Desktop 3D printing: The milk analogy

I was thrilled by the response to the Singer problem post so I’ve decided to once more share another one of my tortured analogies with you.

The Milk analogy

Milk is a huge business. A typical US consumer consumes 178 pounds of milk products a year and drinks 21 gallons of the milk. In 2007 in the US alone the dairy industry had revenues of $99,700,000,000. There are approximately 34,000 supermarkets in the US & 13,000 small grocery stores.
In addition there are many other retail outlets that sell milk. Many of
these milk retailers are well located and have long opening hours,
while others compete on price. It is therefore easy to get milk in the
United States.

There is however a huge business opportunity here.
I will give you this business idea, completely free of charge and even
get you started with the marketing of it. This business idea will turn
the dairy industry and milk consumption on its head as well as
revolutionize food retailing and have a lasting impact on society.
Riches and fame await for the person that is lucky enough to read this
and able to capitalize on this idea.

Major revolutions in
retailing and consumer behavior typically often when distribution
problems are solved in inventive ways. Milk distribution in the US
seems to be an efficient affair with high barriers to entry that
depend on scale and scope. That is until you apply real insight and
inventiveness to the problem. To be “at arms length” at all times to
the customer is the dream of any distributor and retailer in any
industry. You can do this for the dairy industry by implementing this
simple idea:

put a cow in every home.  

Consumers like milk and
respond very well to “freshness”, wholesome, healthy and immediacy. A
cow in every home will allow these consumers to consume milk in the
most healthy, fresh and immediate form. Furthermore, in the long run
they will be able to drink this much fresher much healthier much more
available milk at lower cost than store bought milk. The environmental
benefits would be self evident. A cow could in addition to being a milk
dispensing device also function as a household pet. The cuteness and
docility of cows will drive widespread adoption.   

distribution channel will focus directly on the consumer and allow for
high margins by cutting out the middle man. Furthermore by supplying
the inputs for the cow such as hay and “superfood for your cow” you
have a potential high margin secondary revenue stream similar to
Proctor & Gamble’s razorblades model. If you act quickly the
“aftermarket” in cow accessories such as pens, toys and grooming
equipment could be yours also.

Some of the skeptics will invariably see problems with this idea. They will say things like:

81% of the united States is urbanized.
In a city home there simply will not be any room for a cow. In suburban
homes there might in some case be room but it will be difficult. Cows
also produce a lot of waste, they stink and specialized training will
be needed to care for them. It also takes a lot of work to take care of
cows.The market will soon be saturated with other firms offering cows
for sale even if it does work.”

I will quickly repudiate any and all objections from these naysayers.

genius of this plan is to create: mini-cows. Through genetic
engineering you can create low cost, small footprint, low waste, low
maintenance cows. This proprietary technology is patentable and will
create barriers to entry. No one will be able to sell your excellent
breed of cow at the same price point and with the same low footprint.
The prime mover advantage in this industry will be enormous and while
others struggle to catch up and genetically engineer similar cows you
will be free to conquer and own the market.

As for the branding
and marketing I have some suggestions. I would not go for the Holstein
breed of cattle. Although iconic I would not consider this a wise
choice. The black and white patterning will tend to breed tired visual
and marketing as well as limit your brand. Tucows and Gateway
(remember when they both were cool?) both have copyrights in the tech
arena with the typical black & white patterning. A lot of dairy
firms already use similar patterning in their respective fields.
Instead I would suggest going for a Jerrsey, Brown Swiss Cow or Guernsey breed.
These are all good milk producers, docile, smaller than Holsteins and
have a distinctive look. This will make for easier patent and
intellectual property development and protection. By taking a
distinctive breed you can also dominate and define the “category” of
the in home milk producers more easily.

As for the consumer
marketing itself it would be simple. Focus on cuteness, utility and
originality. I would also advocate for a social media viral marketing campaign.

Take the photograph below, tastefully insert your brand and URL
somewhere and submit a link to the image on your site to services such
as Reddit and Digg and you should be golden.

Please remember to attribute it to Law_keven correctly and realize it is a Share Alike Creative Commons

As you will have gathered this post is not about milk at all but about desktop 3D printers.

Fist some background for you, 3D printing in a nutshell:

A 3D printer today costs from $10,000-$900,000. When I use the term 3D printing I mean it in the broad sense as explained here. 3D printers are typically the size of coke machines.
An experienced 3D modeler has to make a 3D file that is watertight,
2-manifold and with the normals pointing in the right direction. This
3D file has to be checked and often repaired. Then the machine software
takes this file and calculates how the machine will make it. The
support structures
and orientation of the part have to be calculated also. Next, layer by
layer material is built up into the resulting object. There is a manual
process for removing support material
(the brown stuff in the picture right) and many processes require
additional steps such as curing or hardening in some way. The printer
requires a skilled & experienced operator to be run efficiently.

Some machines can work in an office environment but the removing of
support material is often not suited for an office. Other machines
produce a lot of dust or work with toxic materials and I would not want
them within a city block of my home. Price of the machines depend on
build volume, build speed and the types of materials they can work
with. These materials also vary widely in price. There is no clear
“best” 3D printing technology or material, it all depends on the use

The Desktop 3D printing revolution will put small and easy to use versions of these machines in everyone’s home. Inspired by this T-shirt however I have begun to doubt that this revolution will be as easy, quick or complete as promised.

I work for Shapeways which is a 3D printing community which is based on
giving access to the most expensive machines for the lowest prices
through scale I must confess to definitely being biased and perhaps
suffering from confirmation bias. I also realize that I am exposing
myself to a Thomas Watson type of risk as expressed by this comment to the Singer post

But, with that in mind I do want stimulate discussion on desktop 3D
printing and point out that the path to the mainstream will be a bit
slower than anticipated. Furthermore, it will not be a case of
“laser printers were expensive” now they are cheap, ergo the same will
happen with desktop 3D printing. The drivers for the adoption of this technology will be
different ones entirely.

Just because a lot of people like a technology, just because it could change anything and everything does not mean it will be implemented successfully. Good ideas do not win just because they are good ideas. The milk analogy, hopefully, illustrates barriers to profitability and implementation for a commercial company attempting to put a desktop printer on every desk, in every home. Even if the obvious technical challenges could be solved, instantly and right now, would mini-cows or desktop printers in every household work?

The reasons why mini-cows would or would not work depend on consumer behavior, consumer perception, the ability and willingness of someone to engage with a particular new technology and the usefulness of this technology to them at any given moment. A device’s perceived and actual utility to a person making a decision to buy it, or not.      

The $5000 desktop 3D printer does not matter.

Its existence an sich will not magically make a market. If I created a $5000 consumer friendly desktop nuclear reactor tomorrow, would it sell? Regulatory issues aside and imagining that government bodies would certify it as safe. It would be mega high tech and interesting and would be sure to get a lot of techies blood flowing faster. The marketing launch would be immense. I’m sure that there would be people that wanted one. But, would a lot of people buy one? Would you? Why or why not?

To me the price point would not matter at all. The inventiveness and sheer possibility of it would not matter either. Even the fact that the thing would be an incredibly good investment for consumers would not matter. What would kill such an initiative ultimately would be that people are afraid of the word “nuclear.”

The technology itself and the possibilities, although endless, would be immaterial.

As for the desktop 3D printer market: The real barriers to adoption by consumers should be similar to the ones expressed in the Singer post and the drivers for adoption would come from the materials and the ability of everyone to design, not the machine itself.


Pin It


  1. fracai

    Like you said, it’s not the machinery that is the issue, but the design and materials.

    And while your typical individual may not become an expert modeler, I don’t think they’ll need to be. As a first step, there will be (are) sites like Shapeways that will sell models for home printing. Additionally, as software products improve we’ll see a point where we’ll have modeling apps that range from Photoshop (Blender, etc) down to Paint (Google Sketchup is close). You might not get the highest quality or detailed model, but it’ll be good enough for the average or beginning modeler.

    I also have no doubt that support material removal will also diminish as an issue. It’ll get to the point where water rinsing will be enough to remove every bit of support. I even see the possibility for materials where support isn’t needed, even for overhang. I have no idea how that would work, but I have no doubt that we’ll see it arise.

    I think the best comparison is hobby kits all the way up to kit cars and planes. I may not be able to design a model car, but I can buy the kit and assemble it requiring only glue. When I was younger a family friend put together a plane from a kit. He bought the plans and fabricated components and assembled it. He didn’t have the expertise to design the plane, but assembling it based on plans was within reach. The same is true even for sewing, now that I think about it. You can buy patterns to assemble. The assembly involves skill, but the design is taken care of. With 3D printing assembly will be as easy as changing the “ink” cartridge and pressing “print”.

    This makes me think that perhaps the next step in the revolution will be corner Shapeways shops. You pop in with the design you bought online and walk out a few hours or days later with a fabricated product. Instead of assembling your IKEA dresser from the pieces they provide, you’ll just print the design. I think we’ll get that before we see that before we see an affordable desktop printer.

    The nuclear scare, by the way, is diminishing and I see more and more of the populace coming around to the idea of building new reactors. It’s a downright shame that the US hasn’t built a new reactor since even before the non-event that was 3 Mile Island.

    1. Joris


      As easy as Sketchup is, I would like it to be even easier.

      I also would agree that some of the technical challenges such as support removal and the software to make it easier to check and print will come eventually. But, I do think that someone wanting to spread desktop 3D printers should concentrate on improving these things.

      I like the kit idea, I think that this is where mayor improvements will occur. By letting someone use my designs I can accelerate both our developments and make desktop 3D printing more useful for everyone. Again: this is an example that proponents of desktop 3D printing should spend time on.

      The whole “3D printing copyshop” is an interesting idea too. Funnily enough there only seems to be one example of it in Japan and the Fablab idea of course is very similar.

      The nuclear thing was meant to be an example because focus groups and surveys have shown time and time again that the word elicits almost impossibly negative responses & feelings from people.

    2. Peter Hermans

      A while back I made a post on my own blog about examples of 3D software made easier/more intuitive:
      Especially examples like (smooth)teddy, that are aimed at kids might open up the door to home-3D-printing (and design).

      About the 3D printing copyshop, here’s a 2D copyshop in the US that is going to offer 3D printing as well:

      P.S. Wonderful article; loved the “I can gives milk” :D .

  2. Felix

    I think your analogy is somewhat erroneous mainly for the following reasons:
    1° You compare a “concept” for producing basis food/mass product against a technology conceived to produce very small numbers of specific non-food parts….
    2° You compare a “production concept” which involves a living animal against a completely automated electronic/mechanical machine…
    3° Would your “genetically engineered mini-cow” directly produce pasteurized milk ?!? ;-)
    4° milk production isn’t very attractive/creative (in sense of an hobby). Creating objects to print in 3D could (can already) be a hobby… Creating objects seems be an attractive hobby, see the success of Craft ROBO and Cricut for scrapbooking and other usages.

    We cannot guess the usage pattern of a new technology / machine because we are always using the things we already know to imagine the future. One good example for this are the drawings that were made in 1900 depicting the life in year 2000. The people are dressed like 19XX and are playing -underwater :-) – croquet (croquet was a very trendy sport then) !

    1. Joris


      Perhaps the cow could produce pasteurized milk, but you would have to boil it.

      With regards to 4, well maybe milking should be considered a hobby. If one created such a hobby, the mini-cows would be more successful.

      Likewise, if you would want to make desktop 3D printing successful then I would suggest you do what you can to expand the Maker community and teach them how to 3D model. This to me would make sense and might make desktop 3D printing a reality. Not the existence of the machine itself.

      I too have seen a lot of those “peering” into the future photographs and prints: a technological paradise consisting of balloons and very high tech stream engines is the one that sticks in my mind. So, I recognize that I can not predict the future.

      But, I do think that modeling and the ability to model and design will make this either a successful tool or not. I think that developments such as teddy, that Peter mentions on his blog are where the future lies. Here Bryan made a model with Teddy, a super simple tool:

      These tools will make the future.

      The DVD player revolution would not work if you had to teach a large group of people to make movies and had to teach everyone how to use the machine. Imagine: “yeah, one of these DVD players, really cool but you have to spend a few months learning Maya to use it.” Or “yeah, I really want one of those DVD players, too bad there are no movies to watch on them.”

    2. Felix

      I don’t think you can’t compare a (future) desktop 3D printer with a DVD player. DVD players have (primarly) been designed to play films that already exist. DVD was created to replace VHS; 3D printing would be a completely new device in his function and usage (DVD player wasn’t). The targets for these devices aren’t the same at all because the first one is a player and the second one is a manufacturing tool ! Of course, a 3D printer could act like a sort of “player” by printing only downloaded files, but it would be boring unless the online stock is HUGE… Imagine if your desktop printer (2D) could ONLY print downloaded documents (from the manufacturer website)… It would be useless! If you want to find a machine which has more similarities with a 3D printer, you should look at Craft ROBO or Cricut. (I couldn’t resist to put a link to an astonishing example I found on the net. in French. )

  3. Frank Kleemann

    Hi everybody,

    very interresting discussion! I think I have to throw in my humble thoughts…

    To be honest, I don´t think that 3D Printers (I will call them 3DPs now) will spread on a mass-/mainstream market that soon. I will try to explain why.

    A lot of this discussion is about the problems of handling such a machine or the difficulties of creating/acquiring printable 3D files. I also think that these points create lots of drag and prevent 3DPs from being a popular mainstream gadget. But what me really concerns are the capabilities and purpose of home 3DPs – and 3DPs in general.

    As long as even the pricy, big machines still have major flaws – bad resolution, poor precision, weak, unhealthy, extremely expensive materials, or hadly any choice of material (at least one of these flaws applies to any technology) – what can we expect from a cheap 3DP that should fit into a corner of a flat? Many people are very optimistic (which is generally fine) about how fast the development of 3D Printers will be. I don´t share this view. Just look at the existing technologies and how they developed in the last years. There have been many small steps of continuous improvement, but only few that really influenced or expanded the usage of 3DP. And I guess the majority of them happened in the high price areas, especially in metal printing.

    If we assume (at least I do) that in a foreseeable future personal 3DPs wil only be able to produce small and more or less fragile objects: is this really what huge amounts of people want to have?

    Many people compare 3DPs with with 2D paper printers and say „look ,they were so big, crappy and expensive, and now everyone has got one at home“. That is true of course. But I really don´t think that this is a good comparison. Paper printers are cheap and useful. Even if you don´t plan to create any own content there are plenty of useful applications -printing your monthly bank account statement, the invoice of your last shapeways order, an interresting article you found somewhere in the web…

    I would go so far to claim that personal paper printers mostly are used for simple, but „useful“ purposes. In many other cases, especially when we have higher expectations on the end product, we still prefer professional printing services instead of using our own one. Vacation photos, poster size prints, print jobs with 100 or more pages,… We get them faster, cheaper and in a better quality if we bring them to a specialist who uses a big and expensive machine to produce them.

    I can´t really see useful applications for a cheap home 3DP with limited capabilities. And I think the possibility to print some nice decorative objects and birthday presents would not convince many peolpe to spend many hundreds or even thousands of euros for such a machine.

    I don´t want to say that such cheap 3DPs with limited capabilities don´t have a purpose (and with this, a market). In many professions there is a high need for „materialising“ 3D data cheap and fast. Think of architects, engineers and designers (as I am) that want to create quick models or ergonomic checks. Or many scientific professions that need to visualize generated or collected data. Often these objects neither have to be really stable nor to be printed in a fine resolution. It is just draft: build, evaluate, throw away. The biggest advantage would be saving time and being independent: not having to leave the house (like in the corner 3D printing shop model) or even to wait for the parts until they come via post. I would love to have direct access to a machine like that in my daily work as a designer. And I think that they will spread in professional surroundings pretty fast.

    1. Joris


      good point. There could be an alternative revolution in desktop 3D printing whereby instead of making end products they could be used initially for cheap prototyping only. This would mean that they could still get widespread adoption but only in certain niches and widespread adoption would follow later once the capabilities of the machines improve.

      This seems to be very ‘organic’ to me and more logical than the whole “in two years everyone will have a machine on their desk that can make anything.”

    2. Peter Hermans

      After Joris’ initial post I got thinking and got to roughly the same conclusion as Frank: you need a killer application.
      However, I do see a couple (not many I admit) uses:
      -Prototypes of existing products. The threshold for buying products online, without any other reference than pictures and specs could be lowered if you have the ability to print a mockup to validate whether you like the aesthetics/ergonomics etc. This would have to be fullcolor, fullscale and preferably with reuseble materials. Products I’m thinking of: mp3 players, cellphones, digicams etc.
      -Lego’s. Design your own Lego kit as is already possible, but now with the opportunity to design your own custom building blocks. For this application your printer would have to be able to print fairly high-quality plastics in terms of material quality/strengths and resolution. The printer could be very small (no larger than the biggest Lego block). For this application the question still remains if Lego could better solve this problem with their own industrial 3DPs and shipping the custom parts or by selling personal 3DPs.

    3. Joris

      So Peter & Frank,

      are you both moving away from the idea that a Desktop 3D printer is., for right now anyway, not a “make everything machine” but rather a prototyping one?

      I keep wondering if “the dream of the ultimate machine” or a “killer app for right now” will be better for getting 3D printers on desks.

  4. Felix (different one...)

    Very interesting discussion. I’ve just started with 3D modelling, and I mostly do engineering/robotics type work and lots of home-built projects. Would I privately buy a $5000 3D printer? Don’t know, but probably not. For $1000, absolutely yes. I’ve been toying with the idea of privately buying a $3000 CNC machine, but was mainly held back by its weight – you don’t ever want to have to move house again with a 300kg piece of machinery…

    Where I see cheap 3D printers find their niche is with makers, hackers and geeks like me, who want to do fairly simple things very quickly, and maybe a better job some time in the future.
    Here’s an example. At university we have a $150,000 PCB milling and rapid prototyping setup, complete with copper plating and solder mask. We can do 2-layer and 4-layer boards, with through plating, and it looks and works almost like a professionally made board. The quality is good enough to use almost any fine-pitch surface mount technology. Do I use it much? Yes and no. I have done very few boards where I used the copper plating and two layers. The main reason is, it is a bit of work, makes the process more difficult, and takes extreme care and some experience to get it right. Also, while it is possible to achieve 6mil clearances (0.15mm), it might need reworking and can be a pain at times. So, for the final, high quality 2-layer or 4-layer version I would in most cases send it to a professional place where I can get it made perfectly for $30 or so.
    However, I have used the setup very, very often for a quick test of a design, often only single layer boards, 3 times bigger than the final thing. I can think of a new design in the morning, have the schematics and board layouts ready by lunchtime, spend less than 1 hour in the workshop to make the PCB, assemble it and test it within the same day. Often the board is good enough for the purpose. Otherwise I would fix all the problems and send it off for the final version (which takes a week or so). So while I’m waiting for that one board, I can theoretically go through 3-4 design iterations if necessary.

    That’s where I see 3D printers. Often I need a dead primitive part – let’s say a mounting bracket, or a similarly trivial fixture, of a specific shape, with the holes in the right place. Ideally I would just draw it up, and 3D-print it – preferably on the same day. If it sorta works but isn’t quite strong enough, etc. I can still send it off to Shapeways, or other manufacturing places, that would make me a much better quality version. However, I would be able to use the part for those 7-10 days it takes to get it made. It’s like printing your favourite photo on your inkjet – maybe you want a professional poster print later, but the inkjet print is quicker and doesn’t require planning ahead. Maybe it’s just me, but the requirement to have to send something off and waiting for a week often makes me think of other ways how I could make it right now, even if it’s lower quality. I just like to get things done when I’m on a roll….

    Having said all that, I think Shapeways is a fantastic step forward. Having instant pricing and online check out is (for me at least) a much lower hurdle than having to get a quote and entering an email discussion about manufacturability. Before Shapeways I also never had a clear idea how much 3D printing really costs, not even the order of magnitude. It found myself often in the situation where the quote was at least 10 times more than I was willing to spend, so the whole process of obtaining the quote was a complete waste of time, and I ended up producing the part in a completely different way. Having the Shapeways service at my disposal, with transparent pricing, enables me to think of new designs in a different way, because now I know that I can count on making free-form quality plastic parts any time I like, which wasn’t really an option before. Now I can make my home-grown builds look nice by making a nice enclosure. Etc. I’ve also considered printing negative molds for resin casting, but haven’t tried that out yet. Do I still want a CNC mill? yes, sure – because for some parts the surface finish just has to be good enough, and sometimes I need strong, cheap metal parts.
    How about a CNC milling analogy? CNC mills now start at $3000, and some of them (not the one I’m after) fit on a desk, weigh 15kg and are perfectly fine for making small precision parts in metal. Why doesn’t everyone have one at home already? Because for a lot of hobbyists $3000 is still a lot of money, and they need a lot of experience (toolpath generation, clamping, multiple work steps, etc). However, I think 3D printers are mostly easier to use, can make far more complex parts, and, if they ever make it to the desktop, will be almost as easy to use as a laser printer. 3D-modelling software is quickly catching up as well (I’m using ViaCAD and it’s very, very easy to learn, and very cheap – design was printable straight away too).
    Will every house hold have one? no, definitely not. Will every model train enthusiast, model airplane builder, sculptor, tinkerer, hacker, geek and maker have one? I think so, once the price drops below $1000 and that machine can make structural parts. If you look at websites like Sparkfun electronics, instructables, CNCZone, LumenLab, DIYAudio, etc., there are a lot of people like that in all areas.

    For the wider community, I think it’s a chicken-and-egg problem. If many house holds had a 3D printer, I’m sure there would be online companies offering printable 3D-designs for toys (someone said Lego…), spare parts (how about that battery cover for remote controls that got lost?) and lots of other things I can’t think of. I think Shapeways has the right idea with allowing people to share their designs, and in fact it’s a viable path to do all that without requiring everyone to have a printer. So maybe, in 10 years or so, there’ll be enough downloadable 3D design offers that people will seriously consider buying one of those $500 desktop 3D printers…

    By the way, I have a sewing machine, mainly for fixing tents and making mesh seats for recumbent bicycles… bought it for $3 and fixed it :)

    1. Felix (different one...)

      One more thing… :)
      Reading through the “Singer” blog post I noticed that there is a recurring argument that 3D-modelling is hard to learn, and may take months. I don’t think this is entirely true. My own experience was that I tried and gave up on Blender, I tried and gave up on Pro/E, I tried Rhino and it wasn’t so bad but I didn’t get anything done. Then I tried Solidworks, and managed to do something simple in about 3 hours, and within 2 days I had a design for quite a complex 3D-shape with smooth surfaces finished that I previously had modelled in clay (this is for enginieering / product design type stuff, not artistic sculpting). Problem is though, Solidworks costs about $10,000… However, now I found PunchCAD’s Viacad (starts at $100) and Shark (download the free demo – Viacad demo is fully functional for 14 days). I spent 3 hours with almost no prior experience other than my Solidworks attempts once, and I had a nice design of a brushless motor armature and a propeller, which I printed at Solidworks. My point is, it may take a lot of time to learn some CAD tools, but there are very easy to use programs out there. Anything that offers fully implemented CSG (solid geometry) will get most users started very quickly – I find it very intuitive to start with primitives like boxes, cylinders and cones, and use adding/subtracting, slicing and filleting to shape it as I want it to be. Mesh-oriented programs may be more flexible, but in my experience make life hard. I assume it gets easier after some training.

      I do agree though that the average consumer has no interest in being creative, and rather buys the design/gadget/clothes from whatever renowned label. That way they don’t have to own up to their taste and skills, and can just say “but, xyz designed it and I paid $1000 for it”. So we’re not looking at the market of “anyone who is using plastic parts”, but “everyone who is into creating interesting things, including free-form plastic parts”. Those aren’t necessarily the financially strong people though who easily dish out $5000 on a plasma TV.

    2. Joris


      LOL about the moving with 300K of machinery. But, I do think that battles such as this will be won or lost with considerations such as that one.

      I like the idea of having a simple “lets test a design” machine and do think that at least initially the desktop machines would be good only for that.

      If you compare it with CNC, desktop 3D printing does seem more apt for “in home” use.

      I agree that the “makers” will get one as soon as one comes out that is OK. They will drive adoption. And I completely agree with the chicken and egg problem. It would seem to me to be a network problem and indeed lots of files for lots of people will solve that.

      Maybe we should like totally help out by doing something like that…

    3. Felix (different one...)

      > Maybe we should like totally help out by doing something like that…

      lol… yes, someone really should set up a business to get things started… ;)

      Btw. I just followed your link to MakerBot… seems there already is a < $1000 3D printer. Do I have to own up now to that I would buy a $1000 machine?? But, looking at the RepRap extrusion results, and my brand new, smooth, robust and precise Shapeways print on my desk… just not sure if it’s going to be good enough :) You guys ruin all the fun… ;)
      Seriously though, I have noticed a while ago that I seem to be stuck in a boot-strapping spiral. I had a few projects going and thought “this would be much easier if I had a CNC mill” so I started building my own CNC mill (not finished yet… as so many things). Of course to make a CNC mill is much easier if you already have one. Or at least a mill and lathe of some sorts… Now I’m thinking of making a 3D printer… but I’d have to finish the CNC mill first to make the parts… :)
      It’s probably more productive to actually start making things instead of making the tools needed to make things. Yet, making the tools is fun too, and having tools around opens up new avenues for making things.

  5. robot makes music

    “The world will never need more than 5 computers” – IBM
    “No one will ever need more than 640k of RAM” – Bill Gates. – the desktop sub-$1000 build-it-yourself then replicate 60% of the parts you need to make your second one. Fully open-source technology, verified and working.

    The 3d desktop printer revolution is coming faster than you may think. Sure, the RepRap is clunky and ugly BUT IT WORKS. And this is just the hacksaws and rocks version. Their plans include eventually printing metals, then circuits, transistors…. The goal is to have an entirely self-replicating 3d printer that can also print anything of comparable complexity.

    Also, there’s the d_shape BUILDING printer. Yes, it works, download the PDF.

    The future will be replicable, and it will be here pretty soon.

    Oh, and for the “where’s my jetpack?” crowd, I guess ya’ll missed the guy flying over the english channel with one? – No, it’s not a VTOL, but seriously how nitpicky are you going to get? It’s only a matter of time.

    1. Joris


      I love RepRap and what they are doing(although when I wanted to build my own one it was going to be more than a $1000). I am not trying to say that desktop 3d printers are not possible, clearly they are. I want to find out if they will be on everyone’s desk.

      D_shape is awesome, I interviewed the inventor yesterday and will put that live as soon as I can.

      The future is definitely replicable, but when and in what form and for whom? Shapeways is all about creating a world where anyone can make anything. I have drunk the kool aid but am not drunk on the kool aid.

      Yes, but where is my Jetpack. Why do I bike to work? Network effects, adoption rates, marketing, perception these get in the way of or facilitate technological revolutions.

    2. robot makes music

      You bike to work because it’s currently the healthy alternative – if you didn’t you might be too out of shape to get lofted by your future jetpack, right? ;) I’m guessing you could probably go buy one of these, but you’d also need a plane to launch from an a parachute target at work – but give it 50 years. You bike to work so you’ll be healthy enough 50 years from now.

      And if you’re not, you can always go buy a robotic exo-skeleton to help you with the jetpack. :D Well, currently only selling to a limited crowd in Japan and “soon” in Europe. But you definitely get the feeling of future shock when you see someone saying “Yeah, we’ve already started selling some robotic exoskeletons.”

      And, of course, advances in robotics will help advances in robotic ‘printing’ replication. Our first generation of printers are entirely enclosed. What happens when someone builds one into robotic arms on an assembly line?

      It’s not that I think the current technology is all that, I am just continually pleasantly surprised by what our technology has accomplished. Sure, we were promised jetpacks by scifi dreamers in the fifties (and we have quite dangerous proof-of-concepts now) but no one forsaw gene therapy, using viruses for said gene therapy, nanotech, or, heck, computers and personal communications devices. Even as late as the 80s, William Gibson forsaw a global interconnected computer mess, but didn’t get cellphones, and has kicked himself ever since for not seeing that one coming – seriously, in a world with 3d rendered internet beamed into your mind through surface contact electrodes, he had everyone using payphones (it even took me a minute to remember the word for payphone).

      But think about it – how could the prices not drop drastically once “self-replicating except for the computer controller” hits.

      It may be 50 years (25 would be my actual guess) but 3d printing will be cheap and widely available, unless they’re some sort of scarcity issue, like with diamonds (even geology professors at a nearby university are willing to believe DeBeers has more diamonds in storage than in circulation in the world to help prop up the astounding prices people pay for stupid shiny chunks of carbon).

      i mean, the d_shape guy printed a gazebo for 60 GBP worth of sand and binder. You can’t even frame a wooden one that cheaply. And there’s no reason you couldn’t print wood, either, except that it’s a kinda crappy building material (don’t get me wrong, wood is great I just like it better in trees). And ten years ago this would have been laughed off as unrealistic. 20 years ago as impossible. 100 years ago you’d be the laughing stock of an entire community. Jules Verne wouldn’t even have touched something like this. A gigantic, spindly machine that zips back and forth spitting out sand into what will become a building. I mean, it sounds preposterous even now that I’ve *seen* it.

      Surely you remember when the internet was just a passing fad in the early 90s? Imagine what the world would have been like if that really took off… :D

      Oh… by the way… there WAS a time when the cow was in everyone’s back yard, and the milk was always fresh. My great grandmother lived that way on the very outskirts of town, less than a mile from where I live now, smack in the middle of this burgeoning city.

      On the other hand, you may indeed be right – there may never be any “3d desktop printer on everyone’s desk” – why? Like I mentioned above, when you build them into the arms of a robot… well, I guess you could ask the robot to sit on your desk and pretend, for old time’s sake. :D :D

    3. Joris

      LOL, Robot,

      But, an exoskeleton that really would make me unhealthy.

      I do know that we can not predict the future and that any attempt to do so has a much higher chance of making you look ridiculous than anything else.

      Good point on the payphone. A few weeks ago I was standing behind the service counter at the supermarket and the person in front of me asked for something. The clerk had to rummage around and finally produced a little yellow box. I was thinking “what that box.” It was a roll of film. Obsolescence, will catch us all with its speed as will the rapidity of technology adoption.

      The point I was trying to circuitously make is that: technologies are adopted by people, sometimes quickly sometimes slowly. This adoption rate has very little to do with the technology an sich but rather how people use and see it.

    4. robot makes music

      Ah, a very true point.

      Thinking about the robot equipped with 3d print nozzles on it’s … arm? manipulator? that might actually have more far reaching implications than a boring old 3d plotter, no matter how exciting that may seem today.

      Imagine: a robot with above attachment, and a 3d mapping laser attached to the end to orient it – could map out, say, a defect in concrete (crack or what have you) and then apply a properly shaped fix in situ. The fine-ness of the fix would only be limited by our ability to minaturize the print-heads, and if ink jet is any measure we can eventually get very fine indeed. When we have metal-printers, a potential end to welding as you can just print the two surfaces together (possibly – depending on a lot of physical and chemical characteristics of the materials – especially the more advanced ones).

      Oh, I had a thought about the barriers to entry for the at-home printer in relation to the modelling of the objects they’d want to print – model exchange/merchants! Once everyone’s equipped with a sweatshop in a box (hehe) on their desktop they can start paying for other people’s designs to be printed for their own consumption. And what about printing polycarbonate items? Imagine going to the eye doctor and getting your glasses printed right there, or potentially just getting your model on a USB key to take home and print yourself.

      I think, as 3d printed objects become more and more ‘ready to roll’ in the sense that you can print something finished except for the power source, we’ll see an upsurge in designer mass goods – go to home depot, have them print you a customiz(ed/able) drill onto a pre-fabbed electronic base. It could start as simple as a program to add say an embossed name to the drill and go from there. The important part is the materials are cheap to input even if the printer isn’t, so they can charge *bank* to start off with, and it’d mostly be profit, materials wise. And think about this – what contractor wouldn’t pay extra to have his company’s name built into his tools, making them unstealable. Go to the cops, “They all have this logo and design built onto them” and there’s no trouble recognizing them at the flea market or on Craigslist. The clever person would ask for some personalization on the inside of the shell in case the outside was sanded off. :)

      As far as the near-term, my prediction is we’d see bigger stores start to offer this stuff, maybe get to the mall within 50 years, but mostly remain as a desktop hacker toy as far as home penetration goes, until Wal*mart figures out how to give everyone a free printer (print them in the delivery van on the way to the house? that would be pretty awesome) and then charge for materials to be delivered and models, and then gets to pay less people to work in their stores and that leads to step 3, profit!!

      I mean, once you invent the completely self replicating 3d printer, they get cheaper to make every time you make a new one, especially if we start doing carbon-based printing and figure out a way to turn plant waste into nanotubes or something. Now that would be a world-changer – mow your yard, make your nanotubes, print just about anything. But this is more pie in the sky than coming anytime soon, even if it is technically in the realm of possibility (though I did see an article posted at school about how the navy figured out how to add some carbon to a particular epoxy and have massive amounts of nanotubes generated with no energy added).

    5. robot makes music

      Hey speaking of a cow in every home for fresh milk… now that we have the cow genome sequenced, we should be able to clone udders… then all we have to do is combine them with the following technology

      and bam, udder in a jar in every home! Or at least every grocery store, so you can just walk up and milk yaself a jug or two.

      I know our technology isn’t quite there yet but … it looks like it’s heading that way. That lung is *breathing* son.

    6. Joris

      Oh, totally this would be much more efficient, albeit much less cute, than my mini-cows idea.

  6. Shapeways Blog

    Shapeways wants to do for 3D printing, what CSI did for petechial hemorrhaging. Luckily we are not alone, there are many companies working towards similar ends. These fellow travelers all want to in some way enable personal production and give people th

Comments are closed.