Before Wi-Fi and Ethernet, these 3 forgotten standards kept computers connected

Published Oct 31, 2026, 7:30 AM EDT Andrew Heinzman is a seasoned tech writer with over half a decade of experience. He specializes in delivering insightful tech news and detailed product reviews, particularly focusing on audio and video technology, as well as computers and smartphones. His journey in tech writing includes a two-year stint as the News Editor at Review Geek, the former sister site of How-To Geek. Throughout his career, Andrew has consistently provided readers with valuable insights and up-to-date information on the latest developments in the tech world, earning recognition for his clarity, depth, and relevance. In his free time, Andrew likes to explore music and read books, just as he did when earning his bachelor's degree in English literature. Wi-Fi and RJ-45 Ethernet are relatively new inventions in the history of computing. But the idea of two computers talking to each other existed before computers were ever even made. There are dozens of standards that kept early computers connected, but few of them are even brought up when people talk about super-old machines. For reasons that should be obvious, I'm not going to bother writing a giant all-encompassing textbook of networking history. There are college courses for that. But I will highlight three old networking standards that are worth knowing about. Attached Resource Computer Network (ARCnet) The first widely adopted LAN system for offices Attached Resource Computer Network, rightfully abbreviated as ARCnet, was developed in 1976 by John Murphy at Datapoint Corporation as a tool to connect terminals to a floppy disk system. It dominated offices before Ethernet's commercial introduction in the '80s (more on that later) because it utilized an agnostic design. Earlier LAN tools required a homogenous network, while ARCnet let you connect different brands or types of terminals and peripherals together. ARCnet cables are usually RG-62/U coaxial. They look similar to television coaxial cables, but they use a higher impedance (93 ohms over TV coax's 75 ohms) for high signal attenuated data pulse transmission. I'm mentioning this because it rules out high-speed data transmission. Standard ARCnet equipment operates at 2.5 Mbps and can only handle packets 508 bytes in size. Plus, it relies on a token system, meaning that each machine in a network must wait for its turn to send data. The ARCnet system also wastes a lot of bandwidth on its four-packet handshake protocol—a computer has to ask if it's allowed to send data, it gets the confirmation, it sends the data, and when that's all done, the computer that did all the sending needs to receive a signal to confirm that the data was actually sent. These flaws naturally put ARCnet at a disadvantage when Ethernet and other (relatively) fast alternatives hit the scene in the '80s. Though, ironically, ARCnet lived on into the 2000s because its limitations made it an extremely reliable option for some industrial and embedded systems. Errors and mistakes are curbed by "wasteful" handshaking, and the high-impedance, low-bandwidth design of standard ARCnet can support cable runs up to 2,000 feet long. X.25 packet-switching Connecting the world one transaction at a time Credit: Joe Fedewa / How-To GeekCredit: MM Printing / MakerWorld While ARCnet accidentally benefitted from a super-slow, ultra-reliable design, the X.25 protocol was intentionally built around these ideals in order to ensure long-term international use. X.25 isn't a cable, it's a packet-switching protocol for WAN networks, and it gave the world a robust credit card and ATM network long before the average person had even heard the word "internet." Finalized in 1976, the X.25 protocol was quickly adopted by banks and retailers for remote transactions. It worked sort of like a telephone system combined with a mail system. You go to an ATM, request funds, and the ATM forms a virtual circuit with a computer that has access to your account. These two machines then use the X.25 protocol to send data in several chopped-up packets, all of which are numbered in sequential order. If a packet is damaged or lost in transmission, one of the machines can request that the packet be sent again, which ensures an extremely high level of accuracy over a low-bandwidth, unreliable connection. The "unreliable connection" part is important here because, again, X.25 is just a packet-sending standard. It was developed by telecom companies to work over telephone lines, which are noisy and unreliable (though it can operate over other physical networks, too). When a computer using X.25 creates a "virtual circuit," it's just mirroring the process of dialing a phone number, either by reaching out to a specific address or by requesting that address through what is essentially a virtualized telephone operator. X.25 was slowly replaced by a handful of different standards, most notably TCP/IP, in the decades after its 1970s introduction. But it still lingers on in some corners of the world, most notably in hobbyist, maritime, and aviation radio, where X.25 and its variants provide a reliable medium for communication and data transmission. Thicknet and Thinnet These are Ethernet, but not in a form you'd recognize Credit: Fdominec/WikiMedia Commons We tend to associate Ethernet with the RJ-45 connector (or the 8P8C connector, more accurately). But the original Ethernet spec, designed at Xerox PARC by Bob Metcalfe and a team of co-creators in 1973, utilized coaxial cable. When Ethernet finally became commercially available in the 80s, the two most common implementations of the tech were Thicknet and Thinnet, which retained a coaxial design despite the fact that RJ-45 Ethernet was just around the corner. Officially known as 10BASE5 and occasionally referred to by an array of stupid nicknames, such as "Thick Ethernet," Thicknet cables are, perhaps unsurprisingly, thick. Imagine a fatter, denser, more rigid version of a television coaxial cable. Thicknet cables don't like to be snaked around corners, they're kind of heavy, and they're relatively expensive due to shipping and manufacturing costs. Notably, Thicknet cables can't plug directly into anything. They don't have built-in connectors. Instead, they need to be manually soldered or crimped into place. And if you're connecting a Thicknet cable straight to a workstation, you need to use an external unit with a vampire tap, which is essentially a spiked clamp that bites through the cable's insulation to make contact with the internal wires. All that work for a maximum transfer rate of just 10 Mbps, mind you. The Thinnet standard, officially called 10BASE2 and sometimes referred to as "Cheapernet," shrinks things down to a less expensive, more workable cable with proper BNC connectors. This is the standard that appeared inside of office buildings, and although it retained the 10 Mbps throughput of Thicknet, it operated at a shorter range and supported fewer concurrent connections. So, be thankful for modern Ethernet cables and connectors, even if they get stuck sometimes. We're just scratching the surface These are just three of the networking standards that powered the world before Wi-Fi and RJ-45 Ethernet's takeover. And some of you won't even support my choice to include Thicknet or Thinnet, which I just had to mention because they predate the RJ-45 Ethernet that most people are familiar with (though I admit that the funny names were the main reason for their inclusion). The good news is that there's no shortage of LAN and WAN technologies that predate Ethernet. ISDN, IBM Token Ring, and University of Hawaii's wonderfully-named ALOHANET all deserve a mention. Maybe we'll get the chance to spend more time talking about them some other day.