things from anywhere to anywhere and from any computer to any computer-seamlessly. The wall-Windows- Netscape phases paved the way for that by standardizing the ways words, music, pictures, and data would be digitized and transported on the Internet-so e-mail and browsing became a very rich experience.
But for all of us to go to the next stage, to get more out of the Internet, the flattening process had to go another notch. We needed two things. We needed programmers to come along and write new applications– new software-that would enable us really to get the maximum from our computers as we worked with these digitized data, words, music, and pictures and shaped them into products. We also needed more magic pipes, more transmissions protocols, that would ensure that everyone's software applications could connect with everyone else's software applications. In short, we had to go from an Internet that just connected people to people, and people to their own applications, to an Internet that could connect any of my software programs to any of your software programs. Only then could we really work together.
Think of it this way: In the beginning, work flow consisted of your sales department taking an order on paper, walking it over to your shipping department, which shipped the product, and then someone from shipping walking over to billing with a piece of paper and instructing them to churn out an invoice to the customer. As a result of the Berlin Wall-Windows-Netscape phases, work flow took a huge leap forward. Now your sales department could electronically take that order, e-mail it to the shipping department within your own company, and then have the shipping department send out the product to the customer and automatically spit out a bill at the same time. The fact that all the departments within your company were seamlessly interoperable and that work could flow between them was a great boost to productivity-but this could happen only if all your company's departments were using the same software and hardware systems. More often than not, back in the 1980s and early 1990s, a company's sales department was running Microsoft and the inventory department was running Novell, and they could not communicate with each other. So work did not flow as easily as it should.
We often forget that the software industry started out like a bad fire department. Imagine a city where every neighborhood had a different interface for connecting the fire hose to the hydrant. Everything was fine as long as your neighborhood fire department could handle your fire. But when a fire became too big, and the fire engines from the next neighborhood had to be called in, they were useless because they could not connect their hoses to your hydrants.
For the world to get flat, all your internal departments-sales, marketing, manufacturing, billing, and inventory- had to become interoperable, no matter what machines or software each of them was running. And for the world to get really flat, all your systems had to be interoperable with all the systems of any other company. That is, your sales department had to be connected to your supplier's inventory department and your supplier's inventory department had to be seamlessly connected to its supplier's supplier, which was a factory in China. That way, when you made a sale, an item was automatically shipped from your supplier's warehouse, and another item was automatically manufactured by your supplier's supplier, and a bill was generated from your billing department. The disparate computer systems and software applications of three distinctly different companies had to be seamlessly interoperable so that work could flow between them.
In the late 1990s, the software industry began to respond to what its consumers wanted. Technology companies, through much backroom wrangling and trial and error, started to forge more common Web-based standards, more integrated digital plumbing and protocols, so that anyone could fit his hose-his software applications-onto anyone else's hydrant.
This was a quiet revolution. Technically, what made it possible was the development of a new data description language, called XML, and its related transport protocol, called SOAP. IBM, Microsoft, and a host of other companies contributed to the development of both XML and SOAP, and both were subsequently ratified and popularized as the Internet standards. XML and SOAP created the technical foundation for software program-to-software program interaction, which was the foundation for Web-enabled work flow. They enabled digitized data, words, music, and photos to be exchanged between diverse software programs so that they could be shaped, designed, manipulated, edited, reedited, stored, published, and transported-without any regard to where people are physically sitting or what computing devices they are connecting through.
Once this technical foundation was in place, more and more people started writing work flow software programs for more and more different tasks. Wild Brain wanted programs to make animated films with a production team spread out around the world. Boeing wanted them so that its airplane factories in America could constantly resupply different airline customers with parts, through its computer ordering systems, no matter what country those orders came from. Doctors wanted them so that an X-ray taken in Bangor could be read in a hospital in Bangalore, without the doctor in Maine ever having to think about what computers that Indian hospital had. And Mom and Dad wanted them because they wanted their e-banking software, e-brokerage software, office e-mail, and spreadsheet software to all work off their home laptop and be able to interface with their office desktop. And once everyone's applications started to connect to everyone else's applications-which took several years and lot of technology and brainpower to make happen-work could not only flow like never before, but it could be chopped up and disaggregated like never before and sent to the four corners of the world. This meant that work could flow anywhere. Indeed, it was the ability to enable applications to speak to applications, not just people to speak to people, that would soon make outsourcing possible. Thanks to different kinds of Web services-work flow, said Craig Mundie, Microsoft's chief technology officer, “the industry created a global platform for a global workforce of people and computers.”
The vast network of underground plumbing that made it possible for all this work to flow has become quite extensive. It includes all the Internet protocols of the previous era, like TCP/IP and others, which made browsing and e-mail and Web sites possible. It includes newer tools, like XML and SOAP, which enabled Web applications to communicate with each other more seamlessly, and it includes software agents known as middleware, which serves as an intermediary between wildly diverse applications. The nexus of these technologies has been a huge boon to innovation and a huge reducer of friction between companies and applications. Instead of everyone trying to control the fire hydrant nozzle, they made all the nozzles and hoses the same, creating a much bigger market that stretched across every neig
hborhood of the world. Then companies started to compete instead over the quality of the hose, the pump, and the fire truck. That is, they competed over who could make the most useful and nifty applications. Said Joel Cawley, the head of IBM's strategic planning unit, “Standards don't eliminate innovation, they just allow you to focus it. They allow you to focus on where the real value lies, which is usually everything you can add above and around the standard.”
I found this out writing my last book. Once Microsoft Word got established as the global standard, work could flow between people on different continents much more easily, because we were all writing off the same screen with the same basic toolbar. When I was working on my first book, From Beirut to Jerusalem, in 1988,1 spent part of my year's leave in the Middle East and had to take notes with pen and paper, as it was the pre-laptop and pre- Microsoft Word era. When I wrote my second book, The Lexus and the Olive Tree, in 1998, I had to do some of the last-minute editing from the computer behind the front desk at a Swiss hotel in Davos on a German version of Microsoft Word. I could not understand a single word, a single command function, on the toolbar of the German version of Word. But by 1998, I was so familiar with the Word for Windows writing program, and where the various on-screen icons were, that I was able to point and click my way through the editing on the German version and type my corrections with the English letters on the German keyboard. Shared standards are a huge flattener, because they both force and empower more people to communicate and innovate over much wider platforms.
Another of my favorite examples of this is PayPal, which enabled eBay's e-commerce bazaar to become what it
