There are multiple inventors, because additive manufacturing can be implemented in many different ways. In 1967 Wyn Swainson, still a student in Denmark, applied for a patent titled "Method of Producing a 3D Figure by Holography" that was probably the first kind of 3D printing. He completed his studies in chemistry at UC Berkeley, obtained a patent in 1971, and opened a company called Formigraphic in Bolinas, north of San Francisco. In 1974 Formigraphic (later renamed Omtec Replication) demonstrated the printing of a 3D object.

Charles Hull in Los Angeles is credited with inventing the first 3D printer. He filed his patent in 1984 when he was still working for a company called UVP in Los Angeles. His technique is known as "stereolithography" or SLA, a laser-based process that works with liquid resins. In 1986 Carl Deckard at the University of Texas invented Selective Laser Sintering (SLS) that in theory can print a wide range of materials: plastics, ceramic, metal, etc. Sintering is a technique that has been used for thousands of years to create everyday objects like bricks, porcelain and jewelry. The difference is that now we can use the laser. SLS uses a powder of the material and uses a laser to turn it into a solid structure. This was a key technology to make 3D printing more practical, but it took several years before SLS printers became competitive. Someone said that SLS is like building a home by placing bricks in the air. Deckard's idea had been preceded in 1979 by a similar invention by Ross Housholder in Las Vegas. In 1987 Michael Feygin in Los Angeles invented Laminated Object Manufacturing (LOM) or "paper 3D printing" because it uses paper. In 1988 Frank Arcella at Westinghouse in Pittsburgh invented Laser Additive Manufacturing (LAM) for making metal parts, a technique that used a high-power laser and titanium powder. The very popular technique that today is known as Fused Filament Fabrication (FFF) or Fused Deposition Modeling (FDM) was developed by Scott Crump in Minnesota in 1989, and uses a filament of molten plastic.

These are the official dates of their patents. Each of them claims to have invented his technique a few years earlier. As it is often the case with US inventions, some foreigners had come first. In 1980 Hideo Kodama (working at Nagoya Municipal Industrial Research Institute in Japan) had already filed a patent and published the general idea of 3D Printing in a paper titled "Three-Dimensional Data Display by Automatic Preparation of a Three-Dimensional Model". In 1982 Alan Herbert of 3M in Minnesota published a paper titled "Solid Object Generation". Alain LeMehaute working at General Electric in France filed an SLA patent titled "Apparatus for Fabricating a Model of an Industrial Part" a few weeks before Charles Hull, but Hull's patent is the one that stuck. In 1984 Yoji Marutani of the Osaka Prefectural Industrial Research Institute (OPIRI) invented his own version of stereolithography. In 1986 patents for 3D printing were filed by Takashi Morihara of Fujitsu in Japan and Itzchak Pomerantz of Cubital in Israel.

Hull introduced his first commercial 3D printer, called SLA-1, in 1988 with his Los Angeles-based company 3D Systems. Crump founded Stratasys in 1989 and shipped his first 3D printer, called 3D Modeler, in 1991 (FFF was patented by Stratesys, FDM is very similar but more open). In 1989 Deckard founded Nova Automation, later renamed Desk Top Manufacturing (DTM) that manufactured the first SLS printer in 1990, the Mod A, followed in 1992 by the Sinterstation 2000. DTM was acquired by 3D Systems in 2001, so today 3D Systems owns both SLA and SLS technologies. Feygin started selling LOM printers in 1991 under the company name Helisys, later renamed Cubic.

Each technique represented a further step towards the third industrial revolution. 3D Systems and Stratasys caused the conceptual revolution: we can turn a digital file into a three-dimensional object. DTM created the mass market, because SLS is a 3D printing technology that is both low-cost and high-resolution, and it can print virtually any shape you can think of; but not much happened in SLS until in 2006 EOS in Germany introduced two SLS models, the Formiga P100 and the Eosint P730.

At this point in time the 3D printers were still rudimentary and very expensive, and a lot of the pioneers had to close because they didn't make money. But 3D printing got some help from the US agency DARPA, that in 1990 launched the Solid Freeform Fabrication program to invent "tool-less" manufacturing, i.e. removing the traditional tools from the factory.

The USA was not ahead of the rest of the world in commercializing 3D printing technology. In 1990 OPIRI's technology was used by Computer Modeling and Engineering Technology (CMET), basically a subsidiary of Mitsubishi, to make the SOUP printer. In 1989 the chemical giant DuPont announced its own version of stereolithography, the Somos 1000 machine, that was acquired by Teijin Seiki in Japan to make its Soliform printer in 1992. In 1989 Sony and Japan Synthetic Rubber (JSR) formed the Design-Model and Engineering Center (D-MEC) to sell a 3D printer that used Sony's own version of stereolithography, Solid Creation System (SCS). In 1991 an Israeli company, Cubital, started selling the Solider System, using a technique that today is known as Solid Ground Curing (SGC). In 1991 Electro Optical Systems (EOS) of Germany introduced the STEREOS 400 system, the first business stereolithography machine, and it remains Europe's top provider of 3D printers; and in 1994 EOS introduced its SLS system, the second commercial SLS system in the world. In 1993 Denken of Japan introduced a stereolithography machine that was almost desktop-size and was relatively cheap.

Then there are the inkjet printers. The inkjet printer was popularized in the 1980s by Hewlett-Packard and Canon. A more complex kind of inkjet technology, called 3DP (also known as "powder and inkjet" and "Z printing") was developed at the MIT in 1993 by Michael Cima and Emanuel Sachs. In 1996 South Carolina's Z Corp introduced the first 3D printer based on the MIT technology, the Z402, and in 2000 Z Corp would introduce the first multicolor 3D printer, the Z402C. In 1997 Sanders Prototype (later Solidscape) of New Hampshire introduced the ModelMaker wax printer based on the MIT inkjet technology, the progenitor of the cheaper T66 that it introduced in 2002.

3D printers for metal parts are particularly important because so much of our world is still made of metal parts. A crucial invention was the idea of using a laser beam to fuse together powders so that they become solid metals. This technique is called Selective Laser Melting (SLM) and was invented in 1995 at the Fraunhofer Institute in Germany. Electro Optical Systems (EOS) merged SLM with its own SLS technology and with a technology from Electrolux Rapid Development (ERD) of Finland for metal powders, and renamed it as Direct Metal Laser Sintering (DMLS). In 1995 it introduced the first commercial printer for metal parts based on DMLS, the Eosint M250. The SLM technology was also commercialized in 1999 by Fockele & Schwarze and in 2000 by MCP Technologies (later renamed MTT). Collaborations by EOS with 3D Systems (1997) and Trumpf (2002) led to the Eosint M270 of 2004. By the way, when in 2013 Elon Musk tweeted the image of SpaceX's rocket engine being made with a 3D printer, it was one of these EOS printers. In 2006 MCP's SLM machine made parts with aluminum and titanium powders.

An incredible number of terms have been used for the technology of Direct Metal Deposition (DMD) invented at the MIT, a close relative of Arcella's LAM, in which high-power laser beams are used to "weld" metal powders such as titanium, nickel and cobalt. The general name now is Laser Deposition Technology (LDT), but also Laser Metal Deposition (LMD), Direct Laser Deposition (DLD), etc. One such DMD-derived process is Laser Engineered Net Shaping (LENS), invented in 1996 at the Sandia National Laboratories in New Mexico by Dave Keicher, and commercialized by Optomec in 1998. Luckily, in 1994 Sandia had approved a program to encourage its scientists to start companies and commercialize its inventions (the Entrepreneurial Separation to Transfer Technology or ESTT program). Objects created with LENS technology can be very big.

First of all, 3D-printing metals became more common although still very expensive: in 2002 Arcam of Sweden, a spinoff of Chalmers University of Technology, introduced the Electron Beam Melting (EBM) technique; in 2002 Precision Optical Manufacturing (POM) of Michigan started selling a machine using its laser-based Direct Metal Deposition (DMD). in 2007 Accufusion in Canada launched its own version of LENS calling it Laser Consolidation (LC). In 1996 Chicago's Sciaky had invented the Electron Beam Additive Manufacturing (EBAM) technology, similar to SLS but processing metal powders in a vacuum environment using extremely high temperatures (up to 1000 degrees Celsius). In 2009 it had introduced a service (not a machine) called Electron Beam Direct Manufacturing (EBDM), and in 2014 it began selling a giant metal printer for titanium parts.

In 2002 EnvisionTec of Germany started selling its Perfactory machine, which used a new technology. In 1987 Larry Hornbeck at Texas Instruments had invented Digital Light Processing (DLP), a system for high-resolution video projection; and in 1997 a company called Digital Projection had introduced the first DLP-based video projector. The process used by EnvisionTec was similar to stereolithography except that it used a DLP projector to shine ultraviolet light on the resin. This dramatically reduced the cost. DLP and SLA printers are sometimes called simply "resin 3D printers".

Another company that had a 3D inkjet printer was the Israeli company Objet Geometries, that introduced the Quadra in 2001. In 2008 Objet introduced "polyjet printing", the ability to combine different materials in the same printed part. This allowed the same part to be made in different versions with different materials.

Another application that was difficult for 3D printer was electronic circuits. In 1999 DARPA had launched the Mesoscale Integrated Conformal Electronics (MICE) project to revolutionize the printing of electronics, that had ended up creating a technology usually called "direct write". Optomec of New Mexico used this direct-write technology to develop in 2004 its Aerosol Jet machines to print electronic parts, and in 2009 Fujifilm entered the market for printed electronics with its Dimatix Materials Printer (DMP) that is very similar to Aerosol Jet. Of course, 3D printers were not able to compete with the highly sophisticated factories of Intel and Foxcon, but sometimes you just want to embed an electronic sensor into an object. In 2004 this was rare, but in a few years direct-write technology would become very appealing to 3D-print "smaft objects" for the Internet of Things.

There were also the first attempts at making the process cheaper. In 2004 Solidimension of Israel started selling a desktop 3D printer. In 2007 Mcor of Ireland introduced its Matrix printer, based on yet another paper-based process, Selective Deposition Lamination (SDL), but this one used ordinary paper as the build material. In 2007 Solidscape introduced 3D printer specifically for jewelry, the R66, yet another variation on the T66; very expensive, but also very useful because specialized for a high-revenue sector.

Meanwhile, Materialise in Belgium was one of the companies providing software for 3D printers. During the 2000s it began to offer a printing service for anyone who used their software to build a 3D model: no need to buy your own printer, just send us the model and they'll print it for you. This was the beginning of "cloud 3D printing". Their print centers now have more than 100 printers of all kinds (SLS, SLA, FDM, etc), producing almost one million parts per year for customers located all over Europe, one of the largest 3D-printing facilities in the world.

So all these developments helped. 3D printing was still meant for big firms and research laboratories, but the possibility of making cheaper printers was increasing. Still, the motivation for small and mid-size firms was still very low: why invest people and money on something so expensive and so difficult to use and with so many proprietary machines on the market? The word "proprietary" here is important: each company had its own technology and was not sharing it with others.

Up to this point, the leadership had come from small companies and independent inventors, not from the universities. The only exception had been the MIT, with its 3D inkjet technology. This changed in 2005 when Adrian Bowyer at the University of Bath launched an open-source project, RepRap, to develop a self-replicating 3D printer. Bowyer made history in 2008 when his Darwin machine built a copy of itself (RepRap self-replicated!), but, even more importantly, this 2008 machine, that was based on FDM, was the first low-cost 3D printer. If the Macintosh had created publishing for the desktop, RepRap created 3D printing for the desktop. Because everything in RepRap was open source, Bowyer had just started the open-source movement for 3D printing. In 2008 MakerBot in New York launched a website called Thingiverse where hobbyists could share for free their 3D models, and Philips' spinoff Shapeways (based in the Netherlands) launched an online market for 3D models. The users of Shapeways can design a product and upload 3D printable files, that Shapeways turns into products, and sells and ships them. In 2009 BitsFromBytes in Britain launched the RapMan, based on RepRap, a "Do-It-Yourself" kit for hobbyists to build their own 3D printer at home (acquired by 3D Systems in 2010). A few months later MakerBot launched an even more popular kit, CupCake, for $750. These were the first RepRap-based 3D printers to be commercially available, although they were just kits that you had to assemble at home. Ironically, the futuristic project to create a machine that can print parts to make other machines did not succeed but the project to spread the technology of 3D printing and make it more affordable started a revolution. RepRap "democratized" 3D printing. RepRap is still the most widely-used 3D printer by "makers" all over the world. In 2010 BitsFromBytes also introduced an assembled model, the BFB3000; and MakerBot eventually came out with a line of affordable consumer printers starting with the Replicator of 2012.

The RepRap project continued to be a major source of inspiration until our days. For example, in 2011 Solidoodle of New York introduced a consumer 3D printer for $700, another printer based on RepRap but pre-asssembled, and in 2012 a 21-years-old student from the Czech Republic, Josef Prusa, produced another self-cloning machine, the Prusa i3. Combined with crowdfunding websites like Kickstarter, RepRap led to a multiplication of 3D printers. For example, in 2011 Brook Drumm (based between San Francisco and Los Angeles) launched the Printrbot project for yet another kind of desktop 3D printer. In 2011 some Dutch scientists unhappy with RepRap started a competitor of RepRap called Ultimaker, which also became a startup offering a kit (and in 2012 a fully assembled printer).

The 2010s also saw some consolidation in the field of 3D printers: Stratasys acquired Bespoke in 2010, Solidscape in 2011, Objet in 2012 and MakerBot in 2013, as well as several providers of 3D models (The3dstudio, Freedom of Creation, MyRobotNation), while 3D Systems acquired BitsFromBytes in 2010 and Z Corp in 2012 (an acquisition that yielded 3D Systems' ProJet). This was useful: Darwinian selection reduced the number of players.

Innovation picked up speed dramatically. In 2012 Shapeways presented the concept of the "Factory of the Future", and started building it in New York, a factory with dozens of 3D printers that will be able to print five million products a year. In 2013 an MIT-offshoot, Formlabs, introduced a stereolithography 3D printer for the desktop. In 2013 Arburg in Germany introduced the Freeformer, a 3D printer that is compatible with traditional plastic manufacturing because it uses granules rather than (expensive) filaments or powders; and it can also mix different materials. In 2013 Nanoscribe, a spin-off of the Karlsruhe Institute of Technology in Germany, introduced a 3D printer of microstructures and nanostructures, The Photonic Professional GT. In 2013 WobbleWorks, founded in 2010 in San Jose by MIT Media Lab's alumnus Peter Dilworth and Maxwell Bogue, launched the 3Doodler, a 3D printing pen based on FDM that allowed users to create objects in mid-air. In 2014 New Matter, started in Los Angeles by Caltech scientist Steve Schell under the aegis of incubator Idealab (i.e. serial entrepreneur Bill Gross) and funded on IndieGoGo, introduced a 3D printer for the home and school market, the MOD-t (another FDM printer). In 2015 Carbon3D, a spinoff of the University of North Carolina that had relocated to Redwood City, unveiled a 3D printing process named Continuous Liquid Interface Production (CLIP), a kind of liquid-based stereolithography (SLA), that improved the printing speed. 3D printing is another case of technology that evolves and improves rapidly because of the genius of independent inventors, not because of the rich laboratories of the big corporations. The technology is so expensive that the "inventors" are not really garage kids like in Silicon Valley: they are often small and mid-size firms whose founders grew up in the world of the automated factory.

Lots of progress also in printing metal parts. In 2000 a former Ford scientist, Dawn White, invented a technique based on ultrasonic welding, a kind of welding that doesn't require melting the metal, and in 2011 a startup, Fabrisonic of Ohio, applied that technique to 3D printing for complex metal parts made of layers of different metals. In 2012 Beam (France) commercialized yet another laser-based metal printing technology (another LDT/LMD) called EasyCLAD that had been invented at IREPA. In 2015 XJet, founded by Objet's founder Hanan Gothait, launched a new inkjet metal nanotechnology. Toyota's LMD machine is scheduled for 2017.

In 2016 GE acquired Sweden's Arcam and Germany's Concept Laser (a maker of laser 3D-printers for metal parts). In 2017 Desktop Metal, founded by MIT scientists, unveiled a 3D-printer (powered by lasers and high heat) that can make objects out of hundreds of different alloys including steel, aluminum, titanium and copper. Alas, its price was $120,000. The founders include Emanuel Sachs, owner of the 3DP patent, and Yet-Ming Chiang and Ric Fulop, who were previously founders of another much hyped startup, A123, that was supposed to produce the next generation of lithium-ion batteries)

In 2015 Daniel Losinski's Skriware in Poland launched a crowdfunded project for a user-friendly, desktop, 3D printer featuring WiFi connection, touchscreen, and USB port. In 2015 Solido3D in Italy unveiled a $99 gadget called OLO, that turns any smartphone into a 3D printer using the light from the touchscreen to process the plastic. Voxel8, founded in 2015 in Boston by a pioneer of bioprinting, Jennifer Lewis, now at Harvard, received funding from the CIA to build a multi-material 3D printer. Their promotional video showed the printer building a drone using plastic and electrical circuits. In 2016 both Photocentric in Britain and Uniz 3D in San Diego introduced desktop resin-based 3D printers that used LCD (liquid crystal display) technology instead of laser or DLP. LCD technology should allow to 3D print objects faster and to build printers of any size, very small to very big. DLP is still preferred for accuracy (the pixels of DLP projectors are smaller) and for price (LCD uses LED lights which are more expensive and require cooling). I don't know which of these ideas will survive, but each one is potentially capable of expanding the market of 3D printers. We don't have the big success story yet. Google is the big success story of search engines. Facebook is the big success story of social media. Etc. No multi-billion dollar firm in 3D printing yet. In fact, many of the original firms have gone out of business. It is a tough business. And it is still a small market: $4 billion in 2014, a very tiny number compared to the $100 billion market for smartphones. The yearly statistics published by Wohlers are a little depressing: 150,000 3D printers sold, of which about 140,000 are desktop 3D printers, but these desktop models account for only about 15% of the total revenues. The real business is still in the industrial 3D printers that cost $100,000 and sometimes millions. But the market is small because nobody has invented the equivalent of the iPhone. When Apple announced the Apple Watch, i was very disappointed: what's special about a smartwatch? Apple wasted its talented engineers building something that any firm in the world can build. I was disappointed because i was expecting the Apple 3D Printer!

Sometimes the best ideas are not the ones that make the headlines. For example, the most creative idea that i have heard recently came from a team of students at University of Pennsylvania, BAM!3D, that designed a 3-D printer suspended from a balloon in order to get around the size limitations of 3D printers, so that the size of the 3D printer would stop being a limit con the size of the printed object.

Printing circuit boards is not trivial because they consist of hundreds of electronic components organized in multiple layers and connected by copper wires, but a new generation of 3D printers allows hobbyists and schools to quickly and cheaply build electronic circuits for home and classroom experiments. In 2010 Neotech of Germany wed Optomec's AerosolJet technology with to its own 5-axis machine to build a 3D printer for 3D-printed electronics. In 2014 Cartesian in Australia crowdfunded a project for a desktop inkjet-based machine, EX (later renamed Argentum), that can print circuit boards, and in 2015 Voltera in Canada crowdfunded a similar machine, called V-One. In 2016 Nano Dimension of Israel introduced the inkjet-based DragonFly 2020 that can 3D-print multi-layered circuit boards, a product designed by Lena Kotlar. Now we can combine these 3D printers with progress in the circuit boards. For example, in 2013 Gregory Abowd at Georgia Tech, in collaboration with Yoshihiro Kawahara of the University of Tokyo and Steve Hodges of Microsoft in Britain, demonstrated a technique to print electronic circuits of any shape with a regular inkjet printer, and in 2014 Anming Hu at the University of Tennessee used an inkjet printer to print electronic sensors and an "electronic skin" (that could be used to cover a robot so the robot can actually "sense" the environment).

The vast majority of 3D printers still use FDM: the Rostock Delta Printer, invented in 2012 in Seattle by Johann Rocholl and commercialized by SeeMeCNC, WASP's DeltaWASP (Italy, 2013), DreamMaker's Overlord , Skriware, New Matter, and all the printers based on RepRap.

DLP printers also became popular thanks to models introduced by B9 Creator of South Dakora in 2012 and mUve of Michigan in 2013. Autodesk's Ember, Uncia3D (China), Morpheus (South Korea) and Kudo3D (San Francisco) are DLP printers.

Formlabs' Form 1 and 2 (Boston), XYZ's Nobel (Taiwan), Carbon3D's M1 (Redwood City), and of course 3D Systems' ProJet, are SLA printers.

SLS requires the use of high-powered lasers, and therefore it has been traditionally more expensive, but recently prices have gone down thanks to Sintratec (Switzerland), Norge (Britain) and Prodways (France).

3D printing is still far from offering the same quality of conventional manufacturing technology, the so-called "injection molding" technology, but in 2016 two products were announced that would compete with injection molding: Carbon3D's M1 (that one can rent for $40,000) and Hewlett Packard's Multi Jet Fusion (that one can buy for $130,000). HP's Multi Jet Fusion uses the High Speed Sintering (HSS) technology invented by Loughborough University in Britain that should be much faster than today's printers and it is full color. Color is still an issue in 3D printing. Most of the 3D-printed objects are monochromatic. When you see a mono-chromatic object, you think "it must be made with a 3D printer". That has to change. In 2014 DreamMaker of Shanghai in China started selling the Overlord, based on RepRap, while 3D Systems introduced the CubePro Trio, based on the BFB3000 that they acquired from BitsFromBytes. In 2015 Samsung filed a patent for a multicolor technology.

Then there are many ideas on how to make 3D printing more practical for building complex objects. For example, 4D Printing is a joint project between the MIT, Stratasys and Autodesk for printing materials that are both customizable and programmable. The Self-Assembly Lab of the MIT, founded in 2013 by Skylar Tibbits, defines 4D printed object as a 3D printed object that performs an additional function, for example that learns and adapts to its environment, or, another example, that self-assembles after the parts come out of the printer. The idea is to "program" a material (plastic, textiles, wood, rubber, carbon fiber) to not only print the parts but also to make sure that those parts will assemble into the desired final object; and then to change shape in order to better adapt to the environment, for example to an increase in temperature, pressure or moisture. Tibbits uses the software of Project Cyborg, originally developed in 2013 by Carlos Olguin's team at Autodesk for biotech applications.

The concept of the factory is changing. First of all, there are lower technological obstacles to manufacturing, so the companies with giant super-expensive factories that today have a huge advantage tomorrow will not have an advantage, will have a disadvantage.

Secondly, you need less money to make a product. Thirdly, you can find that money through a different channel, namely crowdfunfing: Kickstarter, IndieGogo, GoFundMe, etc. The big manufacturing companies used to have a huge advantage: how can an independent inventor compete with Toyota or General Motors if developing a new car model costs one billion dollars? 3D printing will reduce the cost, and crowdfunding will provide money to independent inventors who have no money.

Then it was also expensive to handle the logistics: storing the products in warehouses and shipping them. Amazon has opened a colossal "fulfilment" center in Arizona called PHX6 that is fully automated with robots and any merchant can "rent" the facility to store and ship its products. Amazon is rapidly moving its online business towards same-day delivery (perhaps with drones) and mobile shopping. Drones and robots are reinventing logistics. This infrastruture will soon be available to any merchant that subscribes to Amazon's program.

Marketing used to be another obstacle for family businesses: how can a family-run business compete with big companies that can advertise in big magazines, television and radio? We are in a new of marketing, in which readers trust the reviews published by users on Amazon, Yelp, and many specialized websites, and the same readers don't really pay attention to the ads on magazines and television. The most powerful form of marketing is the buyer's opinion, not the seller's opinion. Consumers are increasingly steered by the algorithms that websites like Amazon use to recommend purchases based on their past history of shopping. This process is outside the control of the big traditional manufacturers and retailers, no matter how much money they want to invest.

The good news for a manufacturer is that they can get immediate feedback from the users about a new product, and then it can modify that product to increase user satisfaction.

The "time to market" used to be measured in years. Now the loop from prototyping to crowdfunding to manufacturing to retail to user feedback and back to prototyping is closing very rapidly. The era of "user-friendly" is transforming into the era of "object-friendly".

Everything is changing: the actors, the investors, the process, the shop.

The next step will be to connect the 3D-printing manufacturers in a network so we can reinvent the supply chain for the age of home-based 3D printing. If someone can 3D-print a part that someone else needs to assemble a product, the match should be made more or less automatically: each maker can publish the 3D model, which is a digital file, and let other makers (or, better, their software) decide if that's what they need. The system could easily calculate the cost of ordering a 3D-printed part from Germany or China, and make the best decision, place the order for the parts that it needs and pay automatically. The company that assembles the product doesn't even need to know where the parts are coming from.

Industry 4.0 is not a real revolution without 3D printing. Industrial robots have existed for a long time, and putting more robots in the factory or warehouse doesn't constitute a real revolution: it's a simplification of an existing process. Data analytics keep changing name but it has existed since the first mainframe computer entered a data center. Smart machines that communicate with each other have have existed since the 1980s (it used to be calle "distributed numerical control" or DNC). But 3D printing turns objects into digital files, and that is a real revolution.

JP Rangaswami says that the firm of the future has to "design for loss of control". The firm has to turn an organization into a social enterprise: a network instead of a hierarchy, open to the outside instead of closed, and ready to contribute to the open-source economy, not only to its own proprietary technologies.

Of course, 3D printers are still too expensive for the average "maker". But i can see a movement towards a new way of building sophisticated things, using low-budget materials with high-tech tools. Ron Rael at UC Berkeley is an architect who does precisely this: he 3D-prints amazing structures using materials such as sand, clay, rubber tires, ...

In 2016 Steve Keating at the MIT built a 3D-printing system called the Digital Construction Platform to build houses in novel ways: not only a new way of making, but (as Keating said) "an entirely new way of thinking about making" (Keating was hired by Apple in 2017).

When the Islamic State of Iraq and Syria (ISIS) was about to invade the ancient city of Palmyra, the Institute of Digital Archaeology in Britain was asking volunteers to take pictures with 3D cameras of the monuments. The Islamists destroyed the monuments, but the institute has been able to rebuild them, and in 2016 a 3D-printed replica of the Triumphal Arch of the Temple of Bel was placed in Trafalgar Square of London. I wish someone had done the same for the giant sixth-century Buddha statues in Bamiyan (Afghanistan), destroyed by the Taliban in 2001,

Frank Arcella's AeroMet specialized in 3D-printing titanium parts of airplanes using its LAM technology on titanium powder, but it went out of business in 2005. Arcam in Sweden, MCP in England and EOS in Germany (the M270 system) are also making titanium parts for airplanes.

In 2017 Yinmin "Morris" Wang at Lawrence Livermore National Laboratory 3D-printed extremely strong stainless steel.

Unfortunately, in 2013 a Texas-based company, Defense Distributed, used a 3D-printer to make a pistol, and in 2018 it made the file available on the Internet so that anyone can 3D-print the weapon at home, and this is perfectly legal in the USA.

The new generation of the late 2010s included: Los Angeles-based Divergent 3D (that specialized in 3d-printing vehicles), Minnesota's Evolve (a Stratesys spinoff led by Steve Chillscyzn), Boston's Markforged (for composite parts), Chicago's Impossible Objects (for composite parts), Silicon Valley's Velo3D (for high-quality metal parts), Israel's XJet (for ceramic and metal), Australia's Aurora Labs (metal), China's Snapmaker (a compact machine that was also a laser engraver and a CNC carver), Los Angeles' SprintRay (dentistry), San Diego-based Uniz Technology (medical), Germany's Apium (high-temperature polymers), Silicon Valley's Arevo Labs (high-performance and ultra-strong polymers), San Diego-based Robo 3D (a versatile home printer with WiFi), Spain's BCN3D (another versatile home printer), Taiwan's T3D (powered by the light of a smartphone or tablet), etc.

Velo3D, founded in 2014 in Campbell by Benny Buller and Erel Milshtein, developed the Sapphire system (first demonstrated in 2018), capable of 3D-printing complex metal objects which, unlike Desktop Metal's and HP's 3D-printers, were ready for mass production, not just prototypes.

Arevo Labs, founded in 2013 in Santa Clara by Hemant Bheda and others, specialized in 3d-printing high-performance and ultra-strong polymers, a technology demonstrated in 2018 when they 3D-printed a carbon-fiber bicycle.

Prellis Biologics, founded in 2016 in San Francisco by Noelle Mullin and Melanie Matheu, worked on 3D-printing human organs. Theirs was a long-term project, expected to show results within five or six years. And the 3D printer would not be like the plastic-making ones: it was expected that it would take between two and four months to print an organ. Nonetheless, the excitement was significant, as organ transplants was still a rather dangerous operation and the USA was facing a shortage of organs for transplants.

Stuart Williams, founder in 1997 of the Biomedical Engineering program at the University of Arizona, used tools from nScrypt to build the first version of what he called the BioAssembly Tool (BAT), and in 2001 Williams 3D-printed a living tissue. In 2001 EnvisionTec started selling its 3D-Bioplotter, based on a technology invented at the Freiburg Materials Research Centre in Germany. In 2004 Thomas Boland at Clemson University in South Carolina used inkjet technology to manufacture heart tissue. When in 2006 the USA approved Boland's patent for "Ink-jet Printing of Viable Cells", that was the official birth date of 3D bioprinting. In 2006 3D Systems announced its InVision DP (dental professional), a combination of 3D scanner and 3D printer for the dental market, followed in 2007 by Solidscape's D66 also for dental applications. In 2007 an Arcam printer was used to make a hip. In 2008 Stratasys introduced a material that is "biocompatible", i.e. that the human body would not reject. In 2008 Organovo, founded by Gabor Forgacs of the University of Missouri, 3D-printed a blood vessel using a machine by nScrypt of Florida, In 2010 Bespoke Innovations, founded in San Francisco by industrial designer Scott Summit and orthopedic surgeon Kenneth Trauner started 3D-printing prosthetic limbs that are both customized and elegant. and Joel Sadler's team at Stanford University 3D-printed the "JaipurKnee" that was a big news in India. In 2010 Organovo of San Diego introduced a bioprinter that can print human tissue, the MMX (almost ten years after EnvisionTec) and started a collaboration with Autodesk to create 3D-design software for Organovo's NovoGen MMX bioprinter. In 2011 Susmita Bose at Washington State University 3D-printed a material that works like bones. In 2012 Jules Poukens at the Biomedical Research Institute at Hasselt University in Belgium 3D-printed a jaw that was then implanted to an 83-year-old woman. In 2013 a 17-year-old high-school student from Colorado, Easton LaChappelle, 3D-printed a robotic arm (not a real arm) and Ivan Owen at Washington State 3D-printed a robotic hand. Two ears were 3D-printed in 2013, one by Michael McAlpine's team at Princeton University and one by Jason Spector's team at Cornell University. In 2014 a Kentucky-based startup, Advanced Solutions, launched a robot called BioAssemblyBot (BAB) and CAD software for 3D-bioprinting called Tissue Structure Information Modeling (TSIM) In 2015 Matt Ratto's team at the University of Toronto 3D-printed a critical component of the artificial leg for a young Ugandan woman. In 2015 Organovo 3D-printed kidney tissue (but not an entire kidney, like some journalists reported). In 2015 nScrypt of Florida introduced the TE (Tissue Engineering) bioprinter. In 2017 Jose Luis Jorcano at Universidad Carlos III in Spain built a 3D printer that can print human skin and the printer is being marketed by the firm BioDan Group. In 2015 the MIT student Steven Keating 3D-printed his own brain tumor. There are now 3D bioprinters in many countries: regenHu (which stands for Regeneration HUman) of Switzerland, Bioprinting Solutions of Russia, Cyfuse Biomedical of Japan, Regenovo of China, Bio 3D of Singapore, etc. And we also have the first "low-cost" bioprinters, like 3Dynamic of Britain and Biobots of Philadelphia.

The most famous scientist in this field is Anthony Atala, director of the Wake Forest Institute for Regenerative Medicine. It is not completely true that in 1999 he implanted a 3D-printed organ in a human patient (they 3D-printed only some supporting parts of the bladder), and it is not true that in 2011 his team 3D-printed a kidney (it was a miniature kidney), but every year he gets closer to 3D-printing an artificial organ. A scientist that has been working for years on 3D-printing living human tissues is Jennifer Lewis, first at the University of Illinois and then (since 2013) at Harvard University, where she developed special inks that our body does not reject. James Yoo at Wake Forest is working on 3D printed skin layers that could be sprayed on skin burns.

We are still in the prehistory of this discipline. None of the 3D-printed limbs and organs is really a replica of the human body part. More importantly, we are both humans but my limbs and organs are different from yours, and from the limbs and organs of the other seven billion humans. These 3D-printed limbs and organs must be customizable in order to represent real progress compared with today's artificial limbs and organs.

The most exciting experiment for me was done in 2014 by William Root, a student of industrial design in New York. He used a 3D-scanning technology developed at the MIT's Biomechatronics Lab (FitSocket technology) to capture the details of a patient's leg and to design the best possible fit of a prosthetic leg. Then it 3D-printed a prosthetic leg that he calls Exo-Prosthetic, a titanium exoskeleton that should replicate exactly the amputated limb.

Most product designers start with an existing product and then modify it for some new function or a new look. To do the same in the age of 3D printing, we need to have a 3D scanner. We used to have printers that were also scanners: you can scan a page, you modify it on the computer, and then you can print a page. Now we have 3D printers, but they are not 3D scanners. The 3D scanner is a separate machine, and it is not cheap. The problem will be solved if the Peachy Printer succeeds: it is a Kickstarter project that aims to build a $100 3D printer and scanner. Ali Hajimiri's team at Caltech is developing a one-millimiter chip that uses a laser to scan an object. Such a chip could fit inside a smartphone, turning your smartphone into a 3D scanner. In 2015 Ramesh Raskar'steam at the MIT Media Lab invented a technology called "Polarized 3D" that vastly improves the quality of 3D scanning and someday could give us high-quality 3D cameras into smartphones.

So in the future we may be able to take a picture of an object, modify it with some 3D-design software, and then 3D-print a replica. When 3D printers become consumer devices, you will be 3D-printing your own furniture. Scan your tea cup, modify the 3D model, print your own version of the tea cup. Even better: i envision a future in which we will have miniature 3D replicas of our relatives and friends instead of photographs. In 2015 Los Angeles-based CoKreeate started a service to make replicas of people. They use an Artec scanner to build a 3D model of you and then a Z Corp printer to make a miniature replica of you.

The final objective should be "bidirectional fabrication": pick an object from a database of 3D models, 3D-print its digital file, modify it manually, 3D-scan it, upload its new digital file, 3D-print the new digital file. In 2015 Jason Alexander's team at Lancaster University in Britain introduced such a system, called ReForm, for clay objects (the advantage of using clay is that it remains soft until you put it in an oven, so you can reshape it as many times as you like). The advantage of using ReForm is that you don't need CAD software.

3D scanning is not a new technology. If you go back in time, LIDAR (LIght Detection And Ranging) employs a laser to provide three-dimensional information about a surface. It exploits the principle of the radar, except that it uses laser pulses which allow for more precision. In 1971 it was used to map the surface of the Moon. LIDAR is how Google's driverless car "sees". There were 3D scanners in the 1960s, but the modern technology uses lasers, and this new technique was invented in 1977 by Michel Clerget, Francois Germain and Jiri Kryze working at the national French laboratory IRIA (later renamed INRIA). The first commercial product was a head scanner introduced in 1987 by David and Lloyd Addleman's Cyberware Laboratories of Monterey (south of the Bay Area), followed in the 1990s by 3D scanners by NKK in Japan (Voxelan), Vision 3D in France, 3D Scanners in Britain (Replica of 1994 and ModelMaker of 1996), Digibotics of Texas, Ben Kacyra' Cyra Technologies of Los Angeles (the Cyrax of 1998, acquired in 2001 by Leica of Switzerland), and Cyberware itself. Kinect, the motion-sensing device that was introduced by Microsoft for the Xbox-360 videogame console in 2010 but actually invented in 2005 by Alexander Shpunt's PrimeSense, is a by-product of 3D scanning technology.

Today we have handheld 3D scanners like 4DDynamics' IIIDScan PrimeSense from Belgium ($1,500) Shining 3D's EinScan-S from China ($1,300), another crowdfunded project, and Fuel3D's Scanify from Britain ($1,500), based on the Eykona technology developed at Oxford Univ in 2005, probably the fastest of this group; but the handheld 3D scanners that really work well are the expensive ones, like Artec's Eva from Luxembourg ($15,000), Nikon's ModelMaker MMDx, Creaform's HandyScan from Canada, etc. We have 3D scanners that attach to a smartphone or a tablet like 3D Systems' Cubify iSense ($400), EORA ($330), another crowdfunded project, and Occipital's Structure Sensor from Colorado ($400), another crowdfunding project. And we have desktop scanners like Matter & Form's MFS1V1 from Toronto, another crowdfunded project ($600), that uses a turntable to scan the object, NextEngine's Ultra HD from Los Angeles ($3,000), and MakerBot's 2016 Digitizer ($800). These 3D scanners create a digital file that can then be manipulated on a computer.

In 2014 two products changed the history of 3D printing, AIO Robotics' Zeus from Los Angeles ($2,500) and XYZPrinting's Da Vinci AiO from Taiwan ($800): they finally combine 3D scanning and 3D printing.

In parallel, there has been progress in 3D design/modeling. Some of the most influential tools were: CATIA from Dassault in France (1981), Pro/E (later renamed Creo) from Parametric Technology Corporation in Boston (1987), 3D Studio from Gary Yost in San Francisco (1990, acquired in 1992 by Autodesk and renamed 3ds Max), LightWave3D from NewTek in Texas (1990), SolidWorks from Jon Hirschtick in Boston (1995, acquired in 1997 by Dassault), SolidEdge from Intergraph in Alabama (1996, acquired in 2007 by Siemens), ZBrush from Pixologic in Los Angeles (1999), Inventor from Autodesk (1999), SketchUp from Last Software in Colorado (2000, acquired by Google in 2006). The problem with all of them is that they were terribly complicated to use. The features got more and more impressive, but the price and the complexity also became more impressive. However, over the last 20 years many free or cheap tools have appeared with increasingly more sophisticaed functions, like the open-source projects Blender (Ton Roosendaal in the Netherlands, 1995) and FreeCAD (Jurgen Riegel in Germany, 2002), Autodesk 123D (2009), Pixologic's Sculptris (2009), and Kai Backman's TinkerCAD (2011, acquired by Autodesk in 2013).

Recently a big influence has been the videogame Minecraft, introduced in 2011 by the Swedish designer Markus Persson. Even children can create complex 3D models using Minecraft's simple building-block concept. For example, in 2014 Sylvain Huet in France used the same concept to develop 3D Slash. And of course there are many attempts to bring 3D modeling to mobile computers and to use the cloud and to create communities of designers. For example, two former Autodesk employees, Evi Meyer and Erik Sapir, founded uMake in 2014 in San Francisco to offer a smartphone alternative to Autodesk's 3D-design tools; in 2015 Autodesk added a cloud-based service (Forge) for makers to design 3D objects, coupled with a venture fund (Spark) to invest in the boldest ideas; and in 2015 Rita Wong in San Francisco launched Valsfer, a social-networking platform to connect designers and manufacturers.

A 3D model is typically encoded in a format called STL. Then the designer can use STL editors like the Italian open-source MeshLab (developed at the University of Pisa in 2005) or the Canadian tool Meshmixer (developed in 2009 at the University of Toronto by Ryan Schmidt and acquired by Autodesk). When the STL file is ready to be sent to a 3D printer, we need tools to convert the STL into printing instructions for the 3D printer. This means "slicing" the 3D model, so sometimes this software is called a "slicer". Examples are: Repetier, the standard 3D printing software within the RepRap community, the open-source Slic3r, Ultimaker's Cura, and Simplify3D in Ohio.

Complicated, right? But the trend is to simplify this process, so in a few years the industry will be unrecognizable.

Microsoft, for example, has a comprehensive strategy. In 2013 Microsoft introduced 3D Builder and 3D Scan for the Windows 10 operating system. Microsoft users can capture, download, edit and print 3D models. If you don't have a 3D printer, don't worry: Microsoft partners with Materialise to offer a cloud-based 3D printing services, and your 3D-printed object gets shipped to your door. You can also 3D-print from HoloLens. Microsoft has adopted the 3D Manufacturing Format (3MF) standard that was released in 2015 by a consortium that includes 3D Systems, Stratasys, Autodesk, HP, Mcor, Materialise, Shapeways and Siemens.

Some day 3D printing will become a family hobby. Children of my generation played with Lego building blocks. Children of future generations will modify objects, scan them, and order a 3D print. Children of my generation were asked to draw pictures, and the best ones were displayed on the walls of the classroom. Children of future generations will be asked to create objects, and the best ones will be displayed in the hallway of the school. Parents will 3D print objects for useful purposes in the house or simply for fun. You will be able to 3D-print your own jewelry, makeup and clothes. How many times have you wished that your favorite chair were a little higher or a little sturdier, or that the tv stand were a little lower? You will be able to scan, modify and 3D print any object that you have. Humankind will return to the artisan society of centuries ago, but with an important difference: i can email you the digital file of my object and you can upload it and 3D-print it, and this whole operation takes just a few seconds. I have done more than build a copy of my vase for you. I have given you the file that you can use to make an unlimited number of copies.