going to do, then (at present) you've got these same guys driving forklifts or something. It's much safer to have them running a rig that amplifies their muscle power so they can use their experience in hand-loading, than running a fork truck with a half-dozen control levers that do _not_ work intuitively.
More to the point, if you're hired to write a _new_ program to solve a _new_ problem, you aren't going to find an existing program you can copy, and you won't know how to write a new one. Employers are very happy if you can get the job done fast and cheap by finding a solution in existing code, but they hire _programmers_ because they expect to need programs that haven't been written before.
It would seem like this program, or any conceivable cheat detector other than a bug in the student's laptop, would be ineffective on "hello world" programs, because after you eliminate white space, variables, and simple re-arranging, there's only one reasonable algorithm.
So my question is, how in heck can you teach "basketweaving" majors enough programming to where they can write programs that the cheat detector will work on? IIRC, it took half the semester for my CS101 to get a class full of nerds up to the point where we could write programs complex enough to have 20 different variations on the algorithm. (Of course, things might move a little faster now that they don't have to type FORTRAN into punch cards and wait a day for it to be run. But still, business, communications, and PE majors???)
"Forced" isn't the right word -- but it sounds like pre-internet there were very few alternatives for someone who wanted to make a living performing music, except for about five companies all offering pretty much the same contract. This really sounds like a case for anti-trust enforcement, but (a) you knee-jerk libertarians don't believe in that either, and it never seems to be actually effective in cases like this, where several companies just imitate each other's policies. They don't actually collude, but you can't find a different deal anywhere in the market. True, in the long run cartels like this will crack, but people have to find work and buy their necessities _now_, not wait for someone else to enter the market with a different deal.
To take one prime example where America has left us almost choiceless: who did you vote for in 2000, Gore because he planned to take away economic freedom, or Bush because he planned to take away other freedoms? I voted for Harry Browne -- but as far as affecting the leadership of the US goes, that is as effective as not voting at all.
It is true that slavery existed when the original articles of the Constitution were written but nowhere does it actually use the phrase slaves, Indians, or Native Americans. It mentions that other people should be counted as 3/5ths for the purpose of apportioning congressional representatives on the basis of population. This did not condone or condemn slavery.
The constitution also stated that Congress could not ban the importation of slaves until 1808. This looks to me like condoning slavery -- but hoping it would go away in a few generations. The position of slavery in the Constitution is the result of compromises recognizing the impossibility of righting long-standing wrongs all at once. Many southerners such as Jefferson were for ending slavery in principle, but in practice they couldn't see a way to do it in less than 50 years without the whole south going bankrupt. (It wasn't just the south: For instance, New York state passed a law freeing slaves one birth-year at a time. This didn't free Sojourner Truth until the late 1820's, 50 years after the Revolution started. But AFAIK no large property holders went bankrupt because of freeing their slaves...)
So the founding fathers were uncomfortable about slavery, but not sufficiently so as to free the slaves at once and bankrupt the slaveowners. (Tax receipts were nowhere near enough to allow the government to buy them out.) Nor would they risk a rupture with the more adamant of the southern slaveholders by cutting off importation immediately and making it clear that slavery could only continue for a limited time -- a southern secession at that point (or mere failure to ratify the Constitution and give the central government enough tax revenues to maintain an army & navy) could have wound up with _everyone_ being enslaved by King George or some other European monarch...
Note that in 1776, slavery was legal virtually everywhere in the world; England was the first nation to finally ban all slavery around then (there were very few slaves left to be freed), serfdom (white agricultural slave labor) was common in several European countries, and in Russia serfs could still be sold at auction. It was expected that, just as economic and social factors had ended slavery and serfdom of whites in England and other "advanced" European nations a few centuries before, black slavery was going to wither away naturally as the nation grew. And to help hurry this along, the better class of southerners (Lees, Jeffersons, Washingtons) often freed their slaves in their wills (it's one thing to do something when you choose to, and quite another to be forced to by new laws), and the importation of new slaves was stopped as soon as possible while keeping the southern states in the union. With no new imports of slaves, some people freeing their slaves every year, and a rather low reproductive rate among slaves up to that time, they expected to have less slaves every year after 1808 until owning men became rare enough to be a serious faux pas...
The problem is, the invention of the cotton gin in the US turned cotton from an overly costly specialty fabric to one of the cheapest sources of fiber for cloth. This coincided with inventions in textile milling, which greatly reduced the cost of turning the ginned cotton into clothes. Suddenly people of ordinary incomes could buy different clothes for each day of the week. (Previously the poor basically wore one set of clothes until they rotted off, then begged or traded for cast-offs, and even the moderately rich had fewer clothes than a modern welfare queen...) The demand for cotton skyrocketed. Southern landowners could make real money at last, if they could get the brutally hard labor done in the fields -- but most free men would rather take their chances on the frontier than work the cotton fields. So the south imported a lot more slaves than expected before the 1808 deadline, and provident slaveowners set about encouraging their "servants" to reproduce. Yankees were horrified and moralized incessantly, southerners invented all sorts of specious arguments to defend themselves, and so it drifted on until the southern US and Russia were the only places remaining where European Christians were so backwards as to own other men... And the southerners had so convinced themselves they were right, that they were ready to kill over it.
How do you get an "accurate to one millimeter" measurement to an object covered with dust grains, plus pebbles, rocks, boulders, mountains and craters? Doesn't your value change depending on whether you measure to the top of that 1 mm sand particle. or to the rock it's on?
OK, they say they are trying to measure the center to center distance, but they don't get that directly. The real measurement is from a telescope mount on top of a mountain on Earth, to a retro-reflector on the Moon. Do you actually know the height of that mountain at the observatory to 1 mm? And can you correct that height to the day the measurement was taken? (Some sorts of subsoil will shrink and swell depending on water contact, sometimes resulting in the ground rising and falling a few feet annually. I'd think that deep down in a mountain would be rock so it wouldn't do that, but in most cases the whole mountain is either rising or falling by at least millimeters a year, and if there is any soil cover weather changes might change the height by a few millimeters.)
And on the moon, you are measuring to a reflector which is basically laying where the astronauts dropped it 30 years ago. How would the distance from the reflector to the center of the moon be measured? Laser beams & navigational gear in satellites orbiting the moon? What satellites?
Tenure, etc., are definitely supposed to be based primarily on publishing important, well-accepted research (except in the smaller institutions that actually expect some good teaching too). That's the theory anyhow, with the practice never quite coming up to it -- a**kissing has always counted, and handing the Dean a $100K check ought to beat kissing his...
This source code issue doesn't just stand and fall on it's own, it's just a part of the tension between working for corporations that can pay well but don't want too much published and keeping up with the normal requirements of academia. You let them restrict what you can publish, and how can it be properly peer-reviewed? And if your work can't be peer-reviewed, then the U begins to lose the reputation for good research that brought in the top students, top profs, and the grants too. OTOH, there's S^a^t^a^n^ the corporation standing there offering a great big check... It's been 18 years since I've been around a university, so maybe things have got worse.
If the U's can't maintain a good balance, maybe some mandatory requirements _are_ needed. Either the research is 100% privately funded -- including full cost of school staff time and using the labs, offices, and students -- or it's public domain.
Go explain that to your grandmother. You will be able to start a new humor site from her responses... Geeks understand, but Win95 and it's successors sure weren't designed for geeks.
Why didn't they call the button "Menu", because that's what it really is? Because Microsoft never thought beyond the new user's first day!
From the OpenInformatics FAQ:
We are not proposing a specific license; only that whatever license is used provides certain minimal rights to the users of the software:
[1] The Right to View the Source Code: electronic access to the source code, without royalty.
[2] The Right to Redistribution: anyone may redistribute the work.
[3] The Right to Create Derivative Works: anyone may create (and redistribute) a derivative of the work.
[4] The Right of Ownership: the originating author or organization may retain the Copyright for commercial licensing purposes.
This primarily grows from the "open to scrutiny" requirement, although it also goes beyond it a little. [1] places it open to scrutiny, [2] allows wider distribution without burdening the University's bandwidth, [3] allows deeper scrutiny by testing changes in the source code, among other things.
Number [4] on the other hand, allows for researchers, universities, or private sources of partial funding to get the commercial rights. Distribution of source code and the right to run object code are quite separable legally -- although there is a certain practical difficulty in enforcement if a $5,000/seat commercial product can be duplicated by running a free download through gnu c. Just don't call the commercial vendor for support...
So this won't give the University as much immediate financial gain as keeping the source secret might, but it allows some, and a U's two MAIN goals are supposed to be educating students and advancing knowledge; staying solvent is just a means to those ends. Open source (in the generic sense) certainly advances knowledge a lot better than secret source. In the long run, open-sourcing might even speed new developments enough to balance out the reduced payoff per development.
Finally, when the best available starting point for a research project was proprietary code, this arrangement would give the researcher some chance of getting permission to publish the partly-his, partly-theirs code while leaving the ownership with the original proprietor.
I wouldn't go so far as to prohibit researchers starting with proprietary source, but consider this: Profs get their reputations from publishing research results which are thought highly of by knowledgeable colleagues, and tenure, promotions, jobs at better universities, and Nobel prizes all follow. IF most published research comes to include the source code to whatever software was developed to complete the research, then publications which don't include source will look not-so-good. So the profs will have a pretty good reason to avoid using proprietary code given any decent alternative, but the decision is still up to them.
So an illogical user interface is OK because you're used to it? NOT!
In most embedded systems -- things which are made to be used by anyone at all -- push "start" to stop is utterly unacceptable. And I know UML is beginning to be used in designing embedded systems, for the whole machine, not just the software...
[we]write out our documentation in a text editor, and send it to the Rational guy, who puts it in the model. Then we show the pretty diagrams to the customer.
Do the customers ever look at those diagrams and then come back and ask for changes? For instance, suppose a certain design team 8 years ago had shown UML diagrams to an outsider, and explained, following along it: "Now, to shut down, the user moves the mouse to the Start button..." "Wait a minute, you hit START to STOP???" Think they might have changed that?
Out where contracts may actually require programmers to produce software to the customer's satisfaction, one thing like this caught early enough to be easily fixed is worth a hell of a lot of diagrams. Imagine a soda machine where you have to press "SELL" to buy...
a disproportionate amount of the examples and diagrams involve physical systems instead of software systems. It's as though software design is a bit of an afterthought, which is fine, but the book could have been richer had it focused more on this aspect of UML implementation rather than, for instance, how to use the UML to model a soda machine.
Hey, I've got friends who make their living programming the microcontrollers for soda machines, etc. And I'm sure there are many more people doing this sort of programming ("embedded") than hacking OS's, so have some respect.
Anyway, modeling the whole system is what UML is about. Forget that flipper that drops one can, and the code will take your money and say thank you, but you aren't getting any soda.