Inspired by the [del]record-setting hijack[/del] discussion in THIS thread by Tuckerfan and Lobohan. Is this the next big paradigm shift? Something that’s going to introduce a “gee-whiz” alteration of how the world works, akin to the original development of machine tools?
I started reading the thread, but it’s mostly about Star Trek and is many pages long. If there is something in there you want us to read, understand, and discuss, then paraphrase it, quote it, or at least link to the specific part.
Umm… I kinda thought the whole point was to discuss the subject I gave in the title up there in THIS thread, but to briefly summarize: In discussing the plausibility of manual construction of star ships vs robotic and/or autofabrication, Tuckerfan and Lobohan got into an debate about the merits of universal fabricators vs. dedicated mass production facilities. Several posts revolved around emerging autofabrication technologies, which in principle hold out the promise of being able to build something close to a fictional “replicator”. Particularly there is the idea that fabricators may be approaching the “desktop publishing” level of home use rather than as purely industrial technologies.
Not going to read through the thread either, but I’d have to say that ‘autofabrication’ (by which I assume you mean some kind of nano-technology based self replicating assembler/disassembler) is definitely not on the verge of changing the world since it’s not on the verge of being a real, viable technology. I’m not sure we’ll get anything like this in my lifetime in fact, though I know great strides are being made in nano-tech, and some amazing things are going to be coming out in the next few decades (I’m still holding out for DNA repairing nano-bots to extend my lifetime by decades if not centuries :)…but then hope springs eternal).
-XT
I have absolutely no doubt it will change a LOT of things, but like the internet and processed cheese, it will find a niche.
You should not buy your catfood online. You should pay your bills that way.
You should not make fondue out of Velveeta. You should make Mac n Cheese with it.
Likewise: You SHOULD print out your cellphone, kidneys, OLED displays, and car keys with a rapid-prototyper, you should NOT use it to print out the Enterprise. 
Lumpy Yes it will be a paradigm shift. A confluence of inflation (reduction of the value of the dollar) and a deflation of the costs for consumer goods, IE iPod price reductions, combined with newer and cheaper methods of mass assembly will result in a world where consumer goods are so cheap and plentiful that we change fundamentally how we interact with them. I don’t see autofabrication a la Diamond Age or Star Trek entering the world within this century, but I do see industrial fabrication plants becoming the way machine shops are. IE, where you can have people produce circuit boards at the local machine shop. There is already a bit of this, but it’s a niche market.
Companies like emachineshop will become more common. Custom modification to consumer goods will become more ubiquitous. People will be able to interact with a machine shop in the way they interact with their architect or contractor today making specialized industrial goods available at the consumer level, but you’ll still require a vendor who owns the expensive equipment to assemble your order.
Even if this occurs you’ll still have to deal with parasites and viruses.
It depends upon what you mean by “verge.” The fabbers we have today are (roughly) comparable to the early computers which appeared after WWII. The advantage modern fabbers have over those computers is their small size and that we understand what they can be used for.
Right now, fabbers can make not only plastic parts, but metal, human tissue/organs, electronics, casting cores (out of sand or foam), and food. They are, however, not cheap. Costs range anywhere from around $14K to several hundreds of thousands of dollars. The fab@home unit is estimated to cost around US $2300 and its limited in what it can do in comparison to professional units.
There are efforts underway to make ones which can produce a fully finished product on the order of a cellphone, but the technology is not going to stop there. Its still in the “big idea” stage, but there’s proposals to use fabbers to build large scale solar arrays. Folks have also kicked around the idea for constructing colonies on the Moon and Mars.
Once the machines reach the point where they can handle a wide variety of materials at one time, you will see them building everything you can imagine (as well as things you can’t even imagine). They simply offer too many advantages over conventional manufacturing methods for them to be ignored. For one thing, the eliminate the papertrail associated with “mission critical” items like we have now.
I worked in a shop which made fasteners for Boeing and the amount of crap tied to those jobs was huge. First of all, we couldn’t source the raw materials from just any metal supplier, it had to be one certified by Boeing, and every order came with a mountain of paperwork, tying it back to the mine where the ore was extracted from. We had to used Boeing certified tooling on Boeing certified machines, and all used/broken tooling was turned over to Boeing, as was all scrap/execess material. Boeing also had inspectors to observe that we followed all proceedures properly.
The reason for this is that you simply can’t afford to have a nut fall off a jet at 30K feet. The detailed and extensive nature of the paperwork was so that if the fastener failed, they could trace it back and pinpoint the problem. Mind you, it did not take a fatal accident to start an investigation, a fastener just had to fail before the end of its expected life to trigger an investigation, and the guys were thorough. They wanted to inspect your measuring tools, the tools you used to work on the machine, the machine, and generally probed you about everything other than what you had to eat that day.
A fabber which can accept raw materials and spit out a finished product cuts all of that out. You don’t have to do a backtrace through the contractor, sub-contractors, sub-sub-contractors, the various distributors supplying everyone, their suppliers, etc., etc., etc. You’ve got the fabber and the mines, that’s it. That’s a huge time and money saver.
Then there’s the personnel costs saved by using a fabber. Forget the wages and benefits side of the equation, just think about the hassles of having humans work on things. First of all, they have to be trained. Even someone who’s got years of experience as a machinist can’t be pulled off the street and allowed to tap holes. Why? Because the specs dictate everything you can possibly image. Not only the size and location of the hole to be tapped, but the RPM of the tap, the amount of cutting oil (as well as age and type) used, and which side of the part you start the tap on (even if its something as “simple” as a nut), to name but a few things. They do this so that they can more accurately predict when the part will fail and remove it from service before it has a chance to fail.
Next, you have the variablility humans put into the equation. Nobody can duplicate exactly how they did something, so this means that you’re going to have an unaccountable error factor in your estimates for how long a part should last. Not to mention, humans have the annoying habit of ignoring the specs. The spec might say you’re supposed to change the tap out after 10 parts, but the human might not feel like changing the tap then, or might forget.
Finally, there’s the crap which humans leave behind on everything they touch. Dead skin cells, oils, food residue, fecal matter, and more. All of that goes away with fabbing.
Presently, aircraft (and spacecraft) are held to the highest quality standards possible, simply because the cost of having somehing going horribly wrong. Now, lets say that a “Model-T” fabber can build an aircraft with the same level of quality as humans presently do. It does this without the help of hundreds of thousands of human beings. The cost savings afforded by this are enough to justify building one of those fabbers.
But, think about this: You can’t presently buy a car built to the same high standards as an aircraft. You can get close, but even a Ferrari isn’t put together as well as a fighter plane. With a fabber programmed to make cars and not planes, you can. That would be a huge selling point for the cars. You’d be able to know almost exactly when a part was going to fail. No more need to spend the money to have the car checked out before going on a long trip, you’ll know that its okay.
There’s also huge savings for the car makers, and not just from being able to ditch the humans on their payrolls. A customer comes in complaining about a problem with their car, you’ve got a perfect record of the build process of the car (which car makers don’t have), you’ll easily be able to find out if there was some kind of design flaw, material problem, or if the customer has done something in violation of their warranty. No more of this “Is it gonna break or not?” or “Why’d it break and who’s fault is it?”
Additionally, if the market wants SUVs one day, and then convertables the next, no problem. Just load up the new program and you can be turning out convertables as soon as its done loading the appropriate software. Not to mention, adding customized features to a car will be a snap. How many people would pick a car built the traditional way over one that could have exactly those features and options they wanted, for either the same cost, or perhaps less?
I’ll add that all of the Intellectual property law that’s being created now for things like books and music will extend to fabbing in the future. What’s the cost when someone downloads the description of that new Apple MacBook and fabs it out? Heck, it wasn’t STOLEN, nobody was HURT, but when you can fab the physical device as easily as you can copy a file, what’s gonna happen?
Well since we are highly unlikely to see personal fabs held in people’s homes, the companies that perform custom fabrication can be governed by regulations just like any other business. Likely Apple will just provide their product cheaply enough with enough customization options that there will be no point in seeking the product elsewhere.
Open source fabrication? You can operate a computer today without any commercial programs. If home fabbers ever becomes reality, a huge culture of creating and sharing fabrications would likely emerge.
You mean like we are highly unlikely to see people ever own personal computers ?
Well, as the intartubes are showing, intellectual property law as we now know it, is pretty much a goner. It won’t go all at once, however. You will probably have, for things like consumer goods, a patron system for the designers.
Here’s a hypothetical situation. Say you’ve got fabbers which can turn out consumer sized goods, PCs, TVs, etc., and everybody has one. Now the people who design consumer goods know that the moment they release one of their designs to the public, its going to get ripped off, so how do they maintain the “exclusivity” that their design has? A couple of ways come to mind. One is that they constantly rework the design and have a new one ready to go every day. So that people will pay them to be the first to get the design, and when someone shows up with their knockoff, they can look at them and say, “Oh, that design is so yesterday. You really need to get with the version he released half an hour ago.” Another way is for them to provide you something in exchange for the purchase price which can’t be duplicated easily. It’ll be a long before we have home fabbers which can spit out copies of DNA. So a designer could snail you a snippet of his DNA (encased in a nice glowing housing) and you could plunk that into the item you replicated. Anyone wishing to verify that you’ve got a real Archibald Q. Snootington III design and not a cheap knock-off, could scan the DNA section embedded in the item with the DNA scanner built into their cellphone (yes, we’ll have those, too) and see if its real or a knockoff.
People already have them. The base price starts out at $2,300, and goes up from there.
Feeding iron molecules through a tube to a home is a little bit more difficult than transmitting binary signals over a fiber optic line. I’m not saying it’ll never happen, I just don’t see it happening within say the next 20 years.
I stand corrected.
Then I change my stance to:
It will change the world MORE dramatically than the personal computer or the industrial revolution.
The Diamond Age should be mandatory reading in HS.
So you just buy the occasional can full of feedstock for the machine.
You’d need multiple types of feedstocks to make complex items.
The early models, yes, but not the later models. They’ll be able to economically sift through the garbage you dump in them and spit out what you want.
I doubt they will be able to do that in our lifetimes. I’ll be shocked if this becomes reality…pleased but shocked.
-XT