Yeah things like this have been made to work under controlled conditions,
What are you talking about? These technologies (not this one, but similar ones) are in my car right now and occasionally I enter less-than-controlled conditions. Hydroplaning and other less-than-optimal road conditions are common and that's what some of these technologies are there for.
AGAIN: Car manufacturers are not going to add a "safety" feature that causes more accidents than it prevents. Based on your posts, I have to believe that this is what you think.
Sure you can write software that tries to follow a line, but will it be safe?
If implemented correctly, yes. If it applies minor adjustments to compensate for a slow drift, yes, I believe this can be used safely. A system like this would not have the ability to spin the steering wheel around to the right and slam you into the car next to you.
The point is this safety technology is not safe.
And I submit to you that it is (and will be) safe. If statistics get published that indicate these features cause 10 times the accidents they would otherwise prevent, I think someone would have an awful lot of explaining to do.
but allowing it to hit the brakes is crazy.
Some higher-end cars automatically decellerate when they detect a critical engine problem. They've been doing this for years. Maybe car manufacturers have found out how to do this safely already?
the last thing you want is for something to unexpectedly slam on the brakes, period.
Who said anything about slamming on the brakes? You're exaggerating to try and make your point valid. This device lightly applies brakes. Yes, I suppose in a frictionless environment, your wheels will (arguably, given things like traction control) cease to maintain traction on the road. But are you worse off at this point? If lightly hitting the brakes is enough to do this, how would you have avoided the car in front of you anyway?
What's frustrating to me is that you're not even willing to accept that Honda has thought of these situations. Do you believe that your experience with RADAR makes you more qualified to speak about these technologies than the guys actively developing it? Honda has a bit of an interest in keeping its drivers alive. I'm perfectly willing to concede that implementing something like this is going to be tricky. But I'm also going to trust that Honda has thought of all of the ways that this could be dangerous and has built something that addresses most (if not all) of those concerns. An on/off button that could be used in adverse conditions should satisfy a lot of these concerns right up front. The next time I'm following a line of cars across a frozen lake, I might want to turn this off. (Actually, I probably wouldn't, since this device, in combination with traction control, could still likely do a better job than I could if I'm not paying attention to the point where this is activated.)
Perhaps you're right, in that there will be a 1% increase in accidents resulting from an unexpected activation of this device. But if this can reduce rear-end collisions by even 10%, wouldn't that be worth it? And if you don't think so, you should have the ability to turn this off (or not ask for it when you buy your next car).
I'm not asking you to say that this technology is safe, I'm just asking you to consider the possibility that Honda has thought of everything you've thought of here, and that your concerns have been addressed in some fashion.
Let the record show that you're the one that started name-calling here, not me.
Yeah, I must be stupid for actually understanding the limits of today's technology.
You don't appear to, or perhaps you are misunderstanding how that technology is being applied in this situation.
You clearly don't
Whatever, man. The bottom line is that you do not have the specifics here and are pulling shit out of your ass as a result. You may label it highly educated, RADAR engineer shit, but it's still shit. I asked you to consider the possibility that they've done what they claim to have done, and you've rejected that. Sounds pretty closed-minded to me.
Spoken like someone who doesn't understand how to drive in the snow.
I learned to drive in a Ford Escort in the dead of winter in Anchorage, Alaska. Here again, you are armed with no knowledge of something and pull more shit out of your ass.
Driving on a sheet of ice, one can indeed safely apply brakes to slow down. In the event you apply them too hard, additional technology such as skid prevention would step in to prevent my braking from causing me to lose control. How do I know this? Because I have done it.
Are you just unaware of these technologies? Perhaps that's why this "next step" into the world of RADAR seems so strange and impossible to you. It doesn't sound like you've fully come to terms with the stuff that's already out there.
The only info this system would have, would be the range, and amplitude of the reflection. That's it.
Even with one dimension and 16 degrees of arc (from which I might argue a two-dimensional view might be obtained, but since I have no specifics we can just assume it's a single range measurement covering an area of 16 degrees), an image zipping in and out of its field of view still looks considerably different than a car in your lane in front of you (that's probably been there for more than a few seconds). Remember that this thing is intended to prevent rear-end collisions, not random collisions with any object in the road or passing in front of you. Even with a simple/dumb range finder, it's pretty easy to see when the car in front of you is slowing down or stopping suddenly as opposed to seeing a bird or some other object move in front of it.
But still (and again), please consider the possibility that the people developing these technologies might have someone on staff that could come up with these scenarios just as easily as you could.
Let me repeat myself: Car manufacturers are not going to add technology to a car that will cause it to have an accident once a week.
I simply don't understand how you are coming back to the conclusion that what they're doing is impossible. THEY'VE DONE IT! This has been in testing for years. I sincerely doubt you've driven in one of these things or been given a demonstration of the technology or the behaviors that govern its actions. Again, please consider the possibility that your concerns have been addressed and stop going off about how much you know about RADAR and how they must be lying. This whole thread is incredibly silly and you're really starting to look like an ass.
P.S., if this piece of tech has already got you believing Honda is in some fantasy land, note that this model also has image recognition software designed to make small corrections to help keep a vehicle in its lane on the highway. Feel free to say this is impossible as well.
Can we please concede the possibility that car manufacturers will overcome some of the most basic issues you address before putting this into their cars? Nobody is going to drive a car that averages one accident a week. Think about this.
What about snow, mud, gravel (insert low traction condition here)?
If you are driving in conditions that light braking will send your car out of control, and the car is unable to detect and respond to this through other mechanisms (steering control, skid control, ABS, whatever), you are already fucked.
What if you're being tailgated?
I might accept this as an exceptional case. But even so, the light braking that this device does shouldn't be sufficient to cause a tailgater to crash into you. And even if it does, it's his fault. Chances are, if this system is activated and decides to apply some brakes, you're already in a high-risk situation (imminent accident) where this additional risk is probably acceptable.
But given that we have so little information about how these devices make their decisions, this is all possible perceived problems and conjecture pulled from our collective asses, yes?
What if you're across from someone in a left turn lane?
I would be shocked if this common situation were not addressed before this device is rolled out.
What if you're speeding up to try an avoid an accident?
What if the technology noticed this?
What if the system malfunctions?
What if ABS malfunctions? What if dynamic steering or skid control malfunctions? Which technology poses the greater risk here in the event of malfunction?
What if a bird flys in front of your car?
The radar profile of a bird (size, speed, likelyhood of collision) is probably significantly different than a large obstacle or a car. I wouldn't be surprised if a bird was ignored entirely. Even if it wasn't, there would be a tiny fraction of a second where the computer would like to see you slow down a bit. That might not be enough time for it to actually act on that. *shrug*
What if you're making a sharp turn, and any braking will cause you to loose control?
Again, I would honestly believe that a car manufacturer would look at these scenarios and come up with something that does not kill its passengers so easily. Additionally, most cars have features nowadays that will compensate for such actions. If it feels the car losing control, it will adjust your braking power or steering to compensate. So the logic is there. How well it integrates with some feature like this (by reducing or eliminating the braking, for example) isn't something I can speak to.
As smart as you think you are, please consider the likelyhood that the people designing these things might understand the issues better than you.
A directory is a search engine for proper names. There will be a finite number of organizations doing business as "Example Widgets". If the scope of my search is nation-wide, and two or three hits come up with "Example Widgets" incorporated in a few local markets, then I guess I need to do some more research to identify which one I want. Additional fields in the record would allow me to determine which Example Widgets is the one I am searching for, such as the home city.
There isn't a "placement" issue because these aren't really content hits. I'm not searching for a product here and getting a list of organizations that provide that product. I'm searching for an organization. If I choose the wrong one because I don't have enough information to pick between the matches, I'm probably not going to end up where I want to be. This is different than a company trying to be at the top of search rankings because it wishes to get the business of someone doing the search. In that case, I'm searching for the provider of a service. In the case where this directory is most useful, I am searching for a specific organization.
Similarly, a more traditional (but still regulated) search engine would allow me to actually search for product names or other service marks, and have that associated with the organization that owns it (perhaps with a URL). In this case as well, certain marks may be scoped to certain markets (countries?). Similar "placement" rules would have to apply for those rare situations. This could be as simple as a multi-column display with each column containing some 'sort by' mechanics. This isn't any different from any other directory browser we use today.
Together, the two options above would satisfy the 'locator' abuse of DNS, where I need to find a specific company, and assume a 'www' and a 'dot-com' to get there, as well as the 'product search' abuse of DNS, where I need to find a specific product, and assume a 'www' and a 'dot-com' to get there as well. With those qualities removed from DNS, nearly all intellectual property weight is removed and DNS can become a simple hierarchical label again.
Lastly, more traditional search engines would be able to associate generic product terms with more specific products that may satisfy the requirement you're hunting for. We don't care at that point how those are presented or ordered. Traditional search engines or global product catalogs have those issues covered.
I would agree with you, except search engines carry no authority. You could easily manipulate your own pages and things to make search engines "think" that your pages are more relevant for a given topic (or company name) than the company itself.
An authoritative search engine might be a good place to start worrying about trademark and service marks, though.
You still need that central authority if you need the ability for your labels to carry IP weight, but these labels must be scoped within political boundaries for that to be feasible.
DNS could actually still function in that capacity through the elimination of the GTLDs, and enforcement of location-specific hierarchies.
We need to start getting people away from using DNS as a locator service. DNS is meant (mainly) to apply convenient names to IP addresses, not so that my mom can guess "www.example.com" when she wants to know something about "example".
DNS is inappropriate for this because it does not allow two parties with a legitimate interest in the same label to share it, except through the confusion of additional TLDs. Two parties could have a perfectly legal claim to a label that they now have to battle out in courts. A better solution would be to create a directory on top of DNS to map these real-world names to DNS domains. Put all of the trademark and IP crap you want into regulating this directory, but leave it out of DNS.
LDAP or X.500 could do this nicely, with the added benefit of allowing distinguished names to include information about the locality (thus legal jurisdiction). This limits the scope of an IP conflict to the jurisdiction in question.
If I want to load the home page for Example Widgets, I should just need to hunt for the common name "Example Widgets", pick the match that makes the most sense (assuming more than one match is found), and have my lookup return a DNS domain name. My browser would consult DNS for a SRV record associated with that domain, find a host providing HTTP services for it, and request the home page from that host. At no point should my mom ever need to be exposed to a hostname (or a URL for that matter).
Don't confuse the official tolerances for deducting points from your license with "soft" speed limits. On the MIT page (which is what I assume you're talking about), there is a section labeled "Official tolerances" that discusses the subject of points against your license only. They are not suggesting that cops are not within their rights to bust you for going 61 in a 60.
Sometimes you can "get away" with speeding if you're within certain margins of error (e.g. radar), but these margins are most certainly not codified in law and would require some convincing argument in court.
The bottom line: most states consider their speed limits "absolute" in all respects. Some of them allow their lower speed limits to be "soft" if you can demonstrate you were travelling safely even though you were speeding, until you hit the state's absolute limit, which is absolute. Three states even consider their statewide speed limit to be "soft".
(ps. There are more to articles than pretty pictures and charts.)
Don't be an ass. Please concede the possibility that you may be wrong before you go off insulting people who disagree with you.
12v * 3 = 36v, 14v * 3 = 42v. It's a convenience factor for dealing with established 12v systems. Both the battery and charging (alternator) voltages are 3 times larger.
Did you actually read the links you posted for us? It looks to me like 3/4 of the states use absolute limits. When you exceed the posted speed limit, you are guilty of speeding. In fact, only 3 states seem have universally "soft" limits that you can contest if you can prove you were exceeding that limit safely.
So no, your assertion that "in most states the speed limit is a suggested value" does not appear to be accurate.
I personally have a perfect driving record, but there's no way in hell I'd voluntarily turn this feature off. If I get into an accident, I want my accident to help others in any way possible. I don't understand why this 5-second thing is any sort of invasion of privacy. The only time the information could be used is after I've destroyed my car (set off the airbags). The only time the information would be used is when the act of destroying my car created a situation where the police (or myself) would want the information analyzed. If the police are after it, they need to convince a judge to get a search warrant. At that point, any "privacy rights" I think I have are out the window anyway.
NASA is a public institution and the public and publicity surrounding the space program demand it.
I think we are in agreement here.
That is why I think it should be a private endevour.
I might suggest that NASA (and perhaps the public) try to understand why certain levels of risk are acceptable and adapt. I'm not sure that a private endeavor is going to fare any better, given the litigation-happy state of the US. Perhaps another country would be better to host this.
Various mars and moon programs? The moon programs were 30 years ago.
Lunar Prospector was a few years ago. Remember all of the hype about finding water on the moon?
get lucky and land in an interesting place or you could land in the middle of the Sahara.
Not to let us drift too far, but bear in mind that in some respects, we have better maps and imagery of the moon and Mars than we do of many places on earth. If we were investigating the earth, finding an interesting spot wouldn't be very difficult. But given the lack of anything obvious from a distance, I suppose you have a point there. I disagree that giving up before getting started is the right decision, though.
It is the unbridled human spirit and mind, which will find a way to move forward.
This isn't very educated. You're suggesting that since the US promotes some form of freedom, that other governments cannot possibly produce people that feel they're pursuing a dream by going into space?
China isn't that different from Russia back during the great space race. Russia didn't do half bad, and reading interviews from the cosmonauts from that day, they very much felt they were pursuing a dream and very much had their heart and spirits into the efforts. Concede the possibility that there are people in China that feel the same way. The difference here is that the people/government of China let them.
But I think we're on the same page here, at least, even if you don't think China is going to be the next big entity in space.
you are unwilling to see that NASA is one of those impediments to progress
No, I don't see NASA like that. I see the policies of the current administration to be like that. Policies can change without needing to scrap the entire agency. As we've described, NASA's policies have already changed slowly over the last 50 years or so, keeping pace with to focus on safety found most everywhere in the US. I contend that it's going to be easier for NASA than it's going to be for a private industry to emerge to do the same thing NASA does in that regard. Some ventures are just too risky with little chance for return. Investors don't like that, and that's why we have government sponsoring these programs. I don't see that changing in the near future.
faulted NASA for past mismanagement of safety concerns. But my real concern is that they spend billions upon billions of dollars and employ thousands of the brightest engineers and scientists, and still make some of the stupidest mistakes which cost lives and money, but most importantly time.
I get frustrated reading stuff like this all the time. Manned orbital activities are inherently risky. You're strapping people to a huge explosive machine with tons of cargo, sending them into orbit at speeds of tens of thousands of miles an hour, and bringing them back.
Nothing is risk free, activities like this least of all. At some point, people have to decide whether to spend a few million dollars here for a potential safety increase of 0.1%, or live with a certain level of risk and unknown. Agencies like NASA do not employ psychics, or limitless staff capable of simulating with 100% accuracy any potential scenario. They work with real-world engineers, real-world data, and real-world funding, and with that, try to do the best they can.
Occasionally these things add up in a bad way, and risk becomes reality. It's tragic, but preferable to having a program cost so much that it's paralyzed with safety concerns and we end up getting nothing past the drawing board.
Not running a single space program for a single space station. All our eggs in one basket, as it goes.
NASA does a hell of a lot more than operate the space shuttle and ISS. I'm going to assume that the various Mars and moon programs have been temporarily forgotten but you now realize the error of your statement. Sometimes it's hard to take note of the dozens of active NASA programs, since only one or two actually make headlines, but they're very much there and very much providing useful information.
Exploration of space is dangerous and will not survive safety concerns of collective action. Liken it to any human endeavor of significant unknown and danger and you will find it must be done by individuals. Individuals that have clarity of vision and certainty of purpose. It must be done by people, not by institutions or incorporations. People who know the risks, people that see the dangers, people that take the leap because they see the oppurtunity. People that learn and reason.
Exactly. People like those that gave their lives in the pursuit of NASA's goals. We need to recognize this and stop pissing on NASA because our hindsight is better than their foresight.
It's really getting to the point where we've seen the end to most any significant progress or monument in the western world. It's all far too risky now. We have safety agencies, codes and laws all there to prevent us from taking risks without an enormously expensive safety line. Do you think we're ever going to see another St. Louis arch constructed anywhere in the western world? Do we have any idea how much money we could save, how much additional science we could actually do, and how much more progress we could make if we decreased our safety margins just by a tiny bit?
This is one good thing that China has going for its space program. It recognizes that progress cannot occur without risk and is prepared to lose lives for that progress. This will allow it to compete with (and potentially surpass) the western world in this area.
In the real world, with salaried staff, you don't have these luxuries. There is no law requiring that all employees receive overtime beyond their 40-hour work week, nor are there laws requiring compensation for off-hours support (e.g. "on call"). You have every "right" to say "no" to your boss at home, provided you aren't breaking your contract with your employer, and provided you're willing to lose your job.
You might be able to say that the changes to your job aren't within the job description you signed up for, and you might be able to use that as a basis for renegotiating your salary, but if you're too unwilling to adapt to your company's needs, your company can pretty easily replace you with someone who is. If their demands have shifted to the point where your job description is no longer appropriate, they can pretty easily lay you off and create a new position better matching what they want.
Now, I'm not trying to suggest that companies are free to treat their folks like shit, but to suggest that what they're doing here is illegal is a bit inaccurate and doesn't really apply to salaried employees.
This seems like it could be an interesting technique for debugging systems as a whole. Like when my Windows PC hangs randomly for 30 seconds, or does something quirky, it'd be nice if I (or preferably, Microsoft) can understand *why* so that the problem can be corrected. Bit rot sucks.
However, if they were not legally obtained, McCalls has the right to do this; although they should not have used the DMCA, they have a legitimate clame to the patterns.
The patterns (as information), yes, as the copyright holders. But they have no claim to a tangible piece of property thrown in the trash.
Copyright or no, nothing gives me the right to exert control or ownership of a tangible storage medium merely because I hold the copyright on the information stored on it.
the owner of a particular copy or phonorecord lawfully made under this title, or any person authorized by such owner, is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy or phonorecord.
The copy is lawfully made, and discarded. Ownership at this point is decided by the "finder's keepers" doctrine.
As I've noted in other posts, there's a huge difference between making a decision to license your code, and redistributing "other" code using a given license.
Your interpretation is very dangerous and could kill Linux in the corporate marketplace: if a company is routinely contributing or releasing GPL'd software, all a former disgruntled employee (or a malicious kid with a score to settle) needs to do is hunt some secret/proprietary/damaging information, sneak it into the source, and wait for the company to re-release it.
To suggest that companies redistributing software must do an extensive audit to ensure that none of their IP made it into the source illegally is ludicrous. Some companies have millions of lines of code that could potentially be inserted in there. Do you really expect an army of coders to go through GPL'd source hunting for snippets or derivatives before that company can safely redistribute?
Again, a GPL violation is effectively retroactive. If copyrighted code gets dropped in illegally, then it's not legally licensed under the GPL. The rights granted by the GPL are effectively void for every release based upon that, including the release SCO received and redistributed. Their act of redistribution does not constitute an act of licensing of their code! They didn't know it was in there. That act of redistribution doesn't magically "restore" the GPL license on those that downloaded it. The copyright holder remains the copyright holder at all times. They did not grant GPL'ed rights to their IP, so they continue to maintain their rights.
I basically agree with what you say here, but as far as the GPL is concerned, you have to look at the entire work here. If one piece becomes "tainted", the entire work becomes undistributable. You can't just say "well this piece and that piece are under the GPL". You'd have to make your way back to a release that occurred before that code was inserted and work from a clean, untainted copy.
Now, in many respects, this could be identical to just picking out the non-SCO pieces, but keep in mind that some work by other parties could have been based, in part, on the wrongfully inserted code. These derivatives are also an infringement of copyright and would need to be rewritten. That's why you need to realistically go back so far. The GPL is clear in this respect: if a work is tainted by something (patent, copyright, whatever), you lose your rights to distribute under the GPL. This applies to everyone, for every release made since that time. Only rebuilding a new release from known GPL'd code would give you a clean GPL copy.
I do agree with most everything else you've said. I'm not trying to justify SCO's position here, I'm just trying to suggest an interpretation of the licence and the law that makes sense here. There's a lot of anti-SCO bias here and it's making this article a little annoying to read.
They did not release their code. They redistributed the Linux kernel, which wrongfully contained some of their IP (allegedly). The act of redistribution here does not constitute an act of licensing.
Do you have any idea the precedent that your interpretation sets? If I didn't like some company that was shipping a Linux distribution, all I have to do is get my hands on some of their proprietary, secret data, insert it into the Linux kernel while no one's looking, and as soon as they redistribute it, claim that they implicitly licensed it.
Remember here that I am the one performing the act of licensing here by sticking the code into the kernel. If I had no rights to do that, redistribution further down the line doesn't "correct" that.
What's to say the SCO themselves didn't put this mystery code into Linux, get it into the kernel, sit back for 18 months and then cry foul? Nothing.
That's what versioning and software configuration management is for. It should be trivial to identify who checked in the code in question.
Because of the GPL, they had a responsibility to be sure they weren't distributing their own IP
I disagree. Continuing under the assumption that someone's IP was wrongfully included in the Linux kernel source, the contributor of that source did not have a right to contribute it. Since he did not own the copyright, he had no legal right to attempt to license it under the GPL. The rights apparently granted by the GPL are void then for every release based upon that code.
The fact that the original holder of that IP then redistributed it further (again, unknowingly, until you can prove otherwise) doesn't matter. The code they received and redistributed is no longer technically GPL'd. Effectively, anything including or derived from that illegally contributed code is now bound by copyright law retroactive to the original contribution and no rights granted by the GPL apply anymore.
If they can support their case here (and I'm not saying they can), and it turns out that their code was indeed contributed to the Linux kernel without authorization, everything based on that code can no longer be redistributed. We basically end up subject to this portion of the GPL:
If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all.
It's not a matter of "ignorance". If we're going to start saying that the mere act of redistribution means you are implicitly releasing any of your IP (known or unknown), that is going to be a huge loss for Linux in the corporate world. You are now effectively forcing companies to do a tremendously expensive audit of GPL'd source prior to redistribution to be sure that none of their employees or 3rd parties have illegally and surreptitiously or accidentally placed some of their IP into the source. There's no way that would fly. That's an excessive burden to place on the redistributor.
Basically it boils down to the legality of the license once someone attempts to release code they don't own the copyright for. If they're not the holder, then they don't have the right to release it under the GPL, so nothing based off of that code can be legally released under the GPL, including the code that made its way back to the original copyright holder.
If a 3rd party infringed on someone's copyright by adding code to Linux and "releasing" that code under the GPL, that doesn't mean the code can be freely redistributed under the GPL. The person adding it to the Linux kernel did not do so legally, not being the copyright holder nor being granted a license by the copyright holder. He did not have the right to release it under the GPL.
Just because that passed into the hands of what may be the original copyright holder of some of that code, and that party redistributes it further, not realizing that their code made its way into the kernel improperly, doesn't mean that they have made the conscious decision to license their copyrighted works under the GPL. They did not add that code to the Linux kernel, someone else did. You can't very well hold them accountable for that release.
They are redistributing, not licensing their copyrighted works. Someone else has already done that without their knowledge or permission. Redistribution does not implicitly create a license/contract.
that's also the fault of the developers. You can't solve bad coding or protection with lawsuits
While I more or less agree with the rest of what you said, I'm not sure I completely agree with you here. This is like saying that someone exploiting a bug shouldn't be held accountable for the act. Really, when you think about it, nearly any compromise/exploit takes advantage of some bug or process problem somewhere. If there were no bugs, there would be nothing to hack, right?
Sure, acknowledge the bug and fix it, but by all means go after the guy that took advantage of it. He could have simply informed them that the bug existed, after all, but he chose instead to cause trouble. He should pay for that.
No, because US government controls violating its citizens' rights to free speech are illegal.
Government controls that mildly impact their ability to perfectly match colors in their printing/copying do not violate their right to free speech; they simply give a slight push to invest in foreign technologies without these controls for precision printing/copying needs.
Please don't jump to these "What's next?" evil government scenarios without asking yourself if your theory is really sensical. There is tremendous legal precedent that such controls here would never be allowed to happen (without a fundamental change in US law), even if common sense doesn't do the trick.
If you have questions or concerns about your computer equipment, take them up with the manufacturer. If you feel that your equipment isn't functioning as you need it to, and these features are both detectable and unadvertised, you might even have grounds to get your money back. Buy a product not encumbered like this (perhaps from another vendor).
Slashdot Mirror
Yeah things like this have been made to work under controlled conditions,
What are you talking about? These technologies (not this one, but similar ones) are in my car right now and occasionally I enter less-than-controlled conditions. Hydroplaning and other less-than-optimal road conditions are common and that's what some of these technologies are there for.
AGAIN: Car manufacturers are not going to add a "safety" feature that causes more accidents than it prevents. Based on your posts, I have to believe that this is what you think.
Sure you can write software that tries to follow a line, but will it be safe?
If implemented correctly, yes. If it applies minor adjustments to compensate for a slow drift, yes, I believe this can be used safely. A system like this would not have the ability to spin the steering wheel around to the right and slam you into the car next to you.
The point is this safety technology is not safe.
And I submit to you that it is (and will be) safe. If statistics get published that indicate these features cause 10 times the accidents they would otherwise prevent, I think someone would have an awful lot of explaining to do.
but allowing it to hit the brakes is crazy.
Some higher-end cars automatically decellerate when they detect a critical engine problem. They've been doing this for years. Maybe car manufacturers have found out how to do this safely already?
the last thing you want is for something to unexpectedly slam on the brakes, period.
Who said anything about slamming on the brakes? You're exaggerating to try and make your point valid. This device lightly applies brakes. Yes, I suppose in a frictionless environment, your wheels will (arguably, given things like traction control) cease to maintain traction on the road. But are you worse off at this point? If lightly hitting the brakes is enough to do this, how would you have avoided the car in front of you anyway?
What's frustrating to me is that you're not even willing to accept that Honda has thought of these situations. Do you believe that your experience with RADAR makes you more qualified to speak about these technologies than the guys actively developing it? Honda has a bit of an interest in keeping its drivers alive. I'm perfectly willing to concede that implementing something like this is going to be tricky. But I'm also going to trust that Honda has thought of all of the ways that this could be dangerous and has built something that addresses most (if not all) of those concerns. An on/off button that could be used in adverse conditions should satisfy a lot of these concerns right up front. The next time I'm following a line of cars across a frozen lake, I might want to turn this off. (Actually, I probably wouldn't, since this device, in combination with traction control, could still likely do a better job than I could if I'm not paying attention to the point where this is activated.)
Perhaps you're right, in that there will be a 1% increase in accidents resulting from an unexpected activation of this device. But if this can reduce rear-end collisions by even 10%, wouldn't that be worth it? And if you don't think so, you should have the ability to turn this off (or not ask for it when you buy your next car).
I'm not asking you to say that this technology is safe, I'm just asking you to consider the possibility that Honda has thought of everything you've thought of here, and that your concerns have been addressed in some fashion.
Dumbass.
Let the record show that you're the one that started name-calling here, not me.
Yeah, I must be stupid for actually understanding the limits of today's technology.
You don't appear to, or perhaps you are misunderstanding how that technology is being applied in this situation.
You clearly don't
Whatever, man. The bottom line is that you do not have the specifics here and are pulling shit out of your ass as a result. You may label it highly educated, RADAR engineer shit, but it's still shit. I asked you to consider the possibility that they've done what they claim to have done, and you've rejected that. Sounds pretty closed-minded to me.
Spoken like someone who doesn't understand how to drive in the snow.
I learned to drive in a Ford Escort in the dead of winter in Anchorage, Alaska. Here again, you are armed with no knowledge of something and pull more shit out of your ass.
Driving on a sheet of ice, one can indeed safely apply brakes to slow down. In the event you apply them too hard, additional technology such as skid prevention would step in to prevent my braking from causing me to lose control. How do I know this? Because I have done it.
Are you just unaware of these technologies? Perhaps that's why this "next step" into the world of RADAR seems so strange and impossible to you. It doesn't sound like you've fully come to terms with the stuff that's already out there.
The only info this system would have, would be the range, and amplitude of the reflection. That's it.
Even with one dimension and 16 degrees of arc (from which I might argue a two-dimensional view might be obtained, but since I have no specifics we can just assume it's a single range measurement covering an area of 16 degrees), an image zipping in and out of its field of view still looks considerably different than a car in your lane in front of you (that's probably been there for more than a few seconds). Remember that this thing is intended to prevent rear-end collisions, not random collisions with any object in the road or passing in front of you. Even with a simple/dumb range finder, it's pretty easy to see when the car in front of you is slowing down or stopping suddenly as opposed to seeing a bird or some other object move in front of it.
But still (and again), please consider the possibility that the people developing these technologies might have someone on staff that could come up with these scenarios just as easily as you could.
Let me repeat myself: Car manufacturers are not going to add technology to a car that will cause it to have an accident once a week.
I simply don't understand how you are coming back to the conclusion that what they're doing is impossible. THEY'VE DONE IT! This has been in testing for years. I sincerely doubt you've driven in one of these things or been given a demonstration of the technology or the behaviors that govern its actions. Again, please consider the possibility that your concerns have been addressed and stop going off about how much you know about RADAR and how they must be lying. This whole thread is incredibly silly and you're really starting to look like an ass.
P.S., if this piece of tech has already got you believing Honda is in some fantasy land, note that this model also has image recognition software designed to make small corrections to help keep a vehicle in its lane on the highway. Feel free to say this is impossible as well.
What is with all of these pessimistic attitudes?
Can we please concede the possibility that car manufacturers will overcome some of the most basic issues you address before putting this into their cars? Nobody is going to drive a car that averages one accident a week. Think about this.
What about snow, mud, gravel (insert low traction condition here)?
If you are driving in conditions that light braking will send your car out of control, and the car is unable to detect and respond to this through other mechanisms (steering control, skid control, ABS, whatever), you are already fucked.
What if you're being tailgated?
I might accept this as an exceptional case. But even so, the light braking that this device does shouldn't be sufficient to cause a tailgater to crash into you. And even if it does, it's his fault. Chances are, if this system is activated and decides to apply some brakes, you're already in a high-risk situation (imminent accident) where this additional risk is probably acceptable.
But given that we have so little information about how these devices make their decisions, this is all possible perceived problems and conjecture pulled from our collective asses, yes?
What if you're across from someone in a left turn lane?
I would be shocked if this common situation were not addressed before this device is rolled out.
What if you're speeding up to try an avoid an accident?
What if the technology noticed this?
What if the system malfunctions?
What if ABS malfunctions? What if dynamic steering or skid control malfunctions? Which technology poses the greater risk here in the event of malfunction?
What if a bird flys in front of your car?
The radar profile of a bird (size, speed, likelyhood of collision) is probably significantly different than a large obstacle or a car. I wouldn't be surprised if a bird was ignored entirely. Even if it wasn't, there would be a tiny fraction of a second where the computer would like to see you slow down a bit. That might not be enough time for it to actually act on that. *shrug*
What if you're making a sharp turn, and any braking will cause you to loose control?
Again, I would honestly believe that a car manufacturer would look at these scenarios and come up with something that does not kill its passengers so easily. Additionally, most cars have features nowadays that will compensate for such actions. If it feels the car losing control, it will adjust your braking power or steering to compensate. So the logic is there. How well it integrates with some feature like this (by reducing or eliminating the braking, for example) isn't something I can speak to.
As smart as you think you are, please consider the likelyhood that the people designing these things might understand the issues better than you.
A directory is a search engine for proper names. There will be a finite number of organizations doing business as "Example Widgets". If the scope of my search is nation-wide, and two or three hits come up with "Example Widgets" incorporated in a few local markets, then I guess I need to do some more research to identify which one I want. Additional fields in the record would allow me to determine which Example Widgets is the one I am searching for, such as the home city.
There isn't a "placement" issue because these aren't really content hits. I'm not searching for a product here and getting a list of organizations that provide that product. I'm searching for an organization. If I choose the wrong one because I don't have enough information to pick between the matches, I'm probably not going to end up where I want to be. This is different than a company trying to be at the top of search rankings because it wishes to get the business of someone doing the search. In that case, I'm searching for the provider of a service. In the case where this directory is most useful, I am searching for a specific organization.
Similarly, a more traditional (but still regulated) search engine would allow me to actually search for product names or other service marks, and have that associated with the organization that owns it (perhaps with a URL). In this case as well, certain marks may be scoped to certain markets (countries?). Similar "placement" rules would have to apply for those rare situations. This could be as simple as a multi-column display with each column containing some 'sort by' mechanics. This isn't any different from any other directory browser we use today.
Together, the two options above would satisfy the 'locator' abuse of DNS, where I need to find a specific company, and assume a 'www' and a 'dot-com' to get there, as well as the 'product search' abuse of DNS, where I need to find a specific product, and assume a 'www' and a 'dot-com' to get there as well. With those qualities removed from DNS, nearly all intellectual property weight is removed and DNS can become a simple hierarchical label again.
Lastly, more traditional search engines would be able to associate generic product terms with more specific products that may satisfy the requirement you're hunting for. We don't care at that point how those are presented or ordered. Traditional search engines or global product catalogs have those issues covered.
I would agree with you, except search engines carry no authority. You could easily manipulate your own pages and things to make search engines "think" that your pages are more relevant for a given topic (or company name) than the company itself.
An authoritative search engine might be a good place to start worrying about trademark and service marks, though.
You still need that central authority if you need the ability for your labels to carry IP weight, but these labels must be scoped within political boundaries for that to be feasible.
DNS could actually still function in that capacity through the elimination of the GTLDs, and enforcement of location-specific hierarchies.
Some of these ideas are covered in a small paper I wrote last year at http://fastolfe.net/features/directory-service/.
We need to start getting people away from using DNS as a locator service. DNS is meant (mainly) to apply convenient names to IP addresses, not so that my mom can guess "www.example.com" when she wants to know something about "example".
DNS is inappropriate for this because it does not allow two parties with a legitimate interest in the same label to share it, except through the confusion of additional TLDs. Two parties could have a perfectly legal claim to a label that they now have to battle out in courts. A better solution would be to create a directory on top of DNS to map these real-world names to DNS domains. Put all of the trademark and IP crap you want into regulating this directory, but leave it out of DNS.
LDAP or X.500 could do this nicely, with the added benefit of allowing distinguished names to include information about the locality (thus legal jurisdiction). This limits the scope of an IP conflict to the jurisdiction in question.
If I want to load the home page for Example Widgets, I should just need to hunt for the common name "Example Widgets", pick the match that makes the most sense (assuming more than one match is found), and have my lookup return a DNS domain name. My browser would consult DNS for a SRV record associated with that domain, find a host providing HTTP services for it, and request the home page from that host. At no point should my mom ever need to be exposed to a hostname (or a URL for that matter).
My two cents, anyway.
Don't confuse the official tolerances for deducting points from your license with "soft" speed limits. On the MIT page (which is what I assume you're talking about), there is a section labeled "Official tolerances" that discusses the subject of points against your license only. They are not suggesting that cops are not within their rights to bust you for going 61 in a 60.
Sometimes you can "get away" with speeding if you're within certain margins of error (e.g. radar), but these margins are most certainly not codified in law and would require some convincing argument in court.
The bottom line: most states consider their speed limits "absolute" in all respects. Some of them allow their lower speed limits to be "soft" if you can demonstrate you were travelling safely even though you were speeding, until you hit the state's absolute limit, which is absolute. Three states even consider their statewide speed limit to be "soft".
(ps. There are more to articles than pretty pictures and charts.)
Don't be an ass. Please concede the possibility that you may be wrong before you go off insulting people who disagree with you.
12v * 3 = 36v, 14v * 3 = 42v. It's a convenience factor for dealing with established 12v systems. Both the battery and charging (alternator) voltages are 3 times larger.
Did you actually read the links you posted for us? It looks to me like 3/4 of the states use absolute limits. When you exceed the posted speed limit, you are guilty of speeding. In fact, only 3 states seem have universally "soft" limits that you can contest if you can prove you were exceeding that limit safely.
So no, your assertion that "in most states the speed limit is a suggested value" does not appear to be accurate.
So don't sign your name to it.
I personally have a perfect driving record, but there's no way in hell I'd voluntarily turn this feature off. If I get into an accident, I want my accident to help others in any way possible. I don't understand why this 5-second thing is any sort of invasion of privacy. The only time the information could be used is after I've destroyed my car (set off the airbags). The only time the information would be used is when the act of destroying my car created a situation where the police (or myself) would want the information analyzed. If the police are after it, they need to convince a judge to get a search warrant. At that point, any "privacy rights" I think I have are out the window anyway.
NASA is a public institution and the public and publicity surrounding the space program demand it.
I think we are in agreement here.
That is why I think it should be a private endevour.
I might suggest that NASA (and perhaps the public) try to understand why certain levels of risk are acceptable and adapt. I'm not sure that a private endeavor is going to fare any better, given the litigation-happy state of the US. Perhaps another country would be better to host this.
Various mars and moon programs? The moon programs were 30 years ago.
Lunar Prospector was a few years ago. Remember all of the hype about finding water on the moon?
get lucky and land in an interesting place or you could land in the middle of the Sahara.
Not to let us drift too far, but bear in mind that in some respects, we have better maps and imagery of the moon and Mars than we do of many places on earth. If we were investigating the earth, finding an interesting spot wouldn't be very difficult. But given the lack of anything obvious from a distance, I suppose you have a point there. I disagree that giving up before getting started is the right decision, though.
It is the unbridled human spirit and mind, which will find a way to move forward.
This isn't very educated. You're suggesting that since the US promotes some form of freedom, that other governments cannot possibly produce people that feel they're pursuing a dream by going into space?
China isn't that different from Russia back during the great space race. Russia didn't do half bad, and reading interviews from the cosmonauts from that day, they very much felt they were pursuing a dream and very much had their heart and spirits into the efforts. Concede the possibility that there are people in China that feel the same way. The difference here is that the people/government of China let them.
But I think we're on the same page here, at least, even if you don't think China is going to be the next big entity in space.
you are unwilling to see that NASA is one of those impediments to progress
No, I don't see NASA like that. I see the policies of the current administration to be like that. Policies can change without needing to scrap the entire agency. As we've described, NASA's policies have already changed slowly over the last 50 years or so, keeping pace with to focus on safety found most everywhere in the US. I contend that it's going to be easier for NASA than it's going to be for a private industry to emerge to do the same thing NASA does in that regard. Some ventures are just too risky with little chance for return. Investors don't like that, and that's why we have government sponsoring these programs. I don't see that changing in the near future.
faulted NASA for past mismanagement of safety concerns. But my real concern is that they spend billions upon billions of dollars and employ thousands of the brightest engineers and scientists, and still make some of the stupidest mistakes which cost lives and money, but most importantly time.
I get frustrated reading stuff like this all the time. Manned orbital activities are inherently risky. You're strapping people to a huge explosive machine with tons of cargo, sending them into orbit at speeds of tens of thousands of miles an hour, and bringing them back.
Nothing is risk free, activities like this least of all. At some point, people have to decide whether to spend a few million dollars here for a potential safety increase of 0.1%, or live with a certain level of risk and unknown. Agencies like NASA do not employ psychics, or limitless staff capable of simulating with 100% accuracy any potential scenario. They work with real-world engineers, real-world data, and real-world funding, and with that, try to do the best they can.
Occasionally these things add up in a bad way, and risk becomes reality. It's tragic, but preferable to having a program cost so much that it's paralyzed with safety concerns and we end up getting nothing past the drawing board.
Not running a single space program for a single space station. All our eggs in one basket, as it goes.
NASA does a hell of a lot more than operate the space shuttle and ISS. I'm going to assume that the various Mars and moon programs have been temporarily forgotten but you now realize the error of your statement. Sometimes it's hard to take note of the dozens of active NASA programs, since only one or two actually make headlines, but they're very much there and very much providing useful information.
Exploration of space is dangerous and will not survive safety concerns of collective action. Liken it to any human endeavor of significant unknown and danger and you will find it must be done by individuals. Individuals that have clarity of vision and certainty of purpose. It must be done by people, not by institutions or incorporations. People who know the risks, people that see the dangers, people that take the leap because they see the oppurtunity. People that learn and reason.
Exactly. People like those that gave their lives in the pursuit of NASA's goals. We need to recognize this and stop pissing on NASA because our hindsight is better than their foresight.
It's really getting to the point where we've seen the end to most any significant progress or monument in the western world. It's all far too risky now. We have safety agencies, codes and laws all there to prevent us from taking risks without an enormously expensive safety line. Do you think we're ever going to see another St. Louis arch constructed anywhere in the western world? Do we have any idea how much money we could save, how much additional science we could actually do, and how much more progress we could make if we decreased our safety margins just by a tiny bit?
This is one good thing that China has going for its space program. It recognizes that progress cannot occur without risk and is prepared to lose lives for that progress. This will allow it to compete with (and potentially surpass) the western world in this area.
You seem to be stuck in an hourly-wage world.
In the real world, with salaried staff, you don't have these luxuries. There is no law requiring that all employees receive overtime beyond their 40-hour work week, nor are there laws requiring compensation for off-hours support (e.g. "on call"). You have every "right" to say "no" to your boss at home, provided you aren't breaking your contract with your employer, and provided you're willing to lose your job.
You might be able to say that the changes to your job aren't within the job description you signed up for, and you might be able to use that as a basis for renegotiating your salary, but if you're too unwilling to adapt to your company's needs, your company can pretty easily replace you with someone who is. If their demands have shifted to the point where your job description is no longer appropriate, they can pretty easily lay you off and create a new position better matching what they want.
Now, I'm not trying to suggest that companies are free to treat their folks like shit, but to suggest that what they're doing here is illegal is a bit inaccurate and doesn't really apply to salaried employees.
This seems like it could be an interesting technique for debugging systems as a whole. Like when my Windows PC hangs randomly for 30 seconds, or does something quirky, it'd be nice if I (or preferably, Microsoft) can understand *why* so that the problem can be corrected. Bit rot sucks.
The patterns (as information), yes, as the copyright holders. But they have no claim to a tangible piece of property thrown in the trash.
Copyright or no, nothing gives me the right to exert control or ownership of a tangible storage medium merely because I hold the copyright on the information stored on it.
From http://www.copyright.gov/title17/92chap1.html#109
As I've noted in other posts, there's a huge difference between making a decision to license your code, and redistributing "other" code using a given license.
Your interpretation is very dangerous and could kill Linux in the corporate marketplace: if a company is routinely contributing or releasing GPL'd software, all a former disgruntled employee (or a malicious kid with a score to settle) needs to do is hunt some secret/proprietary/damaging information, sneak it into the source, and wait for the company to re-release it.
To suggest that companies redistributing software must do an extensive audit to ensure that none of their IP made it into the source illegally is ludicrous. Some companies have millions of lines of code that could potentially be inserted in there. Do you really expect an army of coders to go through GPL'd source hunting for snippets or derivatives before that company can safely redistribute?
Again, a GPL violation is effectively retroactive. If copyrighted code gets dropped in illegally, then it's not legally licensed under the GPL. The rights granted by the GPL are effectively void for every release based upon that, including the release SCO received and redistributed. Their act of redistribution does not constitute an act of licensing of their code! They didn't know it was in there. That act of redistribution doesn't magically "restore" the GPL license on those that downloaded it. The copyright holder remains the copyright holder at all times. They did not grant GPL'ed rights to their IP, so they continue to maintain their rights.
I basically agree with what you say here, but as far as the GPL is concerned, you have to look at the entire work here. If one piece becomes "tainted", the entire work becomes undistributable. You can't just say "well this piece and that piece are under the GPL". You'd have to make your way back to a release that occurred before that code was inserted and work from a clean, untainted copy.
Now, in many respects, this could be identical to just picking out the non-SCO pieces, but keep in mind that some work by other parties could have been based, in part, on the wrongfully inserted code. These derivatives are also an infringement of copyright and would need to be rewritten. That's why you need to realistically go back so far. The GPL is clear in this respect: if a work is tainted by something (patent, copyright, whatever), you lose your rights to distribute under the GPL. This applies to everyone, for every release made since that time. Only rebuilding a new release from known GPL'd code would give you a clean GPL copy.
I do agree with most everything else you've said. I'm not trying to justify SCO's position here, I'm just trying to suggest an interpretation of the licence and the law that makes sense here. There's a lot of anti-SCO bias here and it's making this article a little annoying to read.
They did not release their code. They redistributed the Linux kernel, which wrongfully contained some of their IP (allegedly). The act of redistribution here does not constitute an act of licensing.
Do you have any idea the precedent that your interpretation sets? If I didn't like some company that was shipping a Linux distribution, all I have to do is get my hands on some of their proprietary, secret data, insert it into the Linux kernel while no one's looking, and as soon as they redistribute it, claim that they implicitly licensed it.
Remember here that I am the one performing the act of licensing here by sticking the code into the kernel. If I had no rights to do that, redistribution further down the line doesn't "correct" that.
That's what versioning and software configuration management is for. It should be trivial to identify who checked in the code in question.
Because of the GPL, they had a responsibility to be sure they weren't distributing their own IP
I disagree. Continuing under the assumption that someone's IP was wrongfully included in the Linux kernel source, the contributor of that source did not have a right to contribute it. Since he did not own the copyright, he had no legal right to attempt to license it under the GPL. The rights apparently granted by the GPL are void then for every release based upon that code.
The fact that the original holder of that IP then redistributed it further (again, unknowingly, until you can prove otherwise) doesn't matter. The code they received and redistributed is no longer technically GPL'd. Effectively, anything including or derived from that illegally contributed code is now bound by copyright law retroactive to the original contribution and no rights granted by the GPL apply anymore.
If they can support their case here (and I'm not saying they can), and it turns out that their code was indeed contributed to the Linux kernel without authorization, everything based on that code can no longer be redistributed. We basically end up subject to this portion of the GPL:It's not a matter of "ignorance". If we're going to start saying that the mere act of redistribution means you are implicitly releasing any of your IP (known or unknown), that is going to be a huge loss for Linux in the corporate world. You are now effectively forcing companies to do a tremendously expensive audit of GPL'd source prior to redistribution to be sure that none of their employees or 3rd parties have illegally and surreptitiously or accidentally placed some of their IP into the source. There's no way that would fly. That's an excessive burden to place on the redistributor.
Basically it boils down to the legality of the license once someone attempts to release code they don't own the copyright for. If they're not the holder, then they don't have the right to release it under the GPL, so nothing based off of that code can be legally released under the GPL, including the code that made its way back to the original copyright holder.
I'm not sure I share this interpretation.
If a 3rd party infringed on someone's copyright by adding code to Linux and "releasing" that code under the GPL, that doesn't mean the code can be freely redistributed under the GPL. The person adding it to the Linux kernel did not do so legally, not being the copyright holder nor being granted a license by the copyright holder. He did not have the right to release it under the GPL.
Just because that passed into the hands of what may be the original copyright holder of some of that code, and that party redistributes it further, not realizing that their code made its way into the kernel improperly, doesn't mean that they have made the conscious decision to license their copyrighted works under the GPL. They did not add that code to the Linux kernel, someone else did. You can't very well hold them accountable for that release.
They are redistributing, not licensing their copyrighted works. Someone else has already done that without their knowledge or permission. Redistribution does not implicitly create a license/contract.
These are pretty basic questions that have already been figured out. A quick Google search brings up this little FAQ that you might find interesting: http://ntrg.cs.tcd.ie/mepeirce/Project/Mlists/mini faq.html
How will subdividing a finite address space into subnets give you infinite addresses?
that's also the fault of the developers. You can't solve bad coding or protection with lawsuits
While I more or less agree with the rest of what you said, I'm not sure I completely agree with you here. This is like saying that someone exploiting a bug shouldn't be held accountable for the act. Really, when you think about it, nearly any compromise/exploit takes advantage of some bug or process problem somewhere. If there were no bugs, there would be nothing to hack, right?
Sure, acknowledge the bug and fix it, but by all means go after the guy that took advantage of it. He could have simply informed them that the bug existed, after all, but he chose instead to cause trouble. He should pay for that.
No, because US government controls violating its citizens' rights to free speech are illegal.
Government controls that mildly impact their ability to perfectly match colors in their printing/copying do not violate their right to free speech; they simply give a slight push to invest in foreign technologies without these controls for precision printing/copying needs.
Please don't jump to these "What's next?" evil government scenarios without asking yourself if your theory is really sensical. There is tremendous legal precedent that such controls here would never be allowed to happen (without a fundamental change in US law), even if common sense doesn't do the trick.
If you have questions or concerns about your computer equipment, take them up with the manufacturer. If you feel that your equipment isn't functioning as you need it to, and these features are both detectable and unadvertised, you might even have grounds to get your money back. Buy a product not encumbered like this (perhaps from another vendor).