Asimov's laws are based on an anthropomorphic view of artificial intelligence. They assume that robots will "think" in such a way that they can follow instructions provided in the form of "ethical principles."
It turns out that cost-effective robots are cost-effective because they've been reduced to the minimum amount of motions, brains, etc. to do a specialized task. And no robots think in a way that would make Asimov's Laws (or Tesuka's Laws) particularly relevant.
In reality, it's a lot simpler and safer to run the robot inside a safety cage that keeps people from wandering into it.
Exchange is MS's product that's in the most danger right now, IMHO. It's expensive, takes a whole server to run, has serious ongoing security issues, and the German government (and others too, ISTR) is sponsoring an open source replacement.
Incidentally, every rocket fuel of one kind or another has to have some oxidizer, or it won't work in space(think that one through.)
Most do, but not all. Hydrogen Peroxide is often used by itself as a monopropellant rocket fuel, for instance: just run it past a platinum screen and it reacts all on its own, no air required. However, it doesn't put out nearly as much thrust as kerosene and LOX.
Lupin - I 've got no good ideas. Based on looks, maybe David Schwimmer? (note: he can do way more than Ross from Friends) Jigen - Edward Norton Goemon - Hopefully someone Japanese. Or at least Asian. Who was the guy they used for the first Onimusha game? Or maybe Robin Shou? Fujiko - With a name like Fujiko, you'd have to go with an Asian, though the character looks caucasian... I hate to say change the name... Zenigata - Fortysomething, Japanese? I really don't know many Japanese live action stars. Sonny Chiba?
I don't think they should specify reusables, and I don't think NASA should be involved except maybe to set some goals.
I'd run it like a utility, where if you generate power, the utils must buy it back. In my plan, if you build a launch system, the government must buy 10 launches.
The steps are:
Step 1: Work to establish a legal umbrella under which such a market can exist.
Step 2: Establish mission parameters for a set of desired missions (W payload to X orbit for Y cost/kg, maximum acceleration of Z). Do not specify reusability, staging, or VTVL/HTHL: we want to try all approaches.
Step 3: Contract to buy 10 launches from any company that can hit the mark. Note: thats ten launches per company for each of the missions specified. Once they come up with a light-medium rocket, they get 10 launches, then to get more money they have to come up with a heavy lifter. If they do another light-lifter, it's on their own dollar. If they come in under budget, they keep the profits. If they go over budget, they eat their expenses.
At the end of this, you should have multiple launch systems that have private investement and can compete with each other. That's the beginning of a market. And the total cost will still be less than a NASA-run development cycle.
NASA didn't spend anything to develop the Fisher Space Pen, Fisher did it on their own dime, and not for billions, either (it's still just a ballpoint pen). BTW, the Soviets use them too now, because pencils in zero-g generate nasty bits of floating lead that can get in an astronaut's eye (ouch!).
Do you know that we would most likely already have a replacement heavy lifter rocket if NASA hadn't been in the way? Beal Aerospace was well along with their KISS-designed staged heavy lifter rocket when NASA announced SLI, killing them dead (you can't compete when your competitors are being funded by your intended buyer). The Beal rocket wasn't a done deal, but it had a strong chance of success.
There are a number of other companies out there that are capable of developing space launch systems at a fraction of the cost of a NASA "let's use 8 new technologies and spread jobs across 45 states" efforts.
Here's the US Dept. of Commerce's Technology Administration report on the suborbital market. It lists a number of companies developing suborbital craft, all at a fraction of the cost of NASA's efforts:
The real root of Newton's woes was that then-CEO Sculley forced them to launch the Newton before it was ready. It got a bad reputation because it shipped with software that wasn't ready for prime time yet, and it never recovered.
How good were Newton's sales, anyway? There were a number of attempts to produce a "PDA" around that time, but the Palm was the first to get form factor, battery life and price right.
While I'm sure that MMORPGs will benefit from this technology, I can't help but wonder if the real agenda at Sony is better game AI.
From what I know (and I'm not a programmer), good AI is a real sticking point in games right now, and the computations are suited to parallell/grid computing.
A console with across-the-board better AI and Sony's market share would be unstoppable.
Spend more time with Grammy & Grampy Acheson, because you will wish you had gotten to know them better when they're gone. Get Grampy to teach you what he knows about building a house and welding.
This implies they're going to die. Don't think about it. Worrying about it will provide no comfort when they're gone, and will only screw up the time you have left.
Your sister really is a bitch. Don't let it get to you. Take advantage of it and use her as a measuring stick to learn which girls to avoid.
The following classes in your high school are all miserably badly taught and wastes of your time: art, music, computer science and foreign languages. Take an immersive summer course for the languages, take shop classes instead of art and music, and read Kernighan and Ritchie's book on C and the Art of Programming by Donald Knuth.
Try out for soccer, and stick with it.
I'd be tempted to say "Don't go to Penn State" but that will still be six years off and I would forget by then anyway. Plus I met all my friends there.
Jon Acheson
EVA: most overrated anime ever.
on
Giant Mecha News
·
· Score: 1
I'm with you. I've seen the whole show, and all the movies, and it was just pretentious and tedious.
It never really makes sense, though some of the background of what is going on is explained. Your sympathy for the characters will probably only decrease as the show drags on. Particularly for Shinji. You suck, Shinji.
I had thought it would get good later on like other Gainax series have (Gunbuster, Nadia). But it never did, and the ending turned into a bad incoherent art movie with no resolution or character growth.
But don't take my word for it:
The second [Evangelion Movie] is one of the most pathetic acts of masturbation disguised as animation (and I'm not talking about just the opening scene) ever made. It made me physically ill to watch. If I woke up one morning and discovered that I had made something like that I would kill myself immediately! - Scott Frazier in EX issue 4.8
Also, the animation quality is not uniformly good. Particularly towards the end of the series, when they have scenes that run for minutes that are just voices talking over a still frame. That's just pathetic.
For a good giant robot show with an actual ending, and likeable characters that grow, watch the Escaflowne tv series.
I wouldn't necessarily take my word as gospel. I hated Final Fantasy 8, for instance, which others liked. (I thought the gameplay was just too tedious.)
My faves are FF7 and Skies of Arcadia on Dreamcast.
You might also try renting the game first.
I never played Lunar, but have heard vaguely good things about it.
Is this the merger that was reported before, or has Enix actually bought out Square lock, stock & barrel?
Either way, hopefully we'll get better RPGs out of it. I'm playing Final Fantasy X now, and my enthusiasm is dropping like a stone. Watch cut scene, walk, watch cut scene, walk, it's like they combined the linearness of a rail-type shooter with the annoying random encounters and levelling-up of an RPG. And the characters just don't emote. Pretty, but shallow, and ultimately boring.
I just finished Grandia, a Sega Saturn game that was ported to PS1. It was far more advanced than FFX.
Not to mention that they will cost several times more than their so-called "labor-intensive" counterparts.
Have you any idea how much it costs to do ones-offs? Getting the mold for the plastic parts along would probably cost you over $10,000 (IIRC).
Yes, but the second one costs 3 cents. If you're talking mass production, this isn't going to replace current methods.
The only way things are as cheap as they are today is due to mass production. If you want to do a small run, it will cost you an insane amount of money. That's where this printer will pay off.
Yes. But Evilviper was referring to the Slashdot article's commentary, which specifically mentioned mass production items. Rapid prototyping or small lot production is a whole different thing, and this technology may find a home there.
Another possibility is that this process may be able to do things that can't easily be done with ohter processes. Like printing a Klein bottle.
And how is it you know that it will always be expensive to produce it this way? Costs will decrease the more it's used, and as the technlogy improves. I thought that it was common knowledge that new technology can be more expensive at first.
Yes, but not all technology scales equally well, or gets equally cheap.
Evilviper is correct for a number of reasons:
This technology won't scale as well as injection molding/printing. It will take longer to do the fabrication, and the machinery to do the fab will be lots more complicated. This might be offset by the fact that the 3D printed product could be one piece instead of several, eliminating assembly, but frankly I doubt it. Assembly is also automated these days.
3D printing technology that prints in metal already exists, and will be used to print the metal moulds for a lot less than machining them by hand costs now. I'd be surprised if this isn't already being done.
A good comparison is offset printing vs. inkjet. Offset costs more for setup, but is dirt cheap in volume. Inkjet has very low setup costs, (i.e. getting the PDF right), but doesn't get that much cheaper in volume. So you use one for one thing and one for another thing. One last bit: Offset printing's higher setup costs have been dropping because offset printing isn't standing still either.
Sometimes I really can't understand the attitude of some people on/.. Here is this really amazing technology...Printing real 3D electronic devices for f's sake. And all you can think up is some poor comment on about how expensive it will be? *sigh*
Why should we turn off our sense of reason just because something is somewhat novel?
Slashdot Mirror
Asimov's laws are based on an anthropomorphic view of artificial intelligence. They assume that robots will "think" in such a way that they can follow instructions provided in the form of "ethical principles."
It turns out that cost-effective robots are cost-effective because they've been reduced to the minimum amount of motions, brains, etc. to do a specialized task. And no robots think in a way that would make Asimov's Laws (or Tesuka's Laws) particularly relevant.
In reality, it's a lot simpler and safer to run the robot inside a safety cage that keeps people from wandering into it.
Jon Acheson
Exchange is MS's product that's in the most danger right now, IMHO. It's expensive, takes a whole server to run, has serious ongoing security issues, and the German government (and others too, ISTR) is sponsoring an open source replacement.
Jon Acheson
Most do, but not all. Hydrogen Peroxide is often used by itself as a monopropellant rocket fuel, for instance: just run it past a platinum screen and it reacts all on its own, no air required. However, it doesn't put out nearly as much thrust as kerosene and LOX.
Jon Acheson
Actually, hydrogen is typically produced FROM gasoline, because it's much cheaper than electrolysis.
Enjoy your crack pipe.
Jon Acheson
OK, that WOULD rock!
Jon Acheson
That seems like a possibility. Raise their profile in the arcades and hope it translates into more sales of home machines.
However, it only works if you have games people want to play in the arcades. Halo isn't going to cut it.
Mechwarrior might, though. Or Steel Battalion...
At any rate, I wish them ill, because Word still sucks.
Jon Acheson
Lupin - I 've got no good ideas. Based on looks, maybe David Schwimmer? (note: he can do way more than Ross from Friends)
Jigen - Edward Norton
Goemon - Hopefully someone Japanese. Or at least Asian. Who was the guy they used for the first Onimusha game? Or maybe Robin Shou?
Fujiko - With a name like Fujiko, you'd have to go with an Asian, though the character looks caucasian... I hate to say change the name...
Zenigata - Fortysomething, Japanese? I really don't know many Japanese live action stars. Sonny Chiba?
What do you think?
Jon Acheson
That's news to me, I always heard him referred to by his penname.
Jon Acheson
I don't think they should specify reusables, and I don't think NASA should be involved except maybe to set some goals.
I'd run it like a utility, where if you generate power, the utils must buy it back. In my plan, if you build a launch system, the government must buy 10 launches.
The steps are:
Step 1: Work to establish a legal umbrella under which such a market can exist.
Step 2: Establish mission parameters for a set of desired missions (W payload to X orbit for Y cost/kg, maximum acceleration of Z). Do not specify reusability, staging, or VTVL/HTHL: we want to try all approaches.
Step 3: Contract to buy 10 launches from any company that can hit the mark. Note: thats ten launches per company for each of the missions specified. Once they come up with a light-medium rocket, they get 10 launches, then to get more money they have to come up with a heavy lifter. If they do another light-lifter, it's on their own dollar. If they come in under budget, they keep the profits. If they go over budget, they eat their expenses.
At the end of this, you should have multiple launch systems that have private investement and can compete with each other. That's the beginning of a market. And the total cost will still be less than a NASA-run development cycle.
Jon Acheson
NASA didn't spend anything to develop the Fisher Space Pen, Fisher did it on their own dime, and not for billions, either (it's still just a ballpoint pen). BTW, the Soviets use them too now, because pencils in zero-g generate nasty bits of floating lead that can get in an astronaut's eye (ouch!).
Jon Acheson
Do you know that we would most likely already have a replacement heavy lifter rocket if NASA hadn't been in the way? Beal Aerospace was well along with their KISS-designed staged heavy lifter rocket when NASA announced SLI, killing them dead (you can't compete when your competitors are being funded by your intended buyer). The Beal rocket wasn't a done deal, but it had a strong chance of success.
There are a number of other companies out there that are capable of developing space launch systems at a fraction of the cost of a NASA "let's use 8 new technologies and spread jobs across 45 states" efforts.
Here's the US Dept. of Commerce's Technology Administration report on the suborbital market. It lists a number of companies developing suborbital craft, all at a fraction of the cost of NASA's efforts:
Suborbital Reusable Launch Vehicles and Applicable Markets, October 2002
There are clearly alternatives to doing it the NASA way, and we should pursue them. If we use NASA's methods, we will get NASA's results.
Jon Acheson
I'm thinking console saved game here, FYI. Primarily because there's nowhere near the same need for this on the PC side.
Even if you go nuts with the data you save, the saved game will be less than a megabyte, and that takes, what, a second to load?
Jon Acheson
The real root of Newton's woes was that then-CEO Sculley forced them to launch the Newton before it was ready. It got a bad reputation because it shipped with software that wasn't ready for prime time yet, and it never recovered.
How good were Newton's sales, anyway? There were a number of attempts to produce a "PDA" around that time, but the Palm was the first to get form factor, battery life and price right.
Jon Acheson
While I'm sure that MMORPGs will benefit from this technology, I can't help but wonder if the real agenda at Sony is better game AI.
From what I know (and I'm not a programmer), good AI is a real sticking point in games right now, and the computations are suited to parallell/grid computing.
A console with across-the-board better AI and Sony's market share would be unstoppable.
Jon Acheson
I'd be tempted to say "Don't go to Penn State" but that will still be six years off and I would forget by then anyway. Plus I met all my friends there.
Jon Acheson
It never really makes sense, though some of the background of what is going on is explained. Your sympathy for the characters will probably only decrease as the show drags on. Particularly for Shinji. You suck, Shinji.
I had thought it would get good later on like other Gainax series have (Gunbuster, Nadia). But it never did, and the ending turned into a bad incoherent art movie with no resolution or character growth.
But don't take my word for it:
Also, the animation quality is not uniformly good. Particularly towards the end of the series, when they have scenes that run for minutes that are just voices talking over a still frame. That's just pathetic.
For a good giant robot show with an actual ending, and likeable characters that grow, watch the Escaflowne tv series.
Jon Acheson
I wouldn't necessarily take my word as gospel. I hated Final Fantasy 8, for instance, which others liked. (I thought the gameplay was just too tedious.)
My faves are FF7 and Skies of Arcadia on Dreamcast.
You might also try renting the game first.
I never played Lunar, but have heard vaguely good things about it.
Jon Acheson
Is this the merger that was reported before, or has Enix actually bought out Square lock, stock & barrel?
Either way, hopefully we'll get better RPGs out of it. I'm playing Final Fantasy X now, and my enthusiasm is dropping like a stone. Watch cut scene, walk, watch cut scene, walk, it's like they combined the linearness of a rail-type shooter with the annoying random encounters and levelling-up of an RPG. And the characters just don't emote. Pretty, but shallow, and ultimately boring.
I just finished Grandia, a Sega Saturn game that was ported to PS1. It was far more advanced than FFX.
Jon Acheson
And I wasn't aware they distributed their games, either. I thought SCEA distributed Square games. Though I don't have a box handy to check that with.
Jon Acheson
So, probably not.
He was a truly unique individual, and will be missed.
Jon Acheson
Samsung has a 27" HDTV monitor you can find on sale for $700.00 at Best Buy. I have one, and like it.
Jon Acheson
Have you any idea how much it costs to do ones-offs? Getting the mold for the plastic parts along would probably cost you over $10,000 (IIRC).
Yes, but the second one costs 3 cents. If you're talking mass production, this isn't going to replace current methods.
The only way things are as cheap as they are today is due to mass production. If you want to do a small run, it will cost you an insane amount of money. That's where this printer will pay off.
Yes. But Evilviper was referring to the Slashdot article's commentary, which specifically mentioned mass production items. Rapid prototyping or small lot production is a whole different thing, and this technology may find a home there.
Another possibility is that this process may be able to do things that can't easily be done with ohter processes. Like printing a Klein bottle.
And how is it you know that it will always be expensive to produce it this way? Costs will decrease the more it's used, and as the technlogy improves. I thought that it was common knowledge that new technology can be more expensive at first.
Yes, but not all technology scales equally well, or gets equally cheap.
Evilviper is correct for a number of reasons:
- This technology won't scale as well as injection molding/printing. It will take longer to do the fabrication, and the machinery to do the fab will be lots more complicated. This might be offset by the fact that the 3D printed product could be one piece instead of several, eliminating assembly, but frankly I doubt it. Assembly is also automated these days.
- 3D printing technology that prints in metal already exists, and will be used to print the metal moulds for a lot less than machining them by hand costs now. I'd be surprised if this isn't already being done.
A good comparison is offset printing vs. inkjet. Offset costs more for setup, but is dirt cheap in volume. Inkjet has very low setup costs, (i.e. getting the PDF right), but doesn't get that much cheaper in volume. So you use one for one thing and one for another thing. One last bit: Offset printing's higher setup costs have been dropping because offset printing isn't standing still either.Sometimes I really can't understand the attitude of some people on
Why should we turn off our sense of reason just because something is somewhat novel?
Jon Acheson
Or those t-shirts that change color, or those coffee mugs....
Serisouly, even if they are changing color dynamically, isn't that basically just wrapping an object in "electronic ink" paper?
Jon Acheson
That's the one.
It worked too, I remember pulling a new modem dialler off of it.
A much better solution than Windows Update.
Jon Acheson
OS/2 also had a program similar to Windows Update in 1995 or so, except that it didn't require a browser and used open protocols to run.
I can't for the life of me remember the exact name, though. Anyone else remember?
Jon Acheson