In mid-twentieth-century computing, early Turing-complete digital machines like ENIAC and Whirlwind relied on individual glass vacuum tubes that were each about the size of an incandescent light bulb. These massive systems consumed as much electrical power as an entire city block while struggling to maintain an operational uptime of more than 50 percent. Inherited from telephony, the delicate glass switches burned out frequently, confining electronic digital computing to heavily funded government research facilities and specialized military installations.
The Shrinking Switch: From Glass Tubes to Integrated Wafers
Telephony soon provided the catalyst for the first major technological shift away from vacuum tubes. In December 1947, engineers at Bell Labs invented the transistor with the primary goal of improving telephone switching circuits. A transistor performed the same logical switching function as a vacuum tube but occupied a minute fraction of the physical space, consumed vastly less power, and operated with far higher reliability. By 1960, the computer industry had permanently embraced discrete transistors for corporate computing. However, manually wiring together tens of thousands of individual transistors remained a daunting task, and high-powered mainframes still filled large rooms. A further breakthrough arrived in 1958, when Jack Kilby of Texas Instruments conceived the integrated circuit, embedding miniaturized transistors directly onto silicon wafers for automated mass production.
The Accidental Revolution of the Single-Chip CPU
In 1965, engineer Gordon Moore formulated his observation on chip fabrication scaling, calculating that the number of transistors on a silicon wafer of a given size would double every year, a forecast he revised in 1975 to a doubling every two years. Seeking to commercialize integrated circuit technology, Moore and Robert Noyce founded Intel in July 1968.
The decisive architectural breakthrough took place in 1971. Intel engineer Federico Faggin consolidated all of the circuits making up a computer’s arithmetic, logic, and control units onto a single chip, creating the Intel 4004 microprocessor. Faggin and his colleagues originally viewed the 4004 not as the central processor of general-purpose home computers, but as a modest component to drive mainframes' peripheral disk controllers. Nevertheless, integrating an entire central processing unit onto one silicon wafer lowered manufacturing costs and drastically reduced machine size.
Desktop Microcomputers and Home Assembly
Miniaturized computing hardware soon reached independent hobbyists. The January 1975 issue of Popular Electronics published a famous cover announcing the Altair 8800, a put-it-together-yourself microcomputer kit priced at $439 from MITS in Albuquerque, New Mexico. Built around the Intel 8080 microprocessor—a successor to Faggin's 4004—the Altair 8800 demonstrated that individuals could buy and assemble dedicated personal machines.
By 1977, consumer computing shifted from self-assembled kits to factory-built appliances. That year brought a trinity of pre-assembled microcomputers: the Apple II, the Radio Shack TRS-80, and the Commodore PET. Although modest in processing power compared to institutional mainframes of the era, these affordable desktop computers allowed individual owners to run software, write code, play games, and attach modems right from their homes.
Dial-Up Services and Walled Communities
As microcomputer modems enabled communication over telephone lines, commercial entities recognized the potential for consumer networking. In July 1979, two subscription services—CompuServe and The Source—launched almost simultaneously, offering dial-up access to digital networks. Subscribers equipped with personal computers and modems dialed into central mainframes to read news, exchange messages, and socialize.
These early commercial online services functioned as centralized walled gardens. Operating from single data centers, subscribers on CompuServe could not communicate directly with users on The Source, and content remained under strict corporate oversight. Despite these limits, subscription services proved that personal computers could foster vibrant online communities over long distances.
Decentralized Networks and the Borderless Internet
A decentralized alternative emerged in late 1979 with Usenet. Built as an asynchronous message-sharing protocol, Usenet operated without a central host mainframe. Independent servers stored local copies of newsgroup discussion forums. When a user posted a message to one server, it passed the text to neighboring servers until the update spread across the network. Usenet scaled rapidly, growing from 3 initial servers in 1979 to 11,000 by 1988.
Because Usenet and the broader internet lacked central authority, they developed open, free-wheeling communities alongside new security vulnerabilities. In December 1988, Robert Tappan Morris, a 23-year-old graduate student at Cornell University, created the Morris Worm. Intended by Morris to measure the network's size, the code contained a replication flaw that repeatedly re-infected Unix computers across the internet. The resulting traffic surge brought down large sections of the network, making front-page headlines and introducing the broader public to the internet.
By consolidating complex logic operations onto pocket-sized microprocessors, early hardware engineers unwittingly laid the structural foundation for global digital interaction. The hardware shift that moved computing from city-block vacuum tubes to affordable silicon chips ultimately transformed isolated desktop terminals into nodes of an interconnected global network.
Cover: AI-generated editorial illustration; not a documentary photograph.
Source: A Web Around the World, Part 10: A Web of Associations
SIGNAL TITANS.jpg){kind=link}
Leave a Reply