3.2 Layer 2 Protocol Overview - LAN Protocols
3.2.1 Brief history of Ethernet
Ethernet was born of internal research conducted at Xerox's famed Palo Alto Research Center in the early 1970s, following pioneering work of Norman Abramson at the University of Hawaii on the ALOHANET project (using packet radio between the islands in the late 1960s). Rather than being some bold and innovative new technology that researchers strove to bring to market, the origins of Ethernet are quite pedestrian. The researchers needed a way to share their expensive laser printer among their desktop workstations. The answer: a crude and somewhat chaotic data link layer protocol that relied on higher-layer protocols to behave properly. Somewhat tongue-in-cheek, these researchers named their new pet Ethernet, a subtle play on words, referring to the 'ether' used by physicists in the 1800s to describe the underlying structure of space (later proved to be scientifically unsound). From this humble beginning spawned a tremendous family of related products, protocols, and technologies.

Today, many different varieties of Ethernet exist. This family includes at least five different frame formats, three different media access arbitration techniques, and an ever-growing collection of medium-dependent interfaces (MDIs). The MDI is the most visible aspect of the 802.3 physical layer; it defines the expected type of transmission medium, as well as its transmission and impedance characteristics. For the LAN to operate properly, the correct transmission media must be connected to the corresponding NIC, which contains the MDI logic.