"ignore the production costs" - what exactly do you think you're paying for when you buy renewables equipment ? "ignore the environmental cost of the equipment" - the energy paybacks on all renewables techs are now very low. Concrete usage in wind turbines is not comparably significant (perhaps you're thinking of dams?). Yes, some producers of solar cells, mainly in China, have had bad waste management practices (like a lot of Chinese industry in general). But compared to the amount of power produced over the lifespan of the products, it's quite small.
Anyway, once again we see that the issue with nuclear is rarely lives - it's cost. Nuclear accidents tend to be accidents in slow motion. Excluding any pressure explosions or the like, they generally give you plenty of time to get away without profound health consequences. But the down side is that, being in slow motion, they just keep on going and going, and keep on costing money. They may be in slow motion, but they don't let you just ignore them. You can't just stay in an area with, say, contaminated water and keep drinking it as if nothing's wrong. You can't just keep operating a facility that's suffered an accident as if it never happened. You have to remedy them and it always costs a fortune. And the potential upper bounds on the costs are almost unlimited (picture, say, the cost of a worst-case scenario at Indian Point with winds pointed at NYC - the cost of even a couple week evacuation of NYC is almost unthinkable).
The nuclear industry has long suffered from a very unfortunate problem: a negative learning curve. With most technology, the longer you use and produce it, the cheaper it gets per unit. The nuclear industry has been one of the few industry where the costs have risen with time as people learn more problems in their designs and more risks that haven't been taken into account. And often the only way to address them is with brand new generations of reactors. Which is great, except that now you're starting your learning curve over from scratch, and your system is most commonly even more complicated to boot. It's really been a curse to the industry, and until it goes away, a true "nuclear rennaissance" is never going to occur. And no amount of government limitations on liability, no amount of municipalities forcing costs on to consumers, no amount of anything will really get the "take over the market" takeoff that proponents really want to see.
That's of course not the only problem nuclear has had. Another is the very long lead times on projects. The consequence is that you have to guess long in advance what the electricity market is going to be like. France suffered from this - they significantly overestimated what electricity consumption was going to be when they built most of their nuclear plants, leading to a generation capacity glut. This led to a lot of really inefficient uses of electricity and much higher investment costs than were necessary to meet demand.
I was fiddling a bit the other day with the NASA Ames Stereo Pipeline. Would be awesome if I could get it to work with series of screenshots from a moving smartphone as a cloud service. Wouldn't that make an awesome app, building *real* 3d scenes, and making the models available for export in a variety of formats and with direct-links for popular functions (editing apps, export to popular 3d printing services, etc)?
Most of this "3d" is just stupid stereoscopy. It's useless except for appearances. Unless your camera is building a 3d *model*, it's just a bit of eye candy - and most commonly, eye candy that requires special glasses or crossed eyes.
I hope that one day we'll come up with a way to recover images and sounds from the past. I've often wondered, for example, if crystal-forming reactions (say, the slow and long-ongoing process of cement hardening) might have statistically measurable shape differences under the influence of different frequency sounds. Or whether photodegradation of materials could yield information about photon fluxes and their direction of travel at each point on the surface, thus allowing for a rough holographic reconstruction of the obstacles around the object at the time it was exposed to light.
I know it's grasping at really weak data signals, but it would be so neat if something proved to be recoverable by some means.
Speaking of old phones, did anyone else know the trick to make payphones start ringing until someone answered them? I loved that trick when I was a kid... it involved dialing a certain code and then hanging up three times in immediate succession. Didn't work on all payphones but it worked on the majority in my area.:)
Yeah, I hate that. Now, you could probably have used Shazam - but it's not always a solution - for example, for obscure or foreign music. A while back I actually went so far at one point as to record a song on the radio that I really liked with my cell phone so that I could play it back to write down the lyrics to google them... but since it was an Icelandic song and thus really obscure even that didn't help. I ultimately ended up - with quite a bit of searching - guessing the title on Youtube. Fallback solution of choice;)
Yeah, I wanted to write some remark about me finding most of my new music and bands by going to concerts for other bands that I like and being unexpectedly impressed by the opening act, which is something that if a person does iteratively is a way to pretty much ensure they ends up a hipster who's crazy into bands that nobody else has ever heard of.;) Then thought, well, what if "at concerts" is an actual category in the study? I didn't want to get a bunch of snide "RTFA" remarks, so I went to check out the article... and saw a nice demonstration of Slashdot's lack of proofreading, to the point of not even checking whether their URL is malformed.
Grasshopper always did something roughly equivalent to a last second suicide burn, why would Falcon be any different? But that's "slowly lowering itself to the deck". Only for a brief moment, but the key is, getting the velocity down. It clearly had the velocity down. But something went wrong.
First off, where did you get the impression that Musk said anything like that? Secondly, where did you get the impression that the person you're replying to said anything like that? They never said "land at the same place".
Probably some truth to this. Russia has been a thorn in OPEC's side for a long time. And they got the perfect opportunity here to hit them when they were weak. Now they've got Russia losing income *and* not able to get foreign investment. The net result, if Russia's isolation and weakness continues, is that future projects will be delayed or cancelled, existing production dries up, Russian hardware wears out, etc - and overall Russia becomes a smaller player in the global oil market.
It's clearly not the only factor in the oil slide, but OPEC seems more than a little content with the price slide. And I don't think there's any love lost between the Saudis and the Russians on non-economic issues either... they seem to be on opposite sides of most international conflicts.
It'll stabilize the fall at the very least, to be sure. But the rest of the oil industry is trying their hardest not to give ground - they don't want to undercut their future on projects that can take years to develop and stay in operation for long periods of time. Most wells (or non-well projects such as surface-mined bitumen) have much longer lifespans. Fracked oil and oil shale will be what gives way most, but the exact nature of where the market will head is yet to be seen.
It is hard - extremely hard. Stability is only part of the problem, hover offers a lot of other problems that don't crop up when the craft is facing strong and roughly steady G forces. But landing like that gives mass benefits. And small mass benefits on one stage means very large benefits for your payload capacity delivered to space. So if you can pull it off, it's a big win.
Landing *anything* that's been to space on a barge has never been done before. Landing a full-size *first stage* that's delivered a payload to space *anywhere* has never been done before (far, far harder problem than landing a crew capsule, they're massive, hard to control, and imbalanced).
The thing is nearly as heavy as the Space Shuttle, much larger, a much more unstable shape (due to the nature of rockets), has only stubby grid fins for manueuvering, is controlled by a computer with no ability for real-time corrections by a pilot, and was landing on a barge barely larger than itself. And was created on what's by comparison a shoestring budget. Let's not pretend that this isn't a massive challenge.
I know it's the nerd-equivalent of being a hipster to berate anyone who expresses any support for SpaceX, but for god's sake, look objectively at the nature of this challenge for a minute. One can hate Elon, but these engineers are not exactly picking the low hanging fruit, and they've still achieved very impressive results thusfar.
(The real question will be, of course, whether they can actually refurbish these first stages cheaper than they can build them... most people assumed the Shuttle would be dirt cheap to refurbish, after all...)
It tells us at least that the rocket didn't impact it moving at hundreds of meters a second. Sabot-launch a streamlined wineglass out of an high power air cannon at a chunk of granite and yeah, you'll almost certainly break off chunks of it.
We can say pretty assuredly say that it got to its final landing stage, was slowly lowering itself to the deck... and then "something" went wrong.
All we can say is, if the ship is fine, just some equipment damaged on deck, then this was a very low speed impact. Sounds more like the rocket falling over than anything else. Why? Not a clue. But I guess we'll find out. Maybe Musk's plan of "have it weigh itself down and then we'll weld it to the deck afterwards" wasn't a good one.
Nobody's going to be taking anything from a nuclear armed state. The most I think anyone would ever dare do is covertly support breakaway rebel movements, but even that would be extremely provocative.
That said, there probably will be an upsurge in internal rebel acticity in Russia as they lose their ability to pay off local tribal leaders with oil money like they've been doing extensively of late....
Putin, the election rigger? Oh please. It's clearly legitimate that 99,5% of Chechnyans voted for "The Butcher of Grozny". Turnouts of over 100% in some precincts in Grozny, well, people were just *that* excited to get to vote for him!
"Somewhat improved properties over plain plastic" is nothing on the order of hand-laid composites, or having properties like a metal, or anything of the sort. That you can print a very mildly-conductive plastic (which this stuff in the article is not) or mildly reinforcing fibers (which the stuff in this article is not) in some extrusion printers is making a mockery of what people *actually* want to print (actual metal, actual strong fibers, etc). And it's what people who didn't know any better were thinking of when they saw this article.
Yes, very good, that was my point. But you are still not getting a cookie.
Just because one *can* make plastic composites with significantly improved properties (for example, carbon fiber panels) doesn't mean it has relevance to 3d printers in general, and it certainly doesn't have any connection to what Makerbot is demoing here. Alumide is all about appearance, plain and simple. And it's not even a metalic appearance, it's more of a sandy appearance.
Very cool:) CNC to get you the rough shape of the large object, then spray to get complex details and advanced materials and coatings. They use a laser jet rather than combustion jet, but the principle is the same. And look at that deposition rate: 3.5 kilograms an hour! Even without the CNC you could print a 1-ton *car* in two weeks at those rate;) Wide range of powders usable. Wall thickness down to 0,1mm - I didn't even think that it'd be able to be that precise. And the results are just beautiful.
I'm so glad to know that someone's working on this 3d printing technology!:) Whenever I write about this possibility I usually just get a bunch of smart-assed "well if it's such a great idea why don't you go build it yourself, genius?" as if it takes 30 seconds and some pocket change to go build a new type of 3d printer.
Little roving glue guns are never going to be the future. But this sort of technology could be.
You're speaking in theoreticals. I'm speaking in facts. Alumide is *not* a conductor, and it is *not* unusually abrasion resistant vs. other hard plastics. Yes, it's possible to make plastic composites that have improved properties. For an extreme example, look to CFRP. This is nothing of the sort. The average grain size for Shapeways' alumide, for example, is 60 microns, so we're talking extremely tiny particles. Alumide's tensile strength is 48 MPA (typical of a hard plastic; metals are usually in the hundreds to thousands). The Shore D hardness is 48 (plastics range from 0 to around 120). The melting point is 172-180C (common rigid plastics are usually 70-260C; metals range into the thousands). And yes, despite the presence of aluminum dust, the resistivity is 3e14 ohms (plastics are broadly in the range of 1e12-24; bulk aluminum is 3e-8).
It's a plastic and it behaves like a plastic, plain and simple. This isn't some theoretical. If you can make a PCB out of this stuff, you can make it out of common everyday plastics too.
Everyone is stuck on the concept of extruders. What I want to see is a 3d printer based on thermal spraying, a technique mainly used today for applying coatings.
1. First off, you don't have to be only the tiniest height above your workpiece like you do with an extruder - you have some degree of distance range (your precision decreases with distance, but there's a tradeoff and it depends on the situation). Your precision vs. flow rate can be chosen by your choice of nozzle size.
2. By varying the fuel and air partial pressures, you can create a radically different propulsion environments and choose whatever impact speeds and temperatures you want for your material - from over a thousand meters per second to a couple meters per second, from thousands of degrees to room temperature. This means your printing materials choice is almost, so long as they can be blown through the nozzle and caused - at some temperature/impact speed combination - to stick to the target. Plastic, metal, fibers, glasses, blends of materials, even food - almost anything, from the same printer.
3. Contrary to extrusion printing which generally produces parts inferior to molding, thermal spraying is used because it produces layers *superior* in properties to what can be otherwise achieved. Major factors in this are the capability to have the particles impact at high speed where desirable and to use materials in situations that normally cannot be done any other way
4. Thermal spraying gives you the ability to do the finishing work from the printer itself. That is after all what it's most commonly used for in the first place - coating with layers to resist scratching, water, corrosion, etc. By proper choice of materials, the same device could paint or sandblast your structure, even sandblasting away temporary supports if your control algorithm was smart enough. You could paint wires onto a PCB, texture a gripping surface into a piece of plastic, pretty much anything.
5. In terms of users providing their own material to print with, a coffee grinder and a sieve is all that one should need for a spraying-based printer. Rolling your own filaments for an extrusion printer, by contrast, is anything but trivial.
Downsides of course would include heat, fuel requirements (either a natural gas connection or fuel bottles), noise (requires a compressor), and potential dust or gas byproducts, depending on how you'd be using it. Hence it would have to be enclosed with filters and sound dampening, and probably be much more suitable for a garage or industrial setting than sitting on the desk next to your computer. Also, the nozzles don't last forever, but then again extrusion printers have set the bar pretty darn low on that one. Lastly, it's not a very fast process. On the upside, that means more precision on your Z axis.
Anyway, I have trouble picturing any other technology for laying down materials that offers that kind of potential versaility. Now, it wouldn't be everything - t's not going to do lithography, it's not going to assemble pieces that need to be made out of multiple parts, etc. But nonetheless, I simply can't envision extrusion as being the tech that ultimately gives people what they dream of from 3d printing (something close to a "replicator"). This potentially could be that.
That's sort of like the saying, "My keys are always in the last place I look!" It's a truism; it's as if we're supposed to believe that there's people out there who are looking for their keys, find them, and then decide to just keep on looking for them.
Slashdot Mirror
"ignore the production costs" - what exactly do you think you're paying for when you buy renewables equipment ?
"ignore the environmental cost of the equipment" - the energy paybacks on all renewables techs are now very low. Concrete usage in wind turbines is not comparably significant (perhaps you're thinking of dams?). Yes, some producers of solar cells, mainly in China, have had bad waste management practices (like a lot of Chinese industry in general). But compared to the amount of power produced over the lifespan of the products, it's quite small.
Anyway, once again we see that the issue with nuclear is rarely lives - it's cost. Nuclear accidents tend to be accidents in slow motion. Excluding any pressure explosions or the like, they generally give you plenty of time to get away without profound health consequences. But the down side is that, being in slow motion, they just keep on going and going, and keep on costing money. They may be in slow motion, but they don't let you just ignore them. You can't just stay in an area with, say, contaminated water and keep drinking it as if nothing's wrong. You can't just keep operating a facility that's suffered an accident as if it never happened. You have to remedy them and it always costs a fortune. And the potential upper bounds on the costs are almost unlimited (picture, say, the cost of a worst-case scenario at Indian Point with winds pointed at NYC - the cost of even a couple week evacuation of NYC is almost unthinkable).
The nuclear industry has long suffered from a very unfortunate problem: a negative learning curve. With most technology, the longer you use and produce it, the cheaper it gets per unit. The nuclear industry has been one of the few industry where the costs have risen with time as people learn more problems in their designs and more risks that haven't been taken into account. And often the only way to address them is with brand new generations of reactors. Which is great, except that now you're starting your learning curve over from scratch, and your system is most commonly even more complicated to boot. It's really been a curse to the industry, and until it goes away, a true "nuclear rennaissance" is never going to occur. And no amount of government limitations on liability, no amount of municipalities forcing costs on to consumers, no amount of anything will really get the "take over the market" takeoff that proponents really want to see.
That's of course not the only problem nuclear has had. Another is the very long lead times on projects. The consequence is that you have to guess long in advance what the electricity market is going to be like. France suffered from this - they significantly overestimated what electricity consumption was going to be when they built most of their nuclear plants, leading to a generation capacity glut. This led to a lot of really inefficient uses of electricity and much higher investment costs than were necessary to meet demand.
Exactly *Real* 3d, not stereoscopy.
I was fiddling a bit the other day with the NASA Ames Stereo Pipeline. Would be awesome if I could get it to work with series of screenshots from a moving smartphone as a cloud service. Wouldn't that make an awesome app, building *real* 3d scenes, and making the models available for export in a variety of formats and with direct-links for popular functions (editing apps, export to popular 3d printing services, etc)?
Most of this "3d" is just stupid stereoscopy. It's useless except for appearances. Unless your camera is building a 3d *model*, it's just a bit of eye candy - and most commonly, eye candy that requires special glasses or crossed eyes.
Point conceded - thanks for the link. :)
I hope that one day we'll come up with a way to recover images and sounds from the past. I've often wondered, for example, if crystal-forming reactions (say, the slow and long-ongoing process of cement hardening) might have statistically measurable shape differences under the influence of different frequency sounds. Or whether photodegradation of materials could yield information about photon fluxes and their direction of travel at each point on the surface, thus allowing for a rough holographic reconstruction of the obstacles around the object at the time it was exposed to light.
I know it's grasping at really weak data signals, but it would be so neat if something proved to be recoverable by some means.
Speaking of old phones, did anyone else know the trick to make payphones start ringing until someone answered them? I loved that trick when I was a kid... it involved dialing a certain code and then hanging up three times in immediate succession. Didn't work on all payphones but it worked on the majority in my area. :)
Yeah, I hate that. Now, you could probably have used Shazam - but it's not always a solution - for example, for obscure or foreign music. A while back I actually went so far at one point as to record a song on the radio that I really liked with my cell phone so that I could play it back to write down the lyrics to google them... but since it was an Icelandic song and thus really obscure even that didn't help. I ultimately ended up - with quite a bit of searching - guessing the title on Youtube. Fallback solution of choice ;)
Yeah, I wanted to write some remark about me finding most of my new music and bands by going to concerts for other bands that I like and being unexpectedly impressed by the opening act, which is something that if a person does iteratively is a way to pretty much ensure they ends up a hipster who's crazy into bands that nobody else has ever heard of. ;) Then thought, well, what if "at concerts" is an actual category in the study? I didn't want to get a bunch of snide "RTFA" remarks, so I went to check out the article... and saw a nice demonstration of Slashdot's lack of proofreading, to the point of not even checking whether their URL is malformed.
[Citation needed]
Grasshopper always did something roughly equivalent to a last second suicide burn, why would Falcon be any different? But that's "slowly lowering itself to the deck". Only for a brief moment, but the key is, getting the velocity down. It clearly had the velocity down. But something went wrong.
First off, where did you get the impression that Musk said anything like that? Secondly, where did you get the impression that the person you're replying to said anything like that? They never said "land at the same place".
Probably some truth to this. Russia has been a thorn in OPEC's side for a long time. And they got the perfect opportunity here to hit them when they were weak. Now they've got Russia losing income *and* not able to get foreign investment. The net result, if Russia's isolation and weakness continues, is that future projects will be delayed or cancelled, existing production dries up, Russian hardware wears out, etc - and overall Russia becomes a smaller player in the global oil market.
It's clearly not the only factor in the oil slide, but OPEC seems more than a little content with the price slide. And I don't think there's any love lost between the Saudis and the Russians on non-economic issues either... they seem to be on opposite sides of most international conflicts.
It'll stabilize the fall at the very least, to be sure. But the rest of the oil industry is trying their hardest not to give ground - they don't want to undercut their future on projects that can take years to develop and stay in operation for long periods of time. Most wells (or non-well projects such as surface-mined bitumen) have much longer lifespans. Fracked oil and oil shale will be what gives way most, but the exact nature of where the market will head is yet to be seen.
It is hard - extremely hard. Stability is only part of the problem, hover offers a lot of other problems that don't crop up when the craft is facing strong and roughly steady G forces. But landing like that gives mass benefits. And small mass benefits on one stage means very large benefits for your payload capacity delivered to space. So if you can pull it off, it's a big win.
Landing *anything* that's been to space on a barge has never been done before.
Landing a full-size *first stage* that's delivered a payload to space *anywhere* has never been done before (far, far harder problem than landing a crew capsule, they're massive, hard to control, and imbalanced).
The thing is nearly as heavy as the Space Shuttle, much larger, a much more unstable shape (due to the nature of rockets), has only stubby grid fins for manueuvering, is controlled by a computer with no ability for real-time corrections by a pilot, and was landing on a barge barely larger than itself. And was created on what's by comparison a shoestring budget. Let's not pretend that this isn't a massive challenge.
I know it's the nerd-equivalent of being a hipster to berate anyone who expresses any support for SpaceX, but for god's sake, look objectively at the nature of this challenge for a minute. One can hate Elon, but these engineers are not exactly picking the low hanging fruit, and they've still achieved very impressive results thusfar.
(The real question will be, of course, whether they can actually refurbish these first stages cheaper than they can build them... most people assumed the Shuttle would be dirt cheap to refurbish, after all...)
It tells us at least that the rocket didn't impact it moving at hundreds of meters a second. Sabot-launch a streamlined wineglass out of an high power air cannon at a chunk of granite and yeah, you'll almost certainly break off chunks of it.
We can say pretty assuredly say that it got to its final landing stage, was slowly lowering itself to the deck... and then "something" went wrong.
All we can say is, if the ship is fine, just some equipment damaged on deck, then this was a very low speed impact. Sounds more like the rocket falling over than anything else. Why? Not a clue. But I guess we'll find out. Maybe Musk's plan of "have it weigh itself down and then we'll weld it to the deck afterwards" wasn't a good one.
Nobody's going to be taking anything from a nuclear armed state. The most I think anyone would ever dare do is covertly support breakaway rebel movements, but even that would be extremely provocative.
That said, there probably will be an upsurge in internal rebel acticity in Russia as they lose their ability to pay off local tribal leaders with oil money like they've been doing extensively of late....
Putin, the election rigger? Oh please. It's clearly legitimate that 99,5% of Chechnyans voted for "The Butcher of Grozny". Turnouts of over 100% in some precincts in Grozny, well, people were just *that* excited to get to vote for him!
"Somewhat improved properties over plain plastic" is nothing on the order of hand-laid composites, or having properties like a metal, or anything of the sort. That you can print a very mildly-conductive plastic (which this stuff in the article is not) or mildly reinforcing fibers (which the stuff in this article is not) in some extrusion printers is making a mockery of what people *actually* want to print (actual metal, actual strong fibers, etc). And it's what people who didn't know any better were thinking of when they saw this article.
Just because one *can* make plastic composites with significantly improved properties (for example, carbon fiber panels) doesn't mean it has relevance to 3d printers in general, and it certainly doesn't have any connection to what Makerbot is demoing here. Alumide is all about appearance, plain and simple. And it's not even a metalic appearance, it's more of a sandy appearance.
Very cool :) CNC to get you the rough shape of the large object, then spray to get complex details and advanced materials and coatings. They use a laser jet rather than combustion jet, but the principle is the same. And look at that deposition rate: 3.5 kilograms an hour! Even without the CNC you could print a 1-ton *car* in two weeks at those rate ;) Wide range of powders usable. Wall thickness down to 0,1mm - I didn't even think that it'd be able to be that precise. And the results are just beautiful.
I'm so glad to know that someone's working on this 3d printing technology! :) Whenever I write about this possibility I usually just get a bunch of smart-assed "well if it's such a great idea why don't you go build it yourself, genius?" as if it takes 30 seconds and some pocket change to go build a new type of 3d printer.
Little roving glue guns are never going to be the future. But this sort of technology could be.
You're speaking in theoreticals. I'm speaking in facts. Alumide is *not* a conductor, and it is *not* unusually abrasion resistant vs. other hard plastics. Yes, it's possible to make plastic composites that have improved properties. For an extreme example, look to CFRP. This is nothing of the sort. The average grain size for Shapeways' alumide, for example, is 60 microns, so we're talking extremely tiny particles. Alumide's tensile strength is 48 MPA (typical of a hard plastic; metals are usually in the hundreds to thousands). The Shore D hardness is 48 (plastics range from 0 to around 120). The melting point is 172-180C (common rigid plastics are usually 70-260C; metals range into the thousands). And yes, despite the presence of aluminum dust, the resistivity is 3e14 ohms (plastics are broadly in the range of 1e12-24; bulk aluminum is 3e-8).
It's a plastic and it behaves like a plastic, plain and simple. This isn't some theoretical. If you can make a PCB out of this stuff, you can make it out of common everyday plastics too.
Everyone is stuck on the concept of extruders. What I want to see is a 3d printer based on thermal spraying, a technique mainly used today for applying coatings.
1. First off, you don't have to be only the tiniest height above your workpiece like you do with an extruder - you have some degree of distance range (your precision decreases with distance, but there's a tradeoff and it depends on the situation). Your precision vs. flow rate can be chosen by your choice of nozzle size.
2. By varying the fuel and air partial pressures, you can create a radically different propulsion environments and choose whatever impact speeds and temperatures you want for your material - from over a thousand meters per second to a couple meters per second, from thousands of degrees to room temperature. This means your printing materials choice is almost, so long as they can be blown through the nozzle and caused - at some temperature/impact speed combination - to stick to the target. Plastic, metal, fibers, glasses, blends of materials, even food - almost anything, from the same printer.
3. Contrary to extrusion printing which generally produces parts inferior to molding, thermal spraying is used because it produces layers *superior* in properties to what can be otherwise achieved. Major factors in this are the capability to have the particles impact at high speed where desirable and to use materials in situations that normally cannot be done any other way
4. Thermal spraying gives you the ability to do the finishing work from the printer itself. That is after all what it's most commonly used for in the first place - coating with layers to resist scratching, water, corrosion, etc. By proper choice of materials, the same device could paint or sandblast your structure, even sandblasting away temporary supports if your control algorithm was smart enough. You could paint wires onto a PCB, texture a gripping surface into a piece of plastic, pretty much anything.
5. In terms of users providing their own material to print with, a coffee grinder and a sieve is all that one should need for a spraying-based printer. Rolling your own filaments for an extrusion printer, by contrast, is anything but trivial.
Downsides of course would include heat, fuel requirements (either a natural gas connection or fuel bottles), noise (requires a compressor), and potential dust or gas byproducts, depending on how you'd be using it. Hence it would have to be enclosed with filters and sound dampening, and probably be much more suitable for a garage or industrial setting than sitting on the desk next to your computer. Also, the nozzles don't last forever, but then again extrusion printers have set the bar pretty darn low on that one. Lastly, it's not a very fast process. On the upside, that means more precision on your Z axis.
Anyway, I have trouble picturing any other technology for laying down materials that offers that kind of potential versaility. Now, it wouldn't be everything - t's not going to do lithography, it's not going to assemble pieces that need to be made out of multiple parts, etc. But nonetheless, I simply can't envision extrusion as being the tech that ultimately gives people what they dream of from 3d printing (something close to a "replicator"). This potentially could be that.
That's sort of like the saying, "My keys are always in the last place I look!" It's a truism; it's as if we're supposed to believe that there's people out there who are looking for their keys, find them, and then decide to just keep on looking for them.