Too long, lacks pithiness. Vin Diesel's character, Riddick, put it better in Pitch Black: "Got it all wrong, Holy Man. I absolutely believe in God. And I absolutely hate the fucker." Much more concise.
So if someone were sitting near a ground-based transmitter, able to take control of the drone at a second's notice, that would be OK? How many drones can their transmitter control at once before you start classifying them as driverless?
Ummm.... if it's "non-satellite imagery," where else could it be from?
I'd think a guy taking pictures out of a Cessna wouldn't be very economical long-term compared to a drone.
Yeah, and I'd think a guy taking pictures out of a car to make a map as he drives up and down every single street in the world wouldn't be very economical... oh wait. (Then again, drones you say? Oh snap.)
Exactly. And most importantly, science is about testing your assumptions in order to verify them. If Newton made a systematic, scientific study of alchemy, then he was practicing science, not "a totemic pseudoscience". He may not have managed to turn lead into gold but I'd bet he learned a lot.
Just because it's not bulletproof enough for you to use it to unlock your bank account, for instance, doesn't mean it's pointless. Maybe if a company phone has multiple users, it could update the company phone list with the person who's currently using it?
Why does it have to be 100% accurate? Passwords aren't 100% accurate. Pin codes aren't. Why does some new technology (that's meant to be used in *conjunction* with said things) have to be absolutely perfect before you'll consider it?
Either way, how did a law saying you have to turn over information that is stored inside your fucking head ever get passed in Britain, anyway? Last time I checked you guys have elections, so which elected officials thought this was something worth voting for and why are their heads still attached to their necks? It sounds like the voters must be as "low-information" as the voters here in the US to let something like that get through.
Well, we don't get quite as much latitude with that whole "it's illegal and unconstitutional but the president said so which makes it OK" loophole that you lads in the U.S. use, so our politicians have discovered something far worse: they simply get together for a huddle behind closed doors, and voila! Suddenly everything remotely controversial is bipartisan. Can't vote out policy if no matter who you vote for, they support it. The whole 'follow the party line' thing is very strong here, it's national news if some guy votes against what their party voted.
If you don't believe me, witness the number of retarded things (like the Great Australian Firewall) that the Australian government has done recently despite NO-ONE supporting them.
This is slightly off topic, but I think the Shanzai industry is fascinating. The lack of regard for patents and copyright makes it the closest thing we have to a commercial open-source culture. They only have 3 months at most after launching a product before competitors have cheaper, better clones on the market. That sounds like an impossible task to keep up with, but (since you can in turn steal most of your design from existing products) startup and R&D costs are trivial compared with traditional western 'closed' development.
For a long time people discounted it as a source of innovation, but the very tight iteration time coupled with the low barriers to entry mean that the only way to survive in the market is to innovate relentlessly. The leading manufacturers are now providing more features for lower prices than western brands, and it can only get better from here.
The Shanzai economy is my pick for the one to watch in the next 20 years. Pretty soon it's going to go absolutely mental.
Hmm. Purely being pragmatic, you're right. It doesn't matter how good EV performance is if people won't buy them. From that point of view, what needs fixing right now is people, not EVs.:P
I'm keeping a really close eye on the Chevy Volt, which is exactly what you describe - a plug-in series hybrid which runs exclusively on battery power for the first 40 miles, then starts recharging itself from an onboard genset. I'm really hoping it takes off. Of course, electricity prices are going up hugely (at least here in Australia) so maybe people won't be quite so keen to plug it in when they realise it costs $17 for 64km of all-electric driving... my 21-year-old turbo 6 is cheaper to run!
You're absolutely right. Sadly, a lot of people apply your logic the other way around - instead of saying "well, my Honda Accord won't tow a horse float or let me shift a ton of sand, so why should I expect an electric car to do so?" they say "what if one day I want to tow a horse float for 600kms on the beach with my dog in the car while carrying four friends and my kids?" And they conclude that despite the fact that they'd be perfectly OK with a Honda Accord for 99.999999% of their driving life, that they in fact really need a Canyonero.
And then they only ever drive the goddamn thing across the street to the supermarket and back.
In a gasoline powered fuel, you walk to the nearest gas station, fill up a container with gas, walk it back to your car, and you are good to go. If we could come up with a similar mechanism for an electric car, and help people feel comfortable that the need to employ it would be rare, then the concern would be mostly gone.
We don't need to come up with one. There is one - at least for lead acid battery vehicles. Lead acid batteries 'recover' a small amount of voltage when left to stand, so if you run out of charge in a lead acid EV, you let the car sit for 10-20 minutes and you'll have enough charge to creep a few kilometers home. And in any EV, 'running out of charge' is a gradual process, it's not like in a fuel powered vehicle where you go from 100% performance to 0% instantly.
I think the best suggestion I've seen so far is make sure the batteries used are modular, and easy to replace. If you run out of power, you can go grab a single replacement, which would get you to the station, where you could swap out the rest.
This actually ain't bad. A 10kg lithium booster battery would hold enough charge to get you to the nearest charge point.
The problem with that line of thought is that battery technology, and electric car technology - is not yet at a point where it is "good enough".
Actually, for the vast majority of purposes and drivers, EV technology IS good enough. Sadly, when people come up against new ideas we love to play "what if" and come up with scenarios where the new thing won't work. It's a cheap, easy way to feel smart. That's why you talk to the guy from Chicago who does 20km a week, and he says "yeah but what if I want to drive to my aunt's house in London?" Bingo, electric cars are obviously useless and he gets to feel smug for 10 minutes knowing that he's smarter than whoever suggested electric cars to him.
The end result is that until a new technology is markedly better in every way than the old technology it replaces, it will see undeserved resistance from the general market.
Batteries that can take charge rates like that don't exist yet. There are claims from the "nanotechnology" crowd that they will be available Real Soon Now. We'll see.
Erm, A123 nanophosphate lithium batteries are available in some Black&Decker and DeWalt cordless tools. They're a real commercial product, it's just that the factory output is booked years in advance.
I didn't say "JPEG compression introduces only visually insignificant artifacts", just that visible differences are "most likely" to be disregarded.
I meant only that the JPEG algorithm, like most lossy media codecs, was designed to favour the signal features to which we are most sensitive (for example, iirc the hue information is encoded at 1/4 the spatial resolution of the luminance information, because our eyes are more accurate detecting changes in luminance than changes in hue).
Yeah, I realised that just after posting. *sigh* My bad.
So I downloaded them and I'm flipping between the two images. I agree that the difference is somewhere between bugger all and diddly squat.
For the preview images on the page I still maintain that presenting the two images side by side as they do is misleading given that they are pretending that it's "JPEG vs. WebP" when in actual fact it's "JPEG vs. PNG", since they seem to have compressed the right hand side with WebP at full resolution then scaled it down and PNG'd it, thus most likely hiding any artifacts.
Subtle minutiae across the entire image makes a difference greater than the sum of its parts.
No it doesn't, unless it's the very specific type of subtle minutae that our visual system considers important. Most likely, said differences fall through the cracks in human visual processing, because they exist precisely BECAUSE the compression method throws away data that doesn't affect how we see the image. That's how lossy compression gets smaller file sizes for visually similar results.
They've taken arbitrary sized images, compressed them in both JPEG and WebP, and then destroyed any visible evidence of differences by downscaling them drastically. Any visible differences are due to those downscaled images *again* being compressed with medium-resolution JPEG on the left, while being stored in lossless PNG format on the right.
If anything, this just proves that JPEG is pretty good at making images that look pretty OK.
I dunno about you but I'm using a Monster(TM) brand DVI cable so I get superior native image resolution on my LCD and 11 bits of resolution per colour channel. I can clearly see the difference between the two formats and one of them is vastly superior.
There is no reference image, so I have no idea which is more correct.
Excellent point. Not only that but there's no indication of what level of compression is being used on the "all other brands" image. A lot of them seem ridiculously large for... *checks*
Wait, what the... I call bullshit. Look at the bottom image (just for one example) and the "jpeg" column lists it as being over a megabyte in size. Right click, image properties... "Size: 65,837 bytes."
Try $1.35/L for 95 RON unless you get very lucky in which case it sometimes gets down to $1.25/L. I probably didn't include all my weekend driving, so call it 50km/day average for 1550km/month. I said 'barely' so say $260/month to account for a bit left over. That gives us ~8km/L which is roughly right for my car in city driving. Not too far off given that my initial numbers were very rough estimates...
Slashdot Mirror
I'm thinking this might be a "woosh" event. Subtly done if so.
The real question is whether there's a "screech" event or a "bang" event at the end of it...
about f* time
The original Rav4 EV was released in 1997. So yeah.
Too long, lacks pithiness. Vin Diesel's character, Riddick, put it better in Pitch Black: "Got it all wrong, Holy Man. I absolutely believe in God. And I absolutely hate the fucker." Much more concise.
Little fleas have lesser fleas upon their backs to bite 'em.
And lesser fleas have smaller fleas, so on ad infinitum.
So if someone were sitting near a ground-based transmitter, able to take control of the drone at a second's notice, that would be OK? How many drones can their transmitter control at once before you start classifying them as driverless?
Ummm.... if it's "non-satellite imagery," where else could it be from?
I'd think a guy taking pictures out of a Cessna wouldn't be very economical long-term compared to a drone.
Yeah, and I'd think a guy taking pictures out of a car to make a map as he drives up and down every single street in the world wouldn't be very economical... oh wait. (Then again, drones you say? Oh snap.)
Exactly. And most importantly, science is about testing your assumptions in order to verify them. If Newton made a systematic, scientific study of alchemy, then he was practicing science, not "a totemic pseudoscience". He may not have managed to turn lead into gold but I'd bet he learned a lot.
Big Brother is watching you.
And masturbating. Ew.
Just because it's not bulletproof enough for you to use it to unlock your bank account, for instance, doesn't mean it's pointless. Maybe if a company phone has multiple users, it could update the company phone list with the person who's currently using it?
Why does it have to be 100% accurate? Passwords aren't 100% accurate. Pin codes aren't. Why does some new technology (that's meant to be used in *conjunction* with said things) have to be absolutely perfect before you'll consider it?
Every single one of them thinks the same thing. Newsflash: you're one of them.
That's a (+eleventy, insightful) right there.
Either way, how did a law saying you have to turn over information that is stored inside your fucking head ever get passed in Britain, anyway? Last time I checked you guys have elections, so which elected officials thought this was something worth voting for and why are their heads still attached to their necks? It sounds like the voters must be as "low-information" as the voters here in the US to let something like that get through.
Well, we don't get quite as much latitude with that whole "it's illegal and unconstitutional but the president said so which makes it OK" loophole that you lads in the U.S. use, so our politicians have discovered something far worse: they simply get together for a huddle behind closed doors, and voila! Suddenly everything remotely controversial is bipartisan. Can't vote out policy if no matter who you vote for, they support it. The whole 'follow the party line' thing is very strong here, it's national news if some guy votes against what their party voted.
If you don't believe me, witness the number of retarded things (like the Great Australian Firewall) that the Australian government has done recently despite NO-ONE supporting them.
This is slightly off topic, but I think the Shanzai industry is fascinating. The lack of regard for patents and copyright makes it the closest thing we have to a commercial open-source culture. They only have 3 months at most after launching a product before competitors have cheaper, better clones on the market. That sounds like an impossible task to keep up with, but (since you can in turn steal most of your design from existing products) startup and R&D costs are trivial compared with traditional western 'closed' development.
For a long time people discounted it as a source of innovation, but the very tight iteration time coupled with the low barriers to entry mean that the only way to survive in the market is to innovate relentlessly. The leading manufacturers are now providing more features for lower prices than western brands, and it can only get better from here.
The Shanzai economy is my pick for the one to watch in the next 20 years. Pretty soon it's going to go absolutely mental.
Ooops, make that $1.70, I slipped a decimal point. That's much better! :)
Hmm. Purely being pragmatic, you're right. It doesn't matter how good EV performance is if people won't buy them. From that point of view, what needs fixing right now is people, not EVs. :P
I'm keeping a really close eye on the Chevy Volt, which is exactly what you describe - a plug-in series hybrid which runs exclusively on battery power for the first 40 miles, then starts recharging itself from an onboard genset. I'm really hoping it takes off. Of course, electricity prices are going up hugely (at least here in Australia) so maybe people won't be quite so keen to plug it in when they realise it costs $17 for 64km of all-electric driving... my 21-year-old turbo 6 is cheaper to run!
You're absolutely right. Sadly, a lot of people apply your logic the other way around - instead of saying "well, my Honda Accord won't tow a horse float or let me shift a ton of sand, so why should I expect an electric car to do so?" they say "what if one day I want to tow a horse float for 600kms on the beach with my dog in the car while carrying four friends and my kids?" And they conclude that despite the fact that they'd be perfectly OK with a Honda Accord for 99.999999% of their driving life, that they in fact really need a Canyonero.
And then they only ever drive the goddamn thing across the street to the supermarket and back.
In a gasoline powered fuel, you walk to the nearest gas station, fill up a container with gas, walk it back to your car, and you are good to go. If we could come up with a similar mechanism for an electric car, and help people feel comfortable that the need to employ it would be rare, then the concern would be mostly gone.
We don't need to come up with one. There is one - at least for lead acid battery vehicles. Lead acid batteries 'recover' a small amount of voltage when left to stand, so if you run out of charge in a lead acid EV, you let the car sit for 10-20 minutes and you'll have enough charge to creep a few kilometers home. And in any EV, 'running out of charge' is a gradual process, it's not like in a fuel powered vehicle where you go from 100% performance to 0% instantly.
I think the best suggestion I've seen so far is make sure the batteries used are modular, and easy to replace. If you run out of power, you can go grab a single replacement, which would get you to the station, where you could swap out the rest.
This actually ain't bad. A 10kg lithium booster battery would hold enough charge to get you to the nearest charge point.
The problem with that line of thought is that battery technology, and electric car technology - is not yet at a point where it is "good enough".
Actually, for the vast majority of purposes and drivers, EV technology IS good enough. Sadly, when people come up against new ideas we love to play "what if" and come up with scenarios where the new thing won't work. It's a cheap, easy way to feel smart. That's why you talk to the guy from Chicago who does 20km a week, and he says "yeah but what if I want to drive to my aunt's house in London?" Bingo, electric cars are obviously useless and he gets to feel smug for 10 minutes knowing that he's smarter than whoever suggested electric cars to him.
The end result is that until a new technology is markedly better in every way than the old technology it replaces, it will see undeserved resistance from the general market.
Batteries that can take charge rates like that don't exist yet. There are claims from the "nanotechnology" crowd that they will be available Real Soon Now. We'll see.
Erm, A123 nanophosphate lithium batteries are available in some Black&Decker and DeWalt cordless tools. They're a real commercial product, it's just that the factory output is booked years in advance.
I didn't say "JPEG compression introduces only visually insignificant artifacts", just that visible differences are "most likely" to be disregarded.
I meant only that the JPEG algorithm, like most lossy media codecs, was designed to favour the signal features to which we are most sensitive (for example, iirc the hue information is encoded at 1/4 the spatial resolution of the luminance information, because our eyes are more accurate detecting changes in luminance than changes in hue).
Yeah, I realised that just after posting. *sigh* My bad.
So I downloaded them and I'm flipping between the two images. I agree that the difference is somewhere between bugger all and diddly squat.
For the preview images on the page I still maintain that presenting the two images side by side as they do is misleading given that they are pretending that it's "JPEG vs. WebP" when in actual fact it's "JPEG vs. PNG", since they seem to have compressed the right hand side with WebP at full resolution then scaled it down and PNG'd it, thus most likely hiding any artifacts.
Subtle minutiae across the entire image makes a difference greater than the sum of its parts.
No it doesn't, unless it's the very specific type of subtle minutae that our visual system considers important. Most likely, said differences fall through the cracks in human visual processing, because they exist precisely BECAUSE the compression method throws away data that doesn't affect how we see the image. That's how lossy compression gets smaller file sizes for visually similar results.
They've taken arbitrary sized images, compressed them in both JPEG and WebP, and then destroyed any visible evidence of differences by downscaling them drastically. Any visible differences are due to those downscaled images *again* being compressed with medium-resolution JPEG on the left, while being stored in lossless PNG format on the right.
If anything, this just proves that JPEG is pretty good at making images that look pretty OK.
I dunno about you but I'm using a Monster(TM) brand DVI cable so I get superior native image resolution on my LCD and 11 bits of resolution per colour channel. I can clearly see the difference between the two formats and one of them is vastly superior.
There is no reference image, so I have no idea which is more correct.
Excellent point. Not only that but there's no indication of what level of compression is being used on the "all other brands" image. A lot of them seem ridiculously large for... *checks*
Wait, what the... I call bullshit. Look at the bottom image (just for one example) and the "jpeg" column lists it as being over a megabyte in size. Right click, image properties... "Size: 65,837 bytes."
Try $1.35/L for 95 RON unless you get very lucky in which case it sometimes gets down to $1.25/L. I probably didn't include all my weekend driving, so call it 50km/day average for 1550km/month. I said 'barely' so say $260/month to account for a bit left over. That gives us ~8km/L which is roughly right for my car in city driving. Not too far off given that my initial numbers were very rough estimates...