menu_open Columnists
We use cookies to provide some features and experiences in QOSHE

More information  .  Close

20 inventions and decisions that had to happen before you could buy anything online

3 0
yesterday

20 inventions and decisions that had to happen before you could buy anything online

Every time you add something to a cart and check out, you're using infrastructure that took a century to build — and most of it has nothing to do with the internet

Kampus Production / Pexels

When Amazon $AMZN 2.90% launched in 1995, it appeared to be a new kind of thing — a bookstore that existed only online, without physical inventory in a storefront, without a cash register, without a salesperson. It was new in those specific respects. In almost every other respect, it was the beneficiary of a century of infrastructure that had been built by other industries for other purposes, and without which no online transaction would have been possible.

The parcel delivery network that brought the book to the door had been built by the postal service and by private couriers over a hundred years. The credit card that paid for it had been invented in 1950 and had spent four decades building the trust, the fraud detection systems, and the merchant acceptance network that made a transaction between strangers feel safe. The relational database that stored the inventory and the customer record had been theorized in 1970 and implemented through the 1970s and 1980s. The encryption standard that protected the payment had been developed by mathematicians working on entirely different problems. The container shipping system that moved the books from warehouse to distribution center had been built in the 1960s to serve the needs of the manufacturing economy.

E-commerce is not a technology. It is the convergence of dozens of technologies, legal frameworks, logistics systems, financial instruments, and consumer behaviors, each of which had to reach a sufficient level of development before the whole system could function. Pull out any single element — the parcel network, the payment rail, the encryption, the consumer trust, the search engine, the product database — and the system fails. The genius of the companies that built e-commerce was partly in the technology they developed and partly in the recognition that the rest of the infrastructure had finally reached the point where the assembly was possible.

This list covers 20 of the essential preconditions — the inventions, the decisions, the standards, and the systemic developments that had to exist before a stranger could sell a book to another stranger across the internet, payment included, delivery guaranteed. Several of them are obvious. Several are not obvious at all. Together they tell the story of how the apparently sudden arrival of online commerce was actually the slow convergence of a century of prerequisite work.

Startup Stock Photos / Pexels

The internet — the global network of interconnected computers that transmits data using the TCP/IP protocol suite — is the obvious prerequisite, but its history as a precondition for e-commerce is more specific and more interesting than its general reputation as a military communications project suggests. The internet's evolution from ARPANET (the U.S. Defense Department's packet-switching network, operational from 1969) to the commercial internet of the early 1990s required specific decisions that were not inevitable and that took decades to make.

The critical decision was the opening of the internet to commercial traffic. ARPANET and its successors operated under an Acceptable Use Policy that prohibited commercial activity — the network was funded by the federal government for research and educational purposes, and using it to sell things was explicitly prohibited. The National Science Foundation Network (NSFNET), which served as the internet's backbone through the late 1980s and early 1990s, maintained this prohibition. Commercial use was permitted only on private networks connected to but not traversing the NSFNET backbone.

The transition began in 1991 when the NSF began allowing commercial entities to connect to the internet through commercial internet service providers, and was effectively completed with the privatization of the NSFNET backbone in 1995 — the same year that Amazon $AMZN 2.90% launched. The decision to privatize and commercialize the internet backbone, made by a relatively small number of policy officials and network engineers, was the specific gate that commercial internet activity had to pass through, and it passed through only months before the first significant e-commerce companies appeared.

Lisa from Pexels / Pexels

The internet existed for two decades before e-commerce was possible because the internet without the Web was a command-line environment accessible only to researchers, academics, and technically sophisticated users. Tim Berners-Lee's invention of the World Wide Web — the system of hypertext documents linked by URLs and rendered by web browsers, first proposed in 1989 and publicly released in 1991 — was the specific layer of internet architecture that made the internet accessible to ordinary people without technical training.

Berners-Lee invented the Web while working at CERN, the European physics laboratory, as a solution to the specific problem of sharing research documents across different computer systems used by researchers at different institutions. The Web's three core components — HTML (the document format), HTTP (the transfer protocol), and URLs (the addressing system) — were all designed for document sharing rather than for commercial transactions. The shopping cart, the checkout process, the product page — none of these were in Berners-Lee's original design, and the Web had to be extended with dynamic content technologies (CGI scripts, server-side processing) before these commercial functions were possible.

Crucially, Berners-Lee and CERN chose not to patent the Web, making it available as an open standard that anyone could implement without licensing fees. This decision — made without the commercial context that might have produced a different outcome — was the specific reason that the Web could spread as rapidly as it did and that no single company could control the infrastructure on which e-commerce was subsequently built.

Bastian Riccardi / Pexels

The World Wide Web's potential as a medium accessible to ordinary people was unlocked specifically by the graphical web browser — software that rendered HTML documents as visual pages with images, formatting, and clickable links rather than as raw text. The first widely used graphical browser, Mosaic, was released by Marc Andreessen and Eric Bina at the National Center for Supercomputing Applications (NCSA) in 1993. Netscape Navigator, developed by Andreessen after he left NCSA, followed in 1994 and became the dominant browser of the early commercial web.

Before the graphical browser, using the Web required comfort with command-line interfaces and the discipline to navigate by typing URLs rather than clicking links. Mosaic's innovation was the inline image — displaying images within the text of a web page rather than as separate files that had to be opened independently — and the intuitive point-and-click navigation that made web pages legible to anyone who could use a mouse.

The browser also introduced the concept of the web page as a visual space that could be designed — that had layout, typography, color, and structure — rather than merely a document that had content. This transition from document to designed page was the specific precondition for the product page, the shopping cart, and the checkout flow that e-commerce required. A commerce experience built on command-line text would have found no mass-market audience.

Netscape's decision to add SSL (Secure Sockets Layer) encryption to Navigator in 1994 was the specific browser decision most directly relevant to e-commerce: it provided the secure transmission of payment information that made online transactions viable.

Markus Spiske / Pexels

SSL — Secure Sockets Layer, the encryption protocol that protects data transmitted between a web browser and a web server — was developed by Netscape and first implemented in Netscape Navigator in 1994, with a publicly released version in 1995. Its successor, TLS (Transport Layer Security), is the protocol that the padlock symbol in modern browsers represents. Without encryption, any payment information entered into a web page — credit card numbers, billing addresses, bank account details — would be transmitted as readable text across the internet, accessible to anyone with the ability to intercept network traffic.

The specific cryptographic foundations of SSL were not invented for commerce. Public-key cryptography — the mathematical framework underlying asymmetric encryption, in which a public key encrypts data that only the corresponding private key can decrypt — was developed by Whitfield Diffie and Martin Hellman in 1976 and independently by Ron Rivest, Adi Shamir, and Leonard Adleman (RSA) in 1977, as contributions to pure cryptography research. The RSA algorithm, in particular, was developed as a theoretical demonstration of public-key cryptography's properties rather than as a commercial product.

The specific........

© Quartz