Gah. I knew I should have read through that more carefully.:-| My brain said one word, my fingers typed another.
Nearly all biometric devices rely intensely on physical security, once you have access to the device and the wire almost all are subject to playback attacks, some as simple as "open circuit this wire" or "close circuit this wire".
And there's no real way to avoid that as long as the device is under someone else's control. At some point in the hardware, the unencrypted data must exist. It's basically the reverse of the DRM problem, and it is unsolvable for precisely the same reason. The only way to avoid the problem in the biometrics case is to deny the bad guy physical access to the hardware, as you said. In effect, this means that you would need to carry your own personal fingerprint scanner around in your pocket.
Short of that, biometrics cannot be made more secure than a (good) password, and in practice, biometrics tend to be much less secure because of how easy it is to obtain them surreptitiously (the old "dust the glass after you leave the restaurant" trick). And as soon as you guard that device as though losing it could mean someone draining your bank account, there's no longer any reason to require the biometric info; it is basically just as secure to authenticate based on whether you have possession of the device... and possibly more so.
Biometrics are a terrible idea. They can't be changed. That means that as soon as somebody lifts your fingerprint off that class, you're 0wn3d. Forever. Sorry, but biometrics are to proper security what the TSA is to proper security—a lot of flashy show with no real function—a tale told by an idiot, full of sound and fury, signifying nothing.
Thanks, but no thanks. Maybe as a fairly weak second factor, but not as a replacement for passwords. A more reasonable solution would be a chip-and-pin scheme in which you enter the pin on the (very simple, hard-to-hack) card rather than on the reader, so that the reader is just a dumb device that passes the authentication request through to a backend server and receives an authorization token a la Kerberos.
But yeah, it feels like the last time things bubbled up to the surface all at once like they're doing now was the 1930s. Not a good feeling at all.
Maybe you missed the McCarthy hearings in the 1950s. We've been in a rapid downhill slide before. Eventually, the public pulls its heads out of its collective asses, shouts "enough", and the politicians stop acting like morons for a decade or two. Then, somebody who wasn't around back then gets it into his/her head that he/she might be able to pull it off without getting caught this time, and the cycle repeats itself.
The U.S. government for at least the last century has been a pendulum endlessly swinging back and forth, with paranoid fascism at one end and relative innocuousness at the other. Right now we're down at the fascist end of the pendulum, but it's starting to look like enough of the public is pissed off enough to start nudging it back the other direction soon. One can only hope.
Warrantless wiretapping affected maybe a few thousand people. Indefinite military detention affected maybe a few thousand people. Criminal prosecution of journalists affects maybe a dozen people in any given year. The TSA forces 600 million people every year to choose between starring in a private porno video and getting felt up. Do the math.
The numbers don't lie. If you want to get the maximum reduction in government invasiveness in a single action, dismantle the TSA. (Better yet, dismantle Homeland Security and all the new agencies within it and go back to the 2001 organizational structure, and in one blow, you'll take care of the TSA and the warrantless wiretapping.)
Actually, correction, on further searching, Data Detectors were initially a Copland UI feature (though it might have appeared in Newton at some point as well). My bad.
Awk isn't prior art. It might be possible to implement a small portion of this using Awk (though it would be a royal pain in the backside—Perl would do a much better job given that we're talking about detecting data in free-form text, not formatted records), but that doesn't make it prior art. Prior art has to do everything that the patent covers, not just one little piece.
Prior art for this would be a graphical (or text-GUI) application that, as part of its normal operation, detected email addresses and other piece of information and acted on them in some interesting way. For example, if you could find a USENET news reader from 1995 that automatically detected email addresses outside the To and From fields and converted them to clickable links, then you would have prior art. The fact that someone could have implemented this using software that existed prior to 1996 is immaterial. As far as patent law is concerned, the only thing that matters is whether someone else actually did implement it prior to 1996.
Email links are explicitly marked. This patent is about detecting things that look like email addresses that are not marked with links in the HTML.
Also, all of your "prior art" examples were invented more than a decade after this patent was filed. Google didn't exist when this was filed, and neither did Microsoft Outlook. Netscape Navigator was still the most popular web browser—Netscape Communicator hadn't scared everyone over to IE yet. JavaScript had just been invented.
Remember, what's obvious now was not necessarily obvious fifteen years ago.
Maybe you didn't read the date on the patent. It was filed in 1996 and granted in 1999. This patent isn't an iPhone patent. It's a Newton patent.
More to the point, the patent application was filed before Google, HTC, or Zimbra were even founded—back in the day when Palm was just starting to take off. You're going to have to look a lot farther back than the Zimlet for prior art.
You've got it half right. What we need are the Facebooks and Googles of the world to work together for the common good.
All it would take is one ad campaign in which the first time each U.S. user logged into Facebook on a particular day, they would initially see a bleak, grey-looking Facebook. Every YouTube video on their site, every photo, and every link to every website would be replaced with the words "This content has been removed due to a copyright claim." Superimposed across the top of the screen in large letters would be the words, "If SOPA and PIPA pass, this is what Facebook will look like." Below that would be two links: "Tell me more" and "I don't care about freedom on the Internet. Just take me to Facebook." Either button would reset Facebook to normalcy, but the first button would take you to a Facebook page explaining the laws and why they are bad. At the bottom of the page would be links to email forms for the appropriate congresspeople based on the user's current address.
The key, however, is to do this not just for Facebook, but also for Google image searches, YouTube searches, etc. Every high profile site that is involve in video sharing has to do it on the same day. Such an ad campaign only has to happen once and I can pretty much guarantee that SOPA and PIPA would be buried for years. The hard part is convincing Facebook and Google to work together to put together such an ad campaign and stick their necks out that far in the interest of protecting their users. It's possible, but I'm not holding my breath.
The military demand for satellite recovery also added a lot to the cost, as I understand it. The shuttle was designed back in the era when they still used film cans in spy satellites, which meant they periodically needed to either restock them or replace them. As a result, the shuttle got a huge payload bay that was only used for an actual satellite recovery four times in the entire history of the shuttle program (five satellites, but Westar-6 and Palapa-B2 were a twofer).
I'm sure it was nice to have the option for some of those Hubble missions in case they weren't able to fix it in orbit, but that's an awful lot of complexity for something that they last did back in 1996. And AFAIK, it was never done for the military. Not once.
Maybe it's that with the overhaul the plants needed, the new production isn't fully debugged yet, so the expected failure rate has increased?
Or that the extra overhead of continuing to produce small numbers of repair parts for older drive mechanisms just became a lot more expensive now that their manufacturing capacity is in short supply. Admittedly, this could also be solved by sometimes shipping people an upgraded drive when theirs goes south, but....
Yes, but statistically speaking, if you're going to have a violation of an open source license, it's going to be GPL. That's the only popular open source license that is so restrictive.
Of course, there are a few licenses that are more restrictive—the Microsoft Shared Source license (which many employers forbid their employees to touch with a ten meter pole) and the Affero GPL (which I generally lump in with GPL even though it's a bit nastier)—but you're not too likely to encounter either one very often.
It's not that. The reason GPL is problematic is that it's all too easy to copy and paste a couple of lines of code out of some open source project into something you're working on. If it's under a BSD license, no problem. If it's under a GPL license, you're screwed. For this reason, the safest default policy for big corporations is to deny all use of GPLed software to remove the temptation.
The result of this is that folks working for those companies are less likely to spend time working on GPLed projects. More importantly, because those companies are not bringing in GPLed source from the outside, they are no longer forced to use that license for their own code. The net effect is that less GPLed code gets produced.
That's pretty much how Congress works, but not quite. In general, they propose something, then if people scream, they wait a while for the anger to die down, then pass almost exactly the same bad bill as soon as they think everyone who cares has stopped paying attention.... If we're lucky, the most outlandish clauses will have been removed, but there's certainly no guarantee....
No, no, you have it wrong. Lobbying is a good thing. It's the way that the public can express their opinions to their elected representatives. The reason lobbying appears to be a bad thing is that only big corporations can afford to lobby in any useful way because our Congress stays in Washington D.C. all year around, and they barely set foot in their actual legislative districts.
If you really want to improve the situation, change the law so that Congress must do their job over the Internet from their districts. This means that your representatives will be accessible to you. This also means that corporations that want to lobby Congress will have to send someone all around the country to lobby instead of just all around a building. This will effectively end the corporate dominance of lobbying and bring the voice of the people back into our government.
A corporation is a tool, nothing more than a way for many not-so-rich people to fund an effort and own the result, rather that the prior model where only the 0.01% could play. Like any tool, it can be used for good or evil.
A corporation is also a means of dodging responsibility for one's actions. Because the leaders of a corporation can't be held responsible for what the company does except in certain extreme situations, corporate speech does not have the same liability/risk that individual speech does. This is why it must be regulated in ways that individual speech is not. No freedom without responsibility and all that.
And the reality is that even if you get together with a million other geeks, you will not be able to do much in terms of lobbying. Congresspeople don't give a rat's ass about any group of people, including corporations, unless that group is creating lots of jobs in their district. What this means is that no political organization has any real bearing on anything in politics beyond perhaps a little lip service from the politicians as they try to make it appear that they still represent the people as a whole. Maybe, maybe you might be able to sway an election to the other candidate. The problem is that the other candidate doesn't give a rat's ass about your opinion, either, which means the best you can really do is nudge the ball a few feet to either side of the fifty yard line.
In short, corporations are so much unlike the general public that any attempt at comparison is meaningless.
I wish so badly that our government would think about how long a law will have any useful reason for existing, and sunset them all appropriately. This one is a great example of a law that will be very nearly useless in twenty years. And yet, even when I'm in a Google-powered autonomous vehicle in five or six years, they will still pull me over and probably ticket me for sitting in the leftmost seat while talking on the phone even though I'm not driving. And this is why passing pencil pusher laws about fairly minor things like talking on the phone while driving is a bad idea—only two things essentially never shrink: the human waistline and the body of law.
By contrast, every single proposal I mentioned will likely be important and useful in the medium to long term unless we switch to roads with physical tracks to limit horizontal deviation or something. (Okay, so the HUD won't be useful, but the HUD could be repurposed for other display purposes, while the sensors that back it could feed into the computer, so everything but the software integration will be useful. You get my point, though.)
Will getting of the phone make the road safer? Maybe slightly. Is an anti-cell-phone law failing to see the forest for the trees? Definitely.
Now to turn at least part of that around. A sizable percentage of the traffic accident reports come from people phoning into the local radio station and telling them about an accident. That's how they're able to know what's going on before even the highway patrol does half the time. The result is that cars hear about the problem and route themselves around the section of highway that is backed up. This significantly reduces the backup.
Statistically, the average backup has something like three additional minor accidents in the backup, and that number is directly proportional to the number of cars involved in the backup. Therefore, people making phone calls while driving almost certainly reduces the number of traffic accidents on the road on the whole. The problem is that the costs are trivial to measure, whereas the benefits are very, very hard to measure because you're trying to compute how many accidents didn't happen....
That and because the people doing the cost-benefit analyses are mostly myopic luddite technophobes who are desperately seeking a scapegoat for traffic accidents so that they can look like they are doing something to make life safer. *sigh*
Want to make driving safer? Do this:
Build vehicles that are safer.
Push hard for autonomous vehicles.
Develop safety systems that make it easier for drivers to see where the vehicles are around them, e.g. a HUD showing a 2D map with your car in the center and colored areas for cars around you on all four sides and all four corners.
Require that every road with a truck speed limit have at least three lanes.
Ban trucks during rush hour from any roads where trucks have a separate speed limit.
Spend more money on alternatives such as high speed rail.
Getting off a bus/train/whatever and walking the last 5-10 blocks is a potentially deadly health risk to the very young and very old at that point.
Only because there aren't enough interior walkways between the bus stop and their destination. That problem is trivially solvable with a sufficient investment in glass walkways, at least in most cities. Cities that are sprawled enough to thwart such a scheme are also usually too sprawled for mass transit to be significantly more efficient than individual cars anyway.
As soon as you say "go to", that's where they made their first mistake. The first rule of sales is that you never, ever lowball your initial asking price. If you do, people come to expect it at that price and won't pay more. That's why most books have a fixed price that doesn't change for the life of the book, sales on the used market notwithstanding. Raising the price of an existing book will always cause sales to dry up.
The second rule of sales is that you should set the price at whatever you think your offering is worth. Commercial book titles cost $9.99 and up, so if you set your price at $0.99 or less, you are telling people that you think your book is worth less than a tenth what a book from a real publisher is worth. That's setting yourself up for failure right off the bat. People are going to immediately think your book is worthless compared with commercial titles because you think it is. And it doesn't mater if you raise the price later; they're going to see it as a $0.99 book, and you as a $0.99 author for a long time to come.
Instead, you should set the initial price to be comparable with commercial titles. Then, offer a one-day or one-week promotional sale of a title at $3.99 or so. Psychologically, this makes people think they're getting a deal by getting it for so much less than list. This creates a flurry of interest for a short term, which drives reviews, which drives both sales rank and word-of-mouth advertisement, which in turn drive future sales.
If you've already made this mistake, raise the price on all your books to $9.99 or more today, and in a couple of years, when you release your next book, you'll start getting sales again. Amusingly, you'll start getting sales for all of your books. That's another psychological trick. When people see a new title at the same cost as commercial offerings by an author with a lot of previous titles, they are likely to assume that the author is a successful writer, and are more likely to give the book a chance.
Finally, you need to produce paper versions of your books. I know that's an unpopular subject among some eBook publishers, but the fact remains that a lot of folks won't touch an eBook without a printed edition. Some people read print books exclusively, for one thing, so you're limiting your market if you don't. Also, a printed edition means that either A. you cared enough about your book and the way it is presented to do the work needed to see it in print, or B. a publisher cared enough about it to do the extra work. Either way, that's often (but not always) a good indication of quality.
Don't get me started ranting about my experience publishing eBook content with LaTeX....
What I want to know is whether there's some way to combine the new content created with the new tool with the old content created with the old tool programmatically. Because KindleGen is remarkably bad at handling CSS, I had to make fairly dramatic (and massively non-spec-compliant) changes to my EPUB doc to create something that still looked reasonably formatted after transforming it into Kindle format. I'd love to use the new tools to generate the new format, but presumably I'd want to start with something approaching proper EPUB instead of a hacked copy. However, unless the KindleGen tool got a lot better, that will probably mean less than acceptable results on older devices that don't use the new format.
Also, where the hell are the new tools for generating these things? The latest version on Amazon's site is the old 1.2 from six months ago. KindleGen 2 is nowhere to be found. How are we supposed to start transitioning to the new tools when we can't get the new tools?
And this is why Amazon needs to get with the program and switch to EPUB. Amazon *and* publishers have to do an absolutely *insane* amount of work to support their hack of a platform because they are the only devices in the world that don't support open standards. What a mess. Amazon, just STOP. Stop releasing new versions of your Kindle format. Stop updating KindleGen. Stop using a nonstandard format on your readers. Nothing makes me long for the relative simplicity of RGB to CMYK conversions than dealing with your crap....:-/
It's 100% bogus thinking in every respect. I don't know anyof my friends' financial situations, or at least every few. Most of my friends on Facebook are people I know from school (many years ago), from outside activities, etc. My friends from musical ensembles are (slightly) more likely to buy musical instruments and supplies, my friends from work are more likely to buy computer stuff, and the remaining folks are just a random cross-section of the population as a whole.
With the possible exception of my work friends, none of them are significantly more likely to have similar credit habits than the general population sampled at random. And my work friends don't really provide any additional insight into my credit habits that you couldn't get just as easily by asking where I work.
In short, it's crap. Maybe in a few edge cases, where someone has no credit history and didn't give a list of employers, you might be able to get some insight about where they have previously worked based on who their friends are, which might have some slight correlation to credit behavior, but even that is a stretch. Either way, anyone clueless enough to knowingly give out their Facebook and Twitter credentials to a bank deserves to have credit denied on principle. That's just not something you do.
Slashdot Mirror
Gah. I knew I should have read through that more carefully. :-| My brain said one word, my fingers typed another.
And there's no real way to avoid that as long as the device is under someone else's control. At some point in the hardware, the unencrypted data must exist. It's basically the reverse of the DRM problem, and it is unsolvable for precisely the same reason. The only way to avoid the problem in the biometrics case is to deny the bad guy physical access to the hardware, as you said. In effect, this means that you would need to carry your own personal fingerprint scanner around in your pocket.
Short of that, biometrics cannot be made more secure than a (good) password, and in practice, biometrics tend to be much less secure because of how easy it is to obtain them surreptitiously (the old "dust the glass after you leave the restaurant" trick). And as soon as you guard that device as though losing it could mean someone draining your bank account, there's no longer any reason to require the biometric info; it is basically just as secure to authenticate based on whether you have possession of the device... and possibly more so.
Biometrics are a terrible idea. They can't be changed. That means that as soon as somebody lifts your fingerprint off that class, you're 0wn3d. Forever. Sorry, but biometrics are to proper security what the TSA is to proper security—a lot of flashy show with no real function—a tale told by an idiot, full of sound and fury, signifying nothing.
Thanks, but no thanks. Maybe as a fairly weak second factor, but not as a replacement for passwords. A more reasonable solution would be a chip-and-pin scheme in which you enter the pin on the (very simple, hard-to-hack) card rather than on the reader, so that the reader is just a dumb device that passes the authentication request through to a backend server and receives an authorization token a la Kerberos.
Maybe you missed the McCarthy hearings in the 1950s. We've been in a rapid downhill slide before. Eventually, the public pulls its heads out of its collective asses, shouts "enough", and the politicians stop acting like morons for a decade or two. Then, somebody who wasn't around back then gets it into his/her head that he/she might be able to pull it off without getting caught this time, and the cycle repeats itself.
The U.S. government for at least the last century has been a pendulum endlessly swinging back and forth, with paranoid fascism at one end and relative innocuousness at the other. Right now we're down at the fascist end of the pendulum, but it's starting to look like enough of the public is pissed off enough to start nudging it back the other direction soon. One can only hope.
Warrantless wiretapping affected maybe a few thousand people. Indefinite military detention affected maybe a few thousand people. Criminal prosecution of journalists affects maybe a dozen people in any given year. The TSA forces 600 million people every year to choose between starring in a private porno video and getting felt up. Do the math.
The numbers don't lie. If you want to get the maximum reduction in government invasiveness in a single action, dismantle the TSA. (Better yet, dismantle Homeland Security and all the new agencies within it and go back to the 2001 organizational structure, and in one blow, you'll take care of the TSA and the warrantless wiretapping.)
Actually, correction, on further searching, Data Detectors were initially a Copland UI feature (though it might have appeared in Newton at some point as well). My bad.
Awk isn't prior art. It might be possible to implement a small portion of this using Awk (though it would be a royal pain in the backside—Perl would do a much better job given that we're talking about detecting data in free-form text, not formatted records), but that doesn't make it prior art. Prior art has to do everything that the patent covers, not just one little piece.
Prior art for this would be a graphical (or text-GUI) application that, as part of its normal operation, detected email addresses and other piece of information and acted on them in some interesting way. For example, if you could find a USENET news reader from 1995 that automatically detected email addresses outside the To and From fields and converted them to clickable links, then you would have prior art. The fact that someone could have implemented this using software that existed prior to 1996 is immaterial. As far as patent law is concerned, the only thing that matters is whether someone else actually did implement it prior to 1996.
Yeah, but Apple did it and patented it seven years before Skype was even founded. Everything seems obvious fifteen years after its invention....
Email links are explicitly marked. This patent is about detecting things that look like email addresses that are not marked with links in the HTML.
Also, all of your "prior art" examples were invented more than a decade after this patent was filed. Google didn't exist when this was filed, and neither did Microsoft Outlook. Netscape Navigator was still the most popular web browser—Netscape Communicator hadn't scared everyone over to IE yet. JavaScript had just been invented.
Remember, what's obvious now was not necessarily obvious fifteen years ago.
Maybe you didn't read the date on the patent. It was filed in 1996 and granted in 1999. This patent isn't an iPhone patent. It's a Newton patent.
More to the point, the patent application was filed before Google, HTC, or Zimbra were even founded—back in the day when Palm was just starting to take off. You're going to have to look a lot farther back than the Zimlet for prior art.
You've got it half right. What we need are the Facebooks and Googles of the world to work together for the common good.
All it would take is one ad campaign in which the first time each U.S. user logged into Facebook on a particular day, they would initially see a bleak, grey-looking Facebook. Every YouTube video on their site, every photo, and every link to every website would be replaced with the words "This content has been removed due to a copyright claim." Superimposed across the top of the screen in large letters would be the words, "If SOPA and PIPA pass, this is what Facebook will look like." Below that would be two links: "Tell me more" and "I don't care about freedom on the Internet. Just take me to Facebook." Either button would reset Facebook to normalcy, but the first button would take you to a Facebook page explaining the laws and why they are bad. At the bottom of the page would be links to email forms for the appropriate congresspeople based on the user's current address.
The key, however, is to do this not just for Facebook, but also for Google image searches, YouTube searches, etc. Every high profile site that is involve in video sharing has to do it on the same day. Such an ad campaign only has to happen once and I can pretty much guarantee that SOPA and PIPA would be buried for years. The hard part is convincing Facebook and Google to work together to put together such an ad campaign and stick their necks out that far in the interest of protecting their users. It's possible, but I'm not holding my breath.
The military demand for satellite recovery also added a lot to the cost, as I understand it. The shuttle was designed back in the era when they still used film cans in spy satellites, which meant they periodically needed to either restock them or replace them. As a result, the shuttle got a huge payload bay that was only used for an actual satellite recovery four times in the entire history of the shuttle program (five satellites, but Westar-6 and Palapa-B2 were a twofer).
I'm sure it was nice to have the option for some of those Hubble missions in case they weren't able to fix it in orbit, but that's an awful lot of complexity for something that they last did back in 1996. And AFAIK, it was never done for the military. Not once.
Or that the extra overhead of continuing to produce small numbers of repair parts for older drive mechanisms just became a lot more expensive now that their manufacturing capacity is in short supply. Admittedly, this could also be solved by sometimes shipping people an upgraded drive when theirs goes south, but....
Yes, but statistically speaking, if you're going to have a violation of an open source license, it's going to be GPL. That's the only popular open source license that is so restrictive.
Of course, there are a few licenses that are more restrictive—the Microsoft Shared Source license (which many employers forbid their employees to touch with a ten meter pole) and the Affero GPL (which I generally lump in with GPL even though it's a bit nastier)—but you're not too likely to encounter either one very often.
Yes, but the silence speaks volumes. How many other workplaces do you know that don't include the company name anywhere on the badge?
It's not that. The reason GPL is problematic is that it's all too easy to copy and paste a couple of lines of code out of some open source project into something you're working on. If it's under a BSD license, no problem. If it's under a GPL license, you're screwed. For this reason, the safest default policy for big corporations is to deny all use of GPLed software to remove the temptation.
The result of this is that folks working for those companies are less likely to spend time working on GPLed projects. More importantly, because those companies are not bringing in GPLed source from the outside, they are no longer forced to use that license for their own code. The net effect is that less GPLed code gets produced.
Hear, hear. I didn't know what Megaupload was until Universal did this.
That's pretty much how Congress works, but not quite. In general, they propose something, then if people scream, they wait a while for the anger to die down, then pass almost exactly the same bad bill as soon as they think everyone who cares has stopped paying attention.... If we're lucky, the most outlandish clauses will have been removed, but there's certainly no guarantee....
No, no, you have it wrong. Lobbying is a good thing. It's the way that the public can express their opinions to their elected representatives. The reason lobbying appears to be a bad thing is that only big corporations can afford to lobby in any useful way because our Congress stays in Washington D.C. all year around, and they barely set foot in their actual legislative districts.
If you really want to improve the situation, change the law so that Congress must do their job over the Internet from their districts. This means that your representatives will be accessible to you. This also means that corporations that want to lobby Congress will have to send someone all around the country to lobby instead of just all around a building. This will effectively end the corporate dominance of lobbying and bring the voice of the people back into our government.
A corporation is also a means of dodging responsibility for one's actions. Because the leaders of a corporation can't be held responsible for what the company does except in certain extreme situations, corporate speech does not have the same liability/risk that individual speech does. This is why it must be regulated in ways that individual speech is not. No freedom without responsibility and all that.
And the reality is that even if you get together with a million other geeks, you will not be able to do much in terms of lobbying. Congresspeople don't give a rat's ass about any group of people, including corporations, unless that group is creating lots of jobs in their district. What this means is that no political organization has any real bearing on anything in politics beyond perhaps a little lip service from the politicians as they try to make it appear that they still represent the people as a whole. Maybe, maybe you might be able to sway an election to the other candidate. The problem is that the other candidate doesn't give a rat's ass about your opinion, either, which means the best you can really do is nudge the ball a few feet to either side of the fifty yard line.
In short, corporations are so much unlike the general public that any attempt at comparison is meaningless.
I wish so badly that our government would think about how long a law will have any useful reason for existing, and sunset them all appropriately. This one is a great example of a law that will be very nearly useless in twenty years. And yet, even when I'm in a Google-powered autonomous vehicle in five or six years, they will still pull me over and probably ticket me for sitting in the leftmost seat while talking on the phone even though I'm not driving. And this is why passing pencil pusher laws about fairly minor things like talking on the phone while driving is a bad idea—only two things essentially never shrink: the human waistline and the body of law.
By contrast, every single proposal I mentioned will likely be important and useful in the medium to long term unless we switch to roads with physical tracks to limit horizontal deviation or something. (Okay, so the HUD won't be useful, but the HUD could be repurposed for other display purposes, while the sensors that back it could feed into the computer, so everything but the software integration will be useful. You get my point, though.)
Will getting of the phone make the road safer? Maybe slightly. Is an anti-cell-phone law failing to see the forest for the trees? Definitely.
Now to turn at least part of that around. A sizable percentage of the traffic accident reports come from people phoning into the local radio station and telling them about an accident. That's how they're able to know what's going on before even the highway patrol does half the time. The result is that cars hear about the problem and route themselves around the section of highway that is backed up. This significantly reduces the backup.
Statistically, the average backup has something like three additional minor accidents in the backup, and that number is directly proportional to the number of cars involved in the backup. Therefore, people making phone calls while driving almost certainly reduces the number of traffic accidents on the road on the whole. The problem is that the costs are trivial to measure, whereas the benefits are very, very hard to measure because you're trying to compute how many accidents didn't happen....
That and because the people doing the cost-benefit analyses are mostly myopic luddite technophobes who are desperately seeking a scapegoat for traffic accidents so that they can look like they are doing something to make life safer. *sigh*
Want to make driving safer? Do this:
Only because there aren't enough interior walkways between the bus stop and their destination. That problem is trivially solvable with a sufficient investment in glass walkways, at least in most cities. Cities that are sprawled enough to thwart such a scheme are also usually too sprawled for mass transit to be significantly more efficient than individual cars anyway.
As soon as you say "go to", that's where they made their first mistake. The first rule of sales is that you never, ever lowball your initial asking price. If you do, people come to expect it at that price and won't pay more. That's why most books have a fixed price that doesn't change for the life of the book, sales on the used market notwithstanding. Raising the price of an existing book will always cause sales to dry up.
The second rule of sales is that you should set the price at whatever you think your offering is worth. Commercial book titles cost $9.99 and up, so if you set your price at $0.99 or less, you are telling people that you think your book is worth less than a tenth what a book from a real publisher is worth. That's setting yourself up for failure right off the bat. People are going to immediately think your book is worthless compared with commercial titles because you think it is. And it doesn't mater if you raise the price later; they're going to see it as a $0.99 book, and you as a $0.99 author for a long time to come.
Instead, you should set the initial price to be comparable with commercial titles. Then, offer a one-day or one-week promotional sale of a title at $3.99 or so. Psychologically, this makes people think they're getting a deal by getting it for so much less than list. This creates a flurry of interest for a short term, which drives reviews, which drives both sales rank and word-of-mouth advertisement, which in turn drive future sales.
If you've already made this mistake, raise the price on all your books to $9.99 or more today, and in a couple of years, when you release your next book, you'll start getting sales again. Amusingly, you'll start getting sales for all of your books. That's another psychological trick. When people see a new title at the same cost as commercial offerings by an author with a lot of previous titles, they are likely to assume that the author is a successful writer, and are more likely to give the book a chance.
Finally, you need to produce paper versions of your books. I know that's an unpopular subject among some eBook publishers, but the fact remains that a lot of folks won't touch an eBook without a printed edition. Some people read print books exclusively, for one thing, so you're limiting your market if you don't. Also, a printed edition means that either A. you cared enough about your book and the way it is presented to do the work needed to see it in print, or B. a publisher cared enough about it to do the extra work. Either way, that's often (but not always) a good indication of quality.
Don't get me started ranting about my experience publishing eBook content with LaTeX....
What I want to know is whether there's some way to combine the new content created with the new tool with the old content created with the old tool programmatically. Because KindleGen is remarkably bad at handling CSS, I had to make fairly dramatic (and massively non-spec-compliant) changes to my EPUB doc to create something that still looked reasonably formatted after transforming it into Kindle format. I'd love to use the new tools to generate the new format, but presumably I'd want to start with something approaching proper EPUB instead of a hacked copy. However, unless the KindleGen tool got a lot better, that will probably mean less than acceptable results on older devices that don't use the new format.
Also, where the hell are the new tools for generating these things? The latest version on Amazon's site is the old 1.2 from six months ago. KindleGen 2 is nowhere to be found. How are we supposed to start transitioning to the new tools when we can't get the new tools?
And this is why Amazon needs to get with the program and switch to EPUB. Amazon *and* publishers have to do an absolutely *insane* amount of work to support their hack of a platform because they are the only devices in the world that don't support open standards. What a mess. Amazon, just STOP. Stop releasing new versions of your Kindle format. Stop updating KindleGen. Stop using a nonstandard format on your readers. Nothing makes me long for the relative simplicity of RGB to CMYK conversions than dealing with your crap.... :-/
It's 100% bogus thinking in every respect. I don't know anyof my friends' financial situations, or at least every few. Most of my friends on Facebook are people I know from school (many years ago), from outside activities, etc. My friends from musical ensembles are (slightly) more likely to buy musical instruments and supplies, my friends from work are more likely to buy computer stuff, and the remaining folks are just a random cross-section of the population as a whole.
With the possible exception of my work friends, none of them are significantly more likely to have similar credit habits than the general population sampled at random. And my work friends don't really provide any additional insight into my credit habits that you couldn't get just as easily by asking where I work.
In short, it's crap. Maybe in a few edge cases, where someone has no credit history and didn't give a list of employers, you might be able to get some insight about where they have previously worked based on who their friends are, which might have some slight correlation to credit behavior, but even that is a stretch. Either way, anyone clueless enough to knowingly give out their Facebook and Twitter credentials to a bank deserves to have credit denied on principle. That's just not something you do.