I was actually more impressed by the acknowledgement by the Catholic church that condoms were perhaps not a bad idea, after all.
I don't know about that, but they *did* make an announcement last year that HIV could pass through the rubber of condoms. Which is pretty funny when you consider that condoms make fine balloons, and the relative sizes of O_2 molecules and viruses. Still, what do you expect from people who believe in fairies?
Documentation volunteers aren't given the same peer recognition and respect as programmers... If you don't submit code patches to a project, you're the invisible man!
There is some truth to this, but I think it's a bit harsh. The truth is that very many good contributors whether they be writers, testers or programmers never get the wide recognition they deserve. There are a dozen Samba committers and probably nearly a hundred working on GNOME, but it's only the high-profile leaders who tend to get interviewed on slashdot... and yes, they tend to be developers. Perhaps the leaders deserve the most credit, but it takes a village, or so they say...
I guess that is just the way that "fame" works in general: it hits a small number of people semi-randomly.
But not being famous doesn't mean you're not appreciated. I've written thankyou emails to the authors of particularly good free manuals or documents. I encourage everyone to do the same.
Kicking out even FSF's documentation, on the grounds that is doesn't meet Debian's criterion of freeness, was a really stupid thing to do.
OK, with you so far...
Writing software is its own reward for a lot of people. Writing docs is a vital chore which nobody likes and which gets little recognition.
Sorry, but this is completely bogus. We went through this argument a few years ago: "people won't write nice GUIs for fun", "people won't write open-source databases", "people won't release open source kernels for big iron". Now it may well be that there is not enough good open documentation on some particular topics at the moment, but that is no reason to assume that it is not worth trying to get free documentation.
The FSF docs are free enough for all practical purposes.
They are free enough for many purposes, but the licence is buggy and needs to be fixed. Some projects such as Arch have gone back to the GPL for documentation, which is the sensible choice at the moment. Both FSF and Debian need to stop fucking around.
Well, I found the thread pretty interesting. Thanks for your thoughts.
I agree that having standard crash tests at higher speeds would be an excellent idea. Some tests are now done at 64kmh/40mph. Even if the government won't run them, I would like to see some manufacturers take the initiative and publish tests for 60MPH crashes.
I agree it would be good to see more non-painted bumpers to make low speed bumps cheaper. I was noticing on the way to work this morning how much more common they were on older cars. I suppose this is a kind of consumer fashion thing. According to the IIHS report there used to be a federal standard requiring no damage at 5mph; it's a shame that is not still required.
if an airbag isn't enough to keep your ribs from being crushed, I can't imagine the tiny benefit from crumple zones changing anything
It may not seem intuitive but I think having extra distance to decelerate makes an enormous difference. Crumpling the hood probably gives you at least a meter, whereas the seatbelt and airbag allow less than a meter of deceleration.
Assuming a linear deceleration, doubling the cushioning distance will halve the deceleration force on the passengers. Chance of injury increases super-linearly with force. Having a hood that crumples is likely to more than halve the chance of serious injury.
I think the point of a rigid passenger cell is to prevent intrusion of impacting objects or car parts. It doesn't absorb energy. The most a rigid cell can do is turn the impact from falling onto a spike into falling onto a flat surface. At 30m/s, even the flat surface is going to hurt.
I'm sure there is lots of room for improvement. I just don't think more rigid == better by definition.
No, the question is, with a trivial design improvement, could the survival rate be even better?
I would guess not: there is a reasonably competitive market for cars. If there were *trivial* changes that could make safety much better, then you would expect it to happen, at least in luxury cars. I can believe that there are changes that are tough tradeoffs or hard to get right or expensive.
I would, in fact, be willing to bet that older cars were safer in roll-overs
That is a pretty interesting question. In principle you could answer it from statistics, though in a little bit of web research I can't find them. Since rollovers seem responsible for a fair fraction of fatalities you would expect there to be a fair selective pressure to improve handling of them.
Since mass-market cars now travel at 75+mph the domain over which they are required to be safe is wider than it once was. So I can believe maybe they get more damaged in a 35mph rollover as a tradeoff for surviving a 75mph rollover.
I have been making the point that we would be much safer if the body around the passenger area was reinforced until it's at least as strong as older cars were. You can maintain your crumple zones, and air bags, but I really want the passenger compartments to be improved.
That's a pretty interesting question, which I am not qualified to answer but will still speculate about.
I think given the energies involved, having just interior padding and airbags would not be enough to decelerate occupants at a reasonable rate.
One comparison that springs to mind is motorcycle helmets: you can buy USD700 helmets pretty close to what racers use, and I think they're probably the best available technology. They layer elements of different strengths but even the hardest outer glass/carbon fiber element is going to abrade, fracture and flex in a heavy crash.
The point is to decelerate the occupants as gently as possible. I think you want to design the system so that in a maximal crash, the protection is maximally deformed without actually coming apart or squashing the occupants. I don't think rigid elements directly help with that. I can imagine that having parts that flex or collapse up to a stronger breaking point would be good.
You obviously haven't spent much time on old cars. Bumpers are great things, and used to be able to handle very powerful forces. It's a molded piece of solid steel, that is connected, through other strong pieces, to the frame of your car. At 15MPH, you'd expect to dent the bumper to the point you'd want to have it replaced, but it would be a very inexpensive replacement.
No, I have to admit I have not. I would still be surprised if all the damage was confined to the bumper, but you may be right.
BTW, a 100MPH crash isn't all that incredible.
I said "much much more than 100MPH", but anyhow: it's 3 times faster than the NCAP tests. I think it would have been nearly unheard of to survive it in the cars of 25 years ago.
Have a crash test with an older car, and you'll see that the car will stay completely in-tact.
And the passengers may well be killed by hitting the dash or steering wheel at nearly full speed. Try dropping an egg in a metal box vs a cardboard egg carton.
I couldn't care less whether the car stays intact. The question is whether the chances of people surviving are greater or less than they used to be. From what I've read, they are greater, but if you have a different reference I'd be interested to see it.
Assuming I got it right the second time: A collision at that speed is equivalent to falling 2.3m onto concrete. I'd be surprised if any bumper could sustain that without damaging the car itself.
I'd be pretty leery of riding in a car during such an experiment. Even at such a relatively low speed compared to normal traffic, I'd be pretty happy for any protection I could get by having airbags go off or the car frame get fucked up.
While I have my dx/dt out: a head-on crash from 100kmh is like falling 40m. And 100kmh can seem so slow..
Well, the article starts out talking about a kid completely rolling a car, which is definitely in the domain of caring about survival, not repair costs.
I agree it would be nice if we could get replaceable rubber-style bumpers to protect against carpark accidents. But 15MPH is not so slow. You'd need a pretty large bumper to absorb it with no other damage, if you could do it at all. 6.7 m/s is equivalent to dropping the car headfirst onto concrete from about 900mm; you'd expect it to make a pretty loud noise.
And additionally, no car I've seen made in the last couple decades has a strong structure.
Really *none*? What are you comparing too?
I've seen a photo of a BMW M3 that flipped on a racetrack at >>100mph. The engine compartment is nearly broken off, but the passenger cell is intact. I'd call that strong -- and strong in the right place, too. No human has a right to survive those kind of energies, but the driver was reported to have walked away.
There was a 6-car low-speed rear-ender outside my office a while ago. It was a pretty interesting demonstration of different construction models: the Kia mini-car folded up by more than a foot and the Mercedes just dinged its bumper. (Possibly the bumper would be more expensive.:-)
Some people suggest that softer bumpers reduce damage to pedestrians. 5-10mph probably won't hurt the occupants, but it's pretty tough on a pedestrian.
The major thrust of car safety design for the last 10-20 years has been that the car should be written off to protect the occupants. Therefore: airbags, crumple zones, seat-belt pre-tensioners and tension limiters, collapsing steering columns, staged failure of structural elements...
The BMW quoted in the article performed very well: the occupant was uninjured, the passenger cell was not breached. The damage looks minor specifically because the structural components are meant to be hardest and fail last. In that minute, the owner spent $30k to prevent their child from being killed or paralyzed. (Whether they still think that was a good deal is another question...)
If they'd been driving an older car, it might well have been repairable after a rollover: more steel, more parts that bend plastically rather than breaking or crumpling, no airbags. On the other hand, if they'd been driving an older car, they might have been dead.
It's pretty simple: if you don't want to write off your car, don't flip it over!
"A good landing is one you walk away from. A perfect landing is where they can use the plane again."
Till then you are just making excuses for petty criminality.
Saying that civil disobediance of a supposedly unjust law is criminal begs the question. Jesus was a criminal, under the laws of his time. The point is whether the law is moral/reasonable/just to begin with.
I have a flash-based MP3 player that doesn't play AAC. If I used iTunes I'd feel it was fair use and morally defensible to transcode songs so I could play them on my Nuvo. I think the law ought to protect this fair use, rather than banning it.
Turn about: yes, I think it's reasonable for someone to make backups of my proprietary software, or to transfer it when they buy a new machine.
The US has done no such thing, and doesn't really have any legal basis for doing so... corporations have free speech rights here.
Don't be silly. US free speech rights are not unlimited, for either natural persons or corporations. One of the US judges already told SCO to tone down their public statements of Linux and IBM, and they seem to have complied.
The particular balance is different between the US and Germany, but both of them have laws against libel.
I heartily agree. (Mod parent up!) Repeatedly using the word "fundamental" to mean "major" makes her look a bit illiterate. It is nearly as annoying as people using "exponential" when they just mean "rapid".
I also think it's pretty damn funny for her to quoteESR describing hackers as "unfortunately intolerant and bigoted".
It's also kind of amusing for a pseudo-socialist to talk about religious blindness causing people to ignore working solutions.
This kind of rant has occurred so many times before: the bits that were good weren't original, and the bits that were original weren't good. In fact the bad bits aren't particularly original either.
In conclusion, to Michelle: shut up and code. (Or write docs or file UI bugs, that's good too. Stop whining.)
Or just install SCons. Great parallelism, smarter rebuilds, cross platforms, more reliable than make, and far faster than automake/libtool. Works well with distcc too.
run _everything_ in parallel
What, even things that shouldn't be parallel? Screw that.
Damned if I'm letting an electric cloud near my machine room.
Those numbers are weighted by the number of google queries the user does. People with modern machines are more likely to have fast and always-on connections; people with old machines are more likely to have modems.
If you have to wait 15s for the results, as you might on a modem line, you're likely to do far fewer queries per day. Even at the same level of user experience I would think old-machine users make far fewer queries...
Slashdot Mirror
I was actually more impressed by the acknowledgement by the Catholic church that condoms were perhaps not a bad idea, after all.
I don't know about that, but they *did* make an announcement last year that HIV could pass through the rubber of condoms. Which is pretty funny when you consider that condoms make fine balloons, and the relative sizes of O_2 molecules and viruses. Still, what do you expect from people who believe in fairies?
Damn. I just got my grandmother comfortable saying that I work on "Leenuks". Now she has to learn "gunookfreebisd". Fat chance!
Documentation volunteers aren't given the same peer recognition and respect as programmers... If you don't submit code patches to a project, you're the invisible man!
There is some truth to this, but I think it's a bit harsh. The truth is that very many good contributors whether they be writers, testers or programmers never get the wide recognition they deserve. There are a dozen Samba committers and probably nearly a hundred working on GNOME, but it's only the high-profile leaders who tend to get interviewed on slashdot... and yes, they tend to be developers. Perhaps the leaders deserve the most credit, but it takes a village, or so they say...
I guess that is just the way that "fame" works in general: it hits a small number of people semi-randomly.
But not being famous doesn't mean you're not appreciated. I've written thankyou emails to the authors of particularly good free manuals or documents. I encourage everyone to do the same.
Kicking out even FSF's documentation, on the grounds that is doesn't meet Debian's criterion of freeness, was a really stupid thing to do.
OK, with you so far...
Writing software is its own reward for a lot of people. Writing docs is a vital chore which nobody likes and which gets little recognition.
Sorry, but this is completely bogus. We went through this argument a few years ago: "people won't write nice GUIs for fun", "people won't write open-source databases", "people won't release open source kernels for big iron". Now it may well be that there is not enough good open documentation on some particular topics at the moment, but that is no reason to assume that it is not worth trying to get free documentation.
The FSF docs are free enough for all practical purposes.
They are free enough for many purposes, but the licence is buggy and needs to be fixed. Some projects such as Arch have gone back to the GPL for documentation, which is the sensible choice at the moment. Both FSF and Debian need to stop fucking around.
I think I preferred the one with the naked lady, the poodle and the salami.
Well, I found the thread pretty interesting. Thanks for your thoughts.
I agree that having standard crash tests at higher speeds would be an excellent idea. Some tests are now done at 64kmh/40mph. Even if the government won't run them, I would like to see some manufacturers take the initiative and publish tests for 60MPH crashes.
I agree it would be good to see more non-painted bumpers to make low speed bumps cheaper. I was noticing on the way to work this morning how much more common they were on older cars. I suppose this is a kind of consumer fashion thing. According to the IIHS report there used to be a federal standard requiring no damage at 5mph; it's a shame that is not still required.
if an airbag isn't enough to keep your ribs from being crushed, I can't imagine the tiny benefit from crumple zones changing anything
It may not seem intuitive but I think having extra distance to decelerate makes an enormous difference. Crumpling the hood probably gives you at least a meter, whereas the seatbelt and airbag allow less than a meter of deceleration.
Assuming a linear deceleration, doubling the cushioning distance will halve the deceleration force on the passengers. Chance of injury increases super-linearly with force. Having a hood that crumples is likely to more than halve the chance of serious injury.
I think the point of a rigid passenger cell is to prevent intrusion of impacting objects or car parts. It doesn't absorb energy. The most a rigid cell can do is turn the impact from falling onto a spike into falling onto a flat surface. At 30m/s, even the flat surface is going to hurt.
I'm sure there is lots of room for improvement. I just don't think more rigid == better by definition.
No, the question is, with a trivial design improvement, could the survival rate be even better?
I would guess not: there is a reasonably competitive market for cars. If there were *trivial* changes that could make safety much better, then you would expect it to happen, at least in luxury cars. I can believe that there are changes that are tough tradeoffs or hard to get right or expensive.
I would, in fact, be willing to bet that older cars were safer in roll-overs
That is a pretty interesting question. In principle you could answer it from statistics, though in a little bit of web research I can't find them. Since rollovers seem responsible for a fair fraction of fatalities you would expect there to be a fair selective pressure to improve handling of them.
Since mass-market cars now travel at 75+mph the domain over which they are required to be safe is wider than it once was. So I can believe maybe they get more damaged in a 35mph rollover as a tradeoff for surviving a 75mph rollover.
I have been making the point that we would be much safer if the body around the passenger area was reinforced until it's at least as strong as older cars were. You can maintain your crumple zones, and air bags, but I really want the passenger compartments to be improved.
That's a pretty interesting question, which I am not qualified to answer but will still speculate about.
I think given the energies involved, having just interior padding and airbags would not be enough to decelerate occupants at a reasonable rate.
One comparison that springs to mind is motorcycle helmets: you can buy USD700 helmets pretty close to what racers use, and I think they're probably the best available technology. They layer elements of different strengths but even the hardest outer glass/carbon fiber element is going to abrade, fracture and flex in a heavy crash.
The point is to decelerate the occupants as gently as possible. I think you want to design the system so that in a maximal crash, the protection is maximally deformed without actually coming apart or squashing the occupants. I don't think rigid elements directly help with that. I can imagine that having parts that flex or collapse up to a stronger breaking point would be good.
You obviously haven't spent much time on old cars. Bumpers are great things, and used to be able to handle very powerful forces. It's a molded piece of solid steel, that is connected, through other strong pieces, to the frame of your car. At 15MPH, you'd expect to dent the bumper to the point you'd want to have it replaced, but it would be a very inexpensive replacement.
No, I have to admit I have not. I would still be surprised if all the damage was confined to the bumper, but you may be right.
BTW, a 100MPH crash isn't all that incredible.
I said "much much more than 100MPH", but anyhow: it's 3 times faster than the NCAP tests. I think it would have been nearly unheard of to survive it in the cars of 25 years ago.
Have a crash test with an older car, and you'll see that the car will stay completely in-tact.
And the passengers may well be killed by hitting the dash or steering wheel at nearly full speed. Try dropping an egg in a metal box vs a cardboard egg carton.
I couldn't care less whether the car stays intact. The question is whether the chances of people surviving are greater or less than they used to be. From what I've read, they are greater, but if you have a different reference I'd be interested to see it.
my02wrxsti (706498) wrote:
:-)
All current model Subarus in Australia have 12500km service intervals (7800 miles).
So what kind of car do you have?
Oops, it's been too long since I did calculus.
Assuming I got it right the second time: A collision at that speed is equivalent to falling 2.3m onto concrete. I'd be surprised if any bumper could sustain that without damaging the car itself.
I'd be pretty leery of riding in a car during such an experiment. Even at such a relatively low speed compared to normal traffic, I'd be pretty happy for any protection I could get by having airbags go off or the car frame get fucked up.
While I have my dx/dt out: a head-on crash from 100kmh is like falling 40m. And 100kmh can seem so slow..
Well, the article starts out talking about a kid completely rolling a car, which is definitely in the domain of caring about survival, not repair costs.
:-)
I agree it would be nice if we could get replaceable rubber-style bumpers to protect against carpark accidents. But 15MPH is not so slow. You'd need a pretty large bumper to absorb it with no other damage, if you could do it at all. 6.7 m/s is equivalent to dropping the car headfirst onto concrete from about 900mm; you'd expect it to make a pretty loud noise.
And additionally, no car I've seen made in the last couple decades has a strong structure.
Really *none*? What are you comparing too?
I've seen a photo of a BMW M3 that flipped on a racetrack at >>100mph. The engine compartment is nearly broken off, but the passenger cell is intact. I'd call that strong -- and strong in the right place, too. No human has a right to survive those kind of energies, but the driver was reported to have walked away.
There was a 6-car low-speed rear-ender outside my office a while ago. It was a pretty interesting demonstration of different construction models: the Kia mini-car folded up by more than a foot and the Mercedes just dinged its bumper. (Possibly the bumper would be more expensive.
Some people suggest that softer bumpers reduce damage to pedestrians. 5-10mph probably won't hurt the occupants, but it's pretty tough on a pedestrian.
What an incredibly whiny article!
The major thrust of car safety design for the last 10-20 years has been that the car should be written off to protect the occupants. Therefore: airbags, crumple zones, seat-belt pre-tensioners and tension limiters, collapsing steering columns, staged failure of structural elements...
The BMW quoted in the article performed very well: the occupant was uninjured, the passenger cell was not breached. The damage looks minor specifically because the structural components are meant to be hardest and fail last. In that minute, the owner spent $30k to prevent their child from being killed or paralyzed. (Whether they still think that was a good deal is another question...)
If they'd been driving an older car, it might well have been repairable after a rollover: more steel, more parts that bend plastically rather than breaking or crumpling, no airbags. On the other hand, if they'd been driving an older car, they might have been dead.
It's pretty simple: if you don't want to write off your car, don't flip it over!
"A good landing is one you walk away from. A perfect landing is where they can use the plane again."
I Have bought cars like a Bic Lighter for years. Get a Cheap one in the 500 to 1000 dollar price range..
Do I have a lighter deal for you!
They distract me because the blue/pink tinges look like police lights in the rear-view mirror...
Till then you are just making excuses for petty criminality.
Saying that civil disobediance of a supposedly unjust law is criminal begs the question. Jesus was a criminal, under the laws of his time. The point is whether the law is moral/reasonable/just to begin with.
I have a flash-based MP3 player that doesn't play AAC. If I used iTunes I'd feel it was fair use and morally defensible to transcode songs so I could play them on my Nuvo. I think the law ought to protect this fair use, rather than banning it.
Turn about: yes, I think it's reasonable for someone to make backups of my proprietary software, or to transfer it when they buy a new machine.
The US has done no such thing, and doesn't really have any legal basis for doing so... corporations have free speech rights here.
Don't be silly. US free speech rights are not unlimited, for either natural persons or corporations. One of the US judges already told SCO to tone down their public statements of Linux and IBM, and they seem to have complied.
The particular balance is different between the US and Germany, but both of them have laws against libel.
I heartily agree. (Mod parent up!) Repeatedly using the word "fundamental" to mean "major" makes her look a bit illiterate. It is nearly as annoying as people using "exponential" when they just mean "rapid".
I also think it's pretty damn funny for her to quote ESR
describing hackers as "unfortunately intolerant and bigoted".
It's also kind of amusing for a pseudo-socialist to talk about religious blindness causing people to ignore working solutions.
This kind of rant has occurred so many times before: the bits that were good weren't original, and the bits that were original weren't good. In fact the bad bits aren't particularly original either.
In conclusion, to Michelle: shut up and code. (Or write docs or file UI bugs, that's good too. Stop whining.)
Could people please stop being so melodramatic with their subject lines?
:-)
Answer: No.
No virtual graphics adapter, so no X-Windows
But how hard can it be to write a dummy framebuffer that maps to a Windows window? Not too hard, I expect.
Or use a Windows X11 server, or a Windows VNC client.
Got a link for that, little troll? I don't see anything on the cisco, distcc or teambuilder sites.
Or just install SCons. Great parallelism, smarter rebuilds, cross platforms, more reliable than make, and far faster than automake/libtool. Works well with distcc too.
run _everything_ in parallel
What, even things that shouldn't be parallel? Screw that.
Damned if I'm letting an electric cloud near my machine room.
What's an iThead? Some kind of new Apple product? They've sold billions already without cellphones? They're taking my job?
Those numbers are weighted by the number of google queries the user does. People with modern machines are more likely to have fast and always-on connections; people with old machines are more likely to have modems.
If you have to wait 15s for the results, as you might on a modem line, you're likely to do far fewer queries per day. Even at the same level of user experience I would think old-machine users make far fewer queries...