I think you bring up a good point, I'd just like to counterweight it a bit.
There are very few technical books that are as well written an clear as Art of Electronics. A robotics book might be more "essential" if you are interested in building a robot, but very few of them will extend your understanding and mastery of robots specifically as AoE will extend your understanding of electronics in general.
"Practical" doesn't just means "arts-and-crafts." Theory is "practical" too, when you are faced with a problem that requires an original solution. AoE was not, if I recall correctly, written with the EE student in mind. It was written for people like experimental scientists from whom the ability to understand and design circuits would be a great practical advantage.
If you want to build other people's designs, then by all means restrict yourself to building things that other people have designed or snapping together modules, you can get by without really knowing much about how electronics works. I expect that most robot hobbyists only have a rudimentary understanding of electronics theory, and that's fine for them. But it certainly won't hurt to be able to analyze circuits even design them, and if you avoid AoE because it is not "practical", you're cheating yourself out of one of the best tech books ever written.
Actually, a book like that falls under other forms of IP. It can contain trade secrets, for example.
It also falls under common law copyright, because it hasn't been published. It could be the equivalent of your personal papers.
Re:Retaining control of the usage
on
Who Owns Software?
·
· Score: 2, Interesting
A few years ago I rented an old vacation cottage, and found there were a number of old hardcover books in it dating from around the time it was built. One of the books, printed in 1903 had something on the front end-paper that was very familiar.
It had a EULA.
The first sale doctrine was recognized by the Supreme Court five years after the book was printed.
The thing is that for whatever reason, the law has chosen to take a more complicated, nuanced/inconsistent stance on software EULAs. Be that as it may, it's dangerous to draw analogies with books for whatever purposes.
The thing about EULAs, is that they bootstrap a whole new set of rights for the copyright holder using copyright as the starting point. That's the whole point of a EULA; if all it did was protect your copyright, you wouldn't need one. You can't put a EULA in a book that says you can only criticize certain parts of it, or that you can't rent it or lend it to other people. But that is routinely done in software. You can't disassemble a piece of softare you "bought", and in many cases you aren't allowed to publish benchmarks.
Except that he admitted he wasn't really making a point, even though if he had the point would be a good one. And if he had been making a point, the punishment would be reasonable.
The point of civil disobedience is not to avoid being caught. It is to be caught in a way that proves the system is corrupt. Punishment is critical to the effectiveness of civil disobedience as a strategy to change the world.
It's also critical for holding back the tide of unthinking self-righteousness in the world. If good intentions were an absolute defense, there would be no end to the crimes people would commit with complete assurance they are on the side of right.
Giving this guy a slap on the wrist is the right thing to do; it serves the purpose of having the rule without doing more damage than breaking the rule did. The rules are there for the guidance of the wise and the protection of fools. The wise might choose to accept punishment in service to a higher cause; the foolish shouldn't be punished more than is necessary to set them on the right track.
You can violate almost any principle of design and show that for small data sets under lab conditions that principle doesn't contribute much to usability. For the dataset in question, you probably could misalign the columns and rows, then display them in six point type and there wouldn't be much difference in speed or accuracy.
The study gives a baseline result that is entirely credible: for small static datasets and a small number of questions, a non-fatigued subject population manages to get through a short list of relatively easy questions about as well with or without the striping.
It's a valid starting point, but it's not really enough for practitioners to pay attention yet.
I always say: a developer who doesn't hate a platform just doesn't know it well enough.
It follows that programmers that rush to the defense of a platform as being for practical purposes the ultimate one aren't worth listening to on that topic.
Anybody who was overbid by Verizon has a legitimate complaint if Verizon gets a break on the conditions of the auction. Suppose you and I bid on a lease to federal land in an auction and the leases did not include mineral rights. You bid one billion, I bid 1.1. After I won the auction, I and my buddies at the BLM announce that surprise! I'm getting the rights to drill for oil and mine gold on the land. In fact I can sublet those rights for cash money if I like.
You'd feel justifiably pissed. The oil on the land is worth ten billion, and you'd have been willing to bid five. Instead, I get 8.9 billion dollars given to me by my friends in the government.
If we want to put the right to decide which applications run on the spectrum is on the table, it should be bid out separately. Of course this would be unfair to Verizon, and for exactly the same reason that giving Verizon a "gift" of additional rights to Verizon is unfair to the other bidders -- indeed to parties that did not bid at all.
It's unfair in either case because the parties bidding didn't know what the winner was actually going to get.
Actually, the stock's performance is surprisingly GOOD.
The 1990s was the end of the era of PC adoption. I started work in the early 80's, which with the introduction of the microcomputer was the star of that era. Back in the late 80's and early 90's, we never bought computers onesies and twosies, we bought them literally by the truckload to computer-up entire departments at a time. It's been widely observed that while Microsoft was strongly against "software piracy" ideologically, it benefited from a certain level of "piracy" through economic network effects. Worrying about "piracy" was like worrying about the little fish that slipped through the holes in your net, whilst your net was completely full of big fish.
Microsoft was a company that was predicated on exponential growth in demand for its products. In the 80s through mid 90s it was driven by PC adoption, but the thoughtful among us always believed that was not sustainable. In the mid to late 90s the era of exponential adoption was extended for a few years by the dot com bubble.
Where are the exponential growth drivers of the twenty-first century? Well, there aren't any like the 80s-90s, but to the degree they exist they are in consumer markets. Microsoft had never been a consumer company. It never had consumer loyalty. It was a company that sold things to people who make purchase decisions on the behalf of others.
Microsoft's XBox and Zune efforts were, in the culture of Microsoft, bold and appropriate steps. Microsoft has for most of its existence been defined by dramatic, market beating growth. That is not in the cards in its PC software business. So it "had to" go where the growth was. They are strategic products. XBox is the more successful of the two, but arguably Zune is the more strategically important, because it is an attempt by Microsoft to leverage its PC monopoly into becoming a pinch point for digital entertainment providers.
It has a formula for digital entertainment, and it's the good old one that's worked so often for them before: appeal to people who make decisions on the behalf of consumers. In this case it's all about DRM. DRM isn't just an ideological choice, it's a strategic choice for Microsoft. What they offer is control of the platform. They offer some of that control to content oriented companies so those companies can extract more revenue from their customers. Consumers go with Microsoft because they can't get the content they want anywhere else. Like a many strategies, it's reasonable on paper, but real world considerations make it a lot harder than it sounds. Microsoft has to deal with a competitor with lots of vision for the future (Apple) and partners with no vision for the future other than to delay its coming as long as possible (the entertainment industry).
Without taking anything away from Bill Gates brilliance as a businessman, Ballmer had it a lot harder than Gates ever had. Bringing back Gates might improve discipline, or it might not. The company is inherently less focused than it was a decade ago.
What Microsoft really needs is new blood.
There are two choices: either it makes a serious bid to become a dominant player in consumer technology, or it becomes more conservative in how it throws money at grand strategies.
They're both reasonable options. I once heard an investment adviser say he had Procter and Gamble in his portfolio because if people stopped buying soap, most of his other assumptions about the world would probably be wrong as well. A company like P&G is continually creating new products, but nobody expects them to double their size every five years. You manage a company like that to produce profit, and growth is a welcome side effect. For years Microsoft ran things the opposite way: aim for growth and profits will come.
The right leader will take them one or the other path, although he'll face a lot of doubters, because neither of those choices is how Microsoft got where it is today. But bringing back Gates won't turn back the clock twenty years.
Lithium ion phosphate technology is almost as good as Li-ion technology, and considerably safer.
Personally, I'd have not much more concern about driving a plain old Li-ion powered car than I have using a Li-ion laptop. Granted, the worst case scenario in a car is much more destructive of the battery, but it doesn't seem to be beyond the capabilities of engineering to render the risk of Li-ion to be on the same order of danger as gasoline or ethanol. If safety is so important, then we should be talking about Li-ion phosphate or NiMH.
What's holding things back in electric cars and plug-in hybrids are all the patents covering the kinds of things you'd need to do to produce large batteries. It's not so much a question of physical practicality than legal practicality, That's why we haven't seen the next logical step on hybrids: the plug-in hybrid. It's not possible to license the technology to scale the NiMH hydride batteries used in current generation vehicles to a size large enough to make the plug-in idea really work.
We're pretty close to being able to make reasonably versatile electric cars economically, and given the popularity of hybrids the plug-in hybrid is a no-brainer. If we don't see those technologies become practical for widespread use in the next decade, it won't be because the world lacks the engineering talent to do it.
So -- if I read what you are saying correctly, if the user has a good experience with vendor supported systems, it's a bad thing because it is unrepresentative of what he'd experience if he did his own installation onto a non-supported system?
Yes, it is true that trouble is a good teacher, but that doesn't make it a good thing. If you have to patch and compile your own kernel and driver, sure you learn stuff, and that's good. But we can't assume that that's a better use of the user's time than what he'd otherwise put it to.
Steer them to a vendor that provides models with a distro of Linux pre-installed. Period, full stop.
Otherwise, they need to have somebody to resolve the problems that come up because no Linux distro can test on every possible combination of hardware. It's not hard to do for US. It's an insurmountable frustration for the. So unless you are prepared to always be there for them, find a vendor that supports them.
Dell, I know, has a narrow but adequate range of choices with Ubuntu preinstalled. Since they provide Ubuntu support, they hopefully chose Linux friendly hardware, and in the worst case there's somebody to call if you aren't there.
I've found Ubuntu kernel releases sometimes break hardware drivers that work fine under stock Debian kernels.
Just make sure you keep your kernel and initramfs in your menu.lst when the update manager wants to do a kernel update. Personally, I keep a bootable external drive for recovery.
This is a classic YMMV situation; I realize that many people have never had this experience. However, given my personal experience, I tend to think of Ubuntu as a "enthusiast" distro. I wouldn't put it on a Linux novice's machine unless I was willing to provide support as well.
I'm generally happy with Ubuntu, but dealing with driver, kernel and ACPI issues has taken a LOT more time over the last two years than is reasonable. Before that I'd been using SUSE and Mandrake, and found them relatively trouble free. I can't speak to their recent releases though.
well, on the plus side, the intel sound card on my laptop works under hardy. on the minus side, i have had issues with the keyboard driver under gnome 9illustrated by this post0. this same bug with modifier keys also seems to lead to program crashes. running the keyboard prefs app fixes the problem when it crops up.
overall, hardy feels very rough, and the upgrade process is even rougher. the upgrade removes the network manager applet, so i had to configure wi-fi from hand and sudo apt-get install network-manager-gnome.
all the problems i've run into have been reported, with the exception of an acpi related boot issue 9e.g., laptop won't boot when running from battery0. i'm going to check my laptop initramfs before looking into that one. i have various acpi related problems including the inability to resume from suspend that may be bios specific.
i'd recommend looking upon hardy as a 'beta' release. it's promising, but not something you want to install on your non-linux friends' machines yet. i've been using it for some months now starting with the beta release, and it's still seems to be in the fix two bugs, create one new bug phase. if feisty doesn't have any obvious hardware problems on your setup, i'd stick with that for a few months yet. i really missed having sound though. the new sound architecture is interesting and has features that might prove useful to some people.
overall, i'm ok with hardy; i can deal with its oddities. i'm a little happier running under xfce, which also leaves me more room to run virtual machines. bottom line -- a few people will be somewhat happier with hardy, many will be considerably happier with feisty, at least for a few months yet.
Well, there is also the common "stability through crisis" leadership anti-pattern.
If People don't like to "change horses in midstream", so as soon as you've got four hooves on the bank you turn around and plunge right back in.
People instinctively pull together when times are tough or dangerous, so keeping people down and endangered unites them behind your leadership.
Here in America, we pride ourselves on rugged individualism. It's a pity that we don't pride ourselves more on rational individualism. We aren't really all that different from people in Iran and Cuba -- we're as easy as anybody else to be led around by the nose. We're just blessed with a system where it's harder to hold on to power for more than a decade.
In America, our Achilles' heel is our fondness for decisive and bold leadership, without any consideration that effective leadership might be possible. This makes it childishly easy to make us vote against our own self interests. It's hard to make progress on problems, but it's always easy and popular to be harsh, aggressive and cruel. It's like heaven for demagogues; when faced with a problem, just do something brutal and loud, and you are rewarded with admiration, while leaving the problem (a very valuable lever on public opinion) intact.
That's why I hate the myth that effective government is impossible. It's not that good government is easy, or even likely. It's not that I expect ever to see really good government in my lifetime. But the possibility of doing better stands as a rebuke to the failures of the current government. You can't complain about bad government if you don't believe good government is possible.
What we are seeing in Iran is the common bad government defense of provoking a crisis that they think they can manage rather than trying to weather the crisis they know they cannot: domestic dissatisfaction with their leadership. They've got our measure all right. They're provoking us into doing something aggressive but unconscionably ineffectual, counting on our being tied down in Afghanistan and Iraq and having very little goodwill among the countries that ought to be our allies.
With any luck, they'll stay in power with this game as long as Castro did.
I thought the advantage of standards was to reduce divergence in systems. . The more implementations of particular items, such as screws, conform to a standard, such as phillips head, the better it is for the people who use screws.
Actually, the advantage of standards is that it makes it easy to compare prices charged by different vendors. So if you've been buying screws from a certain vendor, and you get a lower bid from Acme Tool and Die. Well, it's a hex head, 12mm, and it comes with a matching bolt. It looks good. How strong is it? How did you measure the strength?
The problem with OOXML is that it doesn't do anything to help you choose between vendors.
There is a point at which you need to take some responsibility for your own safety.
For the obvious reason that if you don't, you end up dead.
I agree paying up would be an honorable thing, but it takes chutzpah under these circumstances to demand payment.
I think a policy of charging for a rescue is reasonable, as long as the policy -- and the cost -- is set in advance. There are many reasonable ways to do this. You can require users to carry insurance; you can provide insurance by user fees; or you can just state in advance that various kinds of rescues will be charged at different rates.
I don't think it is reasonable to set a reimbursement number after the fact. Nobody should be thinking about the budget in this kind of situation, or if anybody should it is not the family of the person being rescued. In practice this will spread costs over a number of users, and some will be "luckier" than others.
It is not necessarily the case that more expensive victims are more negligent than less expensive ones. Maybe if Fosset crashed a half mile away from where he did, the rescue would have been successful and cost 1/10 what it did.
Gibbons is the guy who (a) employed an undocumented nanny for six years and (b) as governor tried to get Mexico to fork over $7m to reimburse Nevada for health care spending on undocumented workers.
Voyager came along at a time when keeping the franchise running creatively would have been a huge challenge in any case. TNG extended the formula from TOS as far as the creative team could. But on top of that they were already running another series in the same franchise: DS9.
You'd think that there is enough writing talent in the world to keep two Star Trek series running in parallel. However the number of writers capable of and interested in writing Star Trek scripts, and who were trusted to do so, might narrow down the possibilities a great deal.
For whatever reason, as time went on the creative well was evidently dry, and it was being shared by two series.
There were a lot of very good episodes in DS9. There were not a few reasonably good Voyager episodes. But towards the end of the DS9 run there were an inordinate number of stinkers, and Voyager was breaking out in lousy scripts as well.
If you imagine that there had been a single series, without the worst of the episodes and wth a few more really good episodes, then you would have had something closer to TNG in its popularity.
If you look at what the suit is supposed to do, and remember that there is supposed to be a person in it that is not screaming in agony, the only plausible explanation is that the suit is equipped with exotic physics technology that manipulates gravity or inertia. That would explain why he doesn't need to carry hundreds of liters of reaction mass around.
The "contrails" it leaves might be something like condensation trails in a could chamber, or they might literally be a kind of "smoke screen".
Nuclear power? Far too crude for Stark; it's the world's most dangerously overcomplicated method of boiling water. You'd need all kinds of fuel mass, not to mention complicated machinery, just to eke out a measly few neutrons.
Stark would use something like... duotronically controlled matter antimatter reaction dirven vectorized propulsion. That'd be just the thing for a rocket boot, if you don't mind the risk that your athlete's foot would mutate into something unexpectedly horrible.
Iron man is shown leaving contrails -- or possibly smoke trails. Either way he's burning something that has to have mass and volume. Where does he keep it?
In the trailer, we see him crash the suit from thousand of feet up, breaking his fall by going through meters of reinforced concrete. Granting that the suit might survive this, its occupant is supposed to be human. Keep in mind that he brain and spinal cord have the approximate consistency of toothpaste.
On a side note, is it just me or do Downey's whoops and hollers sound flat? I realize he wasn't actually flying but his yelling just doesn't seem, to me, to be indicative of someone who's flying in a suit of armor.
Oh, come now, would you expect a genius like Stark would miss that bit?
The suit (apparently) has an excellent communications system, that surely include things like noise cancellation and background noise shaping. When he is suited up and taking a call in his Tony Stark persona, he doesn't want to sound like he's talking into a cheap headset, while standing on the deck of an aircraft carrier with with a cookpot over his head.
Come to think of it, the "flatness" is probably a deliberate attempt by the director to insinuate the notion into your head that you are hearing Stark over a mobile communication system. He probably could have made Downey sound ling Bing Crosby crooning into an RCA-44 mic, if he wanted to. In fact, that might have been an interesting choice.
Slashdot Mirror
I think you bring up a good point, I'd just like to counterweight it a bit.
There are very few technical books that are as well written an clear as Art of Electronics. A robotics book might be more "essential" if you are interested in building a robot, but very few of them will extend your understanding and mastery of robots specifically as AoE will extend your understanding of electronics in general.
"Practical" doesn't just means "arts-and-crafts." Theory is "practical" too, when you are faced with a problem that requires an original solution. AoE was not, if I recall correctly, written with the EE student in mind. It was written for people like experimental scientists from whom the ability to understand and design circuits would be a great practical advantage.
If you want to build other people's designs, then by all means restrict yourself to building things that other people have designed or snapping together modules, you can get by without really knowing much about how electronics works. I expect that most robot hobbyists only have a rudimentary understanding of electronics theory, and that's fine for them. But it certainly won't hurt to be able to analyze circuits even design them, and if you avoid AoE because it is not "practical", you're cheating yourself out of one of the best tech books ever written.
Actually, a book like that falls under other forms of IP. It can contain trade secrets, for example.
It also falls under common law copyright, because it hasn't been published. It could be the equivalent of your personal papers.
A few years ago I rented an old vacation cottage, and found there were a number of old hardcover books in it dating from around the time it was built. One of the books, printed in 1903 had something on the front end-paper that was very familiar.
It had a EULA.
The first sale doctrine was recognized by the Supreme Court five years after the book was printed.
The thing is that for whatever reason, the law has chosen to take a more complicated, nuanced/inconsistent stance on software EULAs. Be that as it may, it's dangerous to draw analogies with books for whatever purposes.
The thing about EULAs, is that they bootstrap a whole new set of rights for the copyright holder using copyright as the starting point. That's the whole point of a EULA; if all it did was protect your copyright, you wouldn't need one. You can't put a EULA in a book that says you can only criticize certain parts of it, or that you can't rent it or lend it to other people. But that is routinely done in software. You can't disassemble a piece of softare you "bought", and in many cases you aren't allowed to publish benchmarks.
Except that he admitted he wasn't really making a point, even though if he had the point would be a good one. And if he had been making a point, the punishment would be reasonable.
The point of civil disobedience is not to avoid being caught. It is to be caught in a way that proves the system is corrupt. Punishment is critical to the effectiveness of civil disobedience as a strategy to change the world.
It's also critical for holding back the tide of unthinking self-righteousness in the world. If good intentions were an absolute defense, there would be no end to the crimes people would commit with complete assurance they are on the side of right.
Giving this guy a slap on the wrist is the right thing to do; it serves the purpose of having the rule without doing more damage than breaking the rule did. The rules are there for the guidance of the wise and the protection of fools. The wise might choose to accept punishment in service to a higher cause; the foolish shouldn't be punished more than is necessary to set them on the right track.
You can violate almost any principle of design and show that for small data sets under lab conditions that principle doesn't contribute much to usability. For the dataset in question, you probably could misalign the columns and rows, then display them in six point type and there wouldn't be much difference in speed or accuracy.
The study gives a baseline result that is entirely credible: for small static datasets and a small number of questions, a non-fatigued subject population manages to get through a short list of relatively easy questions about as well with or without the striping.
It's a valid starting point, but it's not really enough for practitioners to pay attention yet.
Well, I'm not giving up laptops, are you?
Point proven.
I always say: a developer who doesn't hate a platform just doesn't know it well enough.
It follows that programmers that rush to the defense of a platform as being for practical purposes the ultimate one aren't worth listening to on that topic.
I disagree.
Anybody who was overbid by Verizon has a legitimate complaint if Verizon gets a break on the conditions of the auction. Suppose you and I bid on a lease to federal land in an auction and the leases did not include mineral rights. You bid one billion, I bid 1.1. After I won the auction, I and my buddies at the BLM announce that surprise! I'm getting the rights to drill for oil and mine gold on the land. In fact I can sublet those rights for cash money if I like.
You'd feel justifiably pissed. The oil on the land is worth ten billion, and you'd have been willing to bid five. Instead, I get 8.9 billion dollars given to me by my friends in the government.
If we want to put the right to decide which applications run on the spectrum is on the table, it should be bid out separately. Of course this would be unfair to Verizon, and for exactly the same reason that giving Verizon a "gift" of additional rights to Verizon is unfair to the other bidders -- indeed to parties that did not bid at all.
It's unfair in either case because the parties bidding didn't know what the winner was actually going to get.
Actually, the stock's performance is surprisingly GOOD.
The 1990s was the end of the era of PC adoption. I started work in the early 80's, which with the introduction of the microcomputer was the star of that era. Back in the late 80's and early 90's, we never bought computers onesies and twosies, we bought them literally by the truckload to computer-up entire departments at a time. It's been widely observed that while Microsoft was strongly against "software piracy" ideologically, it benefited from a certain level of "piracy" through economic network effects. Worrying about "piracy" was like worrying about the little fish that slipped through the holes in your net, whilst your net was completely full of big fish.
Microsoft was a company that was predicated on exponential growth in demand for its products. In the 80s through mid 90s it was driven by PC adoption, but the thoughtful among us always believed that was not sustainable. In the mid to late 90s the era of exponential adoption was extended for a few years by the dot com bubble.
Where are the exponential growth drivers of the twenty-first century? Well, there aren't any like the 80s-90s, but to the degree they exist they are in consumer markets. Microsoft had never been a consumer company. It never had consumer loyalty. It was a company that sold things to people who make purchase decisions on the behalf of others.
Microsoft's XBox and Zune efforts were, in the culture of Microsoft, bold and appropriate steps. Microsoft has for most of its existence been defined by dramatic, market beating growth. That is not in the cards in its PC software business. So it "had to" go where the growth was. They are strategic products. XBox is the more successful of the two, but arguably Zune is the more strategically important, because it is an attempt by Microsoft to leverage its PC monopoly into becoming a pinch point for digital entertainment providers.
It has a formula for digital entertainment, and it's the good old one that's worked so often for them before: appeal to people who make decisions on the behalf of consumers. In this case it's all about DRM. DRM isn't just an ideological choice, it's a strategic choice for Microsoft. What they offer is control of the platform. They offer some of that control to content oriented companies so those companies can extract more revenue from their customers. Consumers go with Microsoft because they can't get the content they want anywhere else. Like a many strategies, it's reasonable on paper, but real world considerations make it a lot harder than it sounds. Microsoft has to deal with a competitor with lots of vision for the future (Apple) and partners with no vision for the future other than to delay its coming as long as possible (the entertainment industry).
Without taking anything away from Bill Gates brilliance as a businessman, Ballmer had it a lot harder than Gates ever had. Bringing back Gates might improve discipline, or it might not. The company is inherently less focused than it was a decade ago.
What Microsoft really needs is new blood.
There are two choices: either it makes a serious bid to become a dominant player in consumer technology, or it becomes more conservative in how it throws money at grand strategies.
They're both reasonable options. I once heard an investment adviser say he had Procter and Gamble in his portfolio because if people stopped buying soap, most of his other assumptions about the world would probably be wrong as well. A company like P&G is continually creating new products, but nobody expects them to double their size every five years. You manage a company like that to produce profit, and growth is a welcome side effect. For years Microsoft ran things the opposite way: aim for growth and profits will come.
The right leader will take them one or the other path, although he'll face a lot of doubters, because neither of those choices is how Microsoft got where it is today. But bringing back Gates won't turn back the clock twenty years.
Lithium ion phosphate technology is almost as good as Li-ion technology, and considerably safer.
Personally, I'd have not much more concern about driving a plain old Li-ion powered car than I have using a Li-ion laptop. Granted, the worst case scenario in a car is much more destructive of the battery, but it doesn't seem to be beyond the capabilities of engineering to render the risk of Li-ion to be on the same order of danger as gasoline or ethanol. If safety is so important, then we should be talking about Li-ion phosphate or NiMH.
What's holding things back in electric cars and plug-in hybrids are all the patents covering the kinds of things you'd need to do to produce large batteries. It's not so much a question of physical practicality than legal practicality, That's why we haven't seen the next logical step on hybrids: the plug-in hybrid. It's not possible to license the technology to scale the NiMH hydride batteries used in current generation vehicles to a size large enough to make the plug-in idea really work.
We're pretty close to being able to make reasonably versatile electric cars economically, and given the popularity of hybrids the plug-in hybrid is a no-brainer. If we don't see those technologies become practical for widespread use in the next decade, it won't be because the world lacks the engineering talent to do it.
So -- if I read what you are saying correctly, if the user has a good experience with vendor supported systems, it's a bad thing because it is unrepresentative of what he'd experience if he did his own installation onto a non-supported system?
Yes, it is true that trouble is a good teacher, but that doesn't make it a good thing. If you have to patch and compile your own kernel and driver, sure you learn stuff, and that's good. But we can't assume that that's a better use of the user's time than what he'd otherwise put it to.
Steer them to a vendor that provides models with a distro of Linux pre-installed. Period, full stop.
Otherwise, they need to have somebody to resolve the problems that come up because no Linux distro can test on every possible combination of hardware. It's not hard to do for US. It's an insurmountable frustration for the. So unless you are prepared to always be there for them, find a vendor that supports them.
Dell, I know, has a narrow but adequate range of choices with Ubuntu preinstalled. Since they provide Ubuntu support, they hopefully chose Linux friendly hardware, and in the worst case there's somebody to call if you aren't there.
I've found Ubuntu kernel releases sometimes break hardware drivers that work fine under stock Debian kernels.
Just make sure you keep your kernel and initramfs in your menu.lst when the update manager wants to do a kernel update. Personally, I keep a bootable external drive for recovery.
This is a classic YMMV situation; I realize that many people have never had this experience. However, given my personal experience, I tend to think of Ubuntu as a "enthusiast" distro. I wouldn't put it on a Linux novice's machine unless I was willing to provide support as well.
I'm generally happy with Ubuntu, but dealing with driver, kernel and ACPI issues has taken a LOT more time over the last two years than is reasonable. Before that I'd been using SUSE and Mandrake, and found them relatively trouble free. I can't speak to their recent releases though.
well, on the plus side, the intel sound card on my laptop works under hardy. on the minus side, i have had issues with the keyboard driver under gnome 9illustrated by this post0. this same bug with modifier keys also seems to lead to program crashes. running the keyboard prefs app fixes the problem when it crops up.
overall, hardy feels very rough, and the upgrade process is even rougher. the upgrade removes the network manager applet, so i had to configure wi-fi from hand and sudo apt-get install network-manager-gnome.
all the problems i've run into have been reported, with the exception of an acpi related boot issue 9e.g., laptop won't boot when running from battery0. i'm going to check my laptop initramfs before looking into that one. i have various acpi related problems including the inability to resume from suspend that may be bios specific.
i'd recommend looking upon hardy as a 'beta' release. it's promising, but not something you want to install on your non-linux friends' machines yet. i've been using it for some months now starting with the beta release, and it's still seems to be in the fix two bugs, create one new bug phase. if feisty doesn't have any obvious hardware problems on your setup, i'd stick with that for a few months yet. i really missed having sound though. the new sound architecture is interesting and has features that might prove useful to some people.
overall, i'm ok with hardy; i can deal with its oddities. i'm a little happier running under xfce, which also leaves me more room to run virtual machines. bottom line -- a few people will be somewhat happier with hardy, many will be considerably happier with feisty, at least for a few months yet.
Well, there is also the common "stability through crisis" leadership anti-pattern.
If People don't like to "change horses in midstream", so as soon as you've got four hooves on the bank you turn around and plunge right back in.
People instinctively pull together when times are tough or dangerous, so keeping people down and endangered unites them behind your leadership.
Here in America, we pride ourselves on rugged individualism. It's a pity that we don't pride ourselves more on rational individualism. We aren't really all that different from people in Iran and Cuba -- we're as easy as anybody else to be led around by the nose. We're just blessed with a system where it's harder to hold on to power for more than a decade.
In America, our Achilles' heel is our fondness for decisive and bold leadership, without any consideration that effective leadership might be possible. This makes it childishly easy to make us vote against our own self interests. It's hard to make progress on problems, but it's always easy and popular to be harsh, aggressive and cruel. It's like heaven for demagogues; when faced with a problem, just do something brutal and loud, and you are rewarded with admiration, while leaving the problem (a very valuable lever on public opinion) intact.
That's why I hate the myth that effective government is impossible. It's not that good government is easy, or even likely. It's not that I expect ever to see really good government in my lifetime. But the possibility of doing better stands as a rebuke to the failures of the current government. You can't complain about bad government if you don't believe good government is possible.
What we are seeing in Iran is the common bad government defense of provoking a crisis that they think they can manage rather than trying to weather the crisis they know they cannot: domestic dissatisfaction with their leadership. They've got our measure all right. They're provoking us into doing something aggressive but unconscionably ineffectual, counting on our being tied down in Afghanistan and Iraq and having very little goodwill among the countries that ought to be our allies.
With any luck, they'll stay in power with this game as long as Castro did.
It's attractive, but remember to size your wiring accordingly. To deliver the same power as a 120V line, a 12V line needs to carry 10x the current.
It's one thing to run a 12V line to a 12V light fixture. It's another thing to design a general purpose power distribution system for your house.
Well, the pessimistic view might be that it already has...
Actually, the advantage of standards is that it makes it easy to compare prices charged by different vendors. So if you've been buying screws from a certain vendor, and you get a lower bid from Acme Tool and Die. Well, it's a hex head, 12mm, and it comes with a matching bolt. It looks good. How strong is it? How did you measure the strength?
The problem with OOXML is that it doesn't do anything to help you choose between vendors.
For the obvious reason that if you don't, you end up dead.
I agree paying up would be an honorable thing, but it takes chutzpah under these circumstances to demand payment.
I think a policy of charging for a rescue is reasonable, as long as the policy -- and the cost -- is set in advance. There are many reasonable ways to do this. You can require users to carry insurance; you can provide insurance by user fees; or you can just state in advance that various kinds of rescues will be charged at different rates.
I don't think it is reasonable to set a reimbursement number after the fact. Nobody should be thinking about the budget in this kind of situation, or if anybody should it is not the family of the person being rescued. In practice this will spread costs over a number of users, and some will be "luckier" than others.
It is not necessarily the case that more expensive victims are more negligent than less expensive ones. Maybe if Fosset crashed a half mile away from where he did, the rescue would have been successful and cost 1/10 what it did.
Gibbons is the guy who (a) employed an undocumented nanny for six years and (b) as governor tried to get Mexico to fork over $7m to reimburse Nevada for health care spending on undocumented workers.
He's evidently a pretty nasty piece of work.
It's all about the writing.
Voyager came along at a time when keeping the franchise running creatively would have been a huge challenge in any case. TNG extended the formula from TOS as far as the creative team could. But on top of that they were already running another series in the same franchise: DS9.
You'd think that there is enough writing talent in the world to keep two Star Trek series running in parallel. However the number of writers capable of and interested in writing Star Trek scripts, and who were trusted to do so, might narrow down the possibilities a great deal.
For whatever reason, as time went on the creative well was evidently dry, and it was being shared by two series.
There were a lot of very good episodes in DS9. There were not a few reasonably good Voyager episodes. But towards the end of the DS9 run there were an inordinate number of stinkers, and Voyager was breaking out in lousy scripts as well.
If you imagine that there had been a single series, without the worst of the episodes and wth a few more really good episodes, then you would have had something closer to TNG in its popularity.
If you look at what the suit is supposed to do, and remember that there is supposed to be a person in it that is not screaming in agony, the only plausible explanation is that the suit is equipped with exotic physics technology that manipulates gravity or inertia. That would explain why he doesn't need to carry hundreds of liters of reaction mass around.
The "contrails" it leaves might be something like condensation trails in a could chamber, or they might literally be a kind of "smoke screen".
Nuclear power? Far too crude for Stark; it's the world's most dangerously overcomplicated method of boiling water. You'd need all kinds of fuel mass, not to mention complicated machinery, just to eke out a measly few neutrons.
... duotronically controlled matter antimatter reaction dirven vectorized propulsion. That'd be just the thing for a rocket boot, if you don't mind the risk that your athlete's foot would mutate into something unexpectedly horrible.
Stark would use something like
Iron man is shown leaving contrails -- or possibly smoke trails. Either way he's burning something that has to have mass and volume. Where does he keep it?
In the trailer, we see him crash the suit from thousand of feet up, breaking his fall by going through meters of reinforced concrete. Granting that the suit might survive this, its occupant is supposed to be human. Keep in mind that he brain and spinal cord have the approximate consistency of toothpaste.
Oh, come now, would you expect a genius like Stark would miss that bit?
The suit (apparently) has an excellent communications system, that surely include things like noise cancellation and background noise shaping. When he is suited up and taking a call in his Tony Stark persona, he doesn't want to sound like he's talking into a cheap headset, while standing on the deck of an aircraft carrier with with a cookpot over his head.
Come to think of it, the "flatness" is probably a deliberate attempt by the director to insinuate the notion into your head that you are hearing Stark over a mobile communication system. He probably could have made Downey sound ling Bing Crosby crooning into an RCA-44 mic, if he wanted to. In fact, that might have been an interesting choice.