Separate datum, data, and information, and show how processing turns one into the next.

• Data are the fuel of computing — data are the reason for computing, networking, automation, and all digital processing; we constantly take in data through our senses.
• Datum vs data — a datum is a single value, attribute, characteristic, or identity of something (Merriam-Webster: “something given…as a basis for reasoning”); data is the plural of datum, so two datums are data.
• Information — the result or product of data being processed; it has context relevant to an inquiry. “Information without context is just data.”
• Context gives meaning — “blue” and “sky” are each a datum with little meaning alone; combined as “blue sky” they gain context and can answer “What color is the sky?”
• IPO model — the most basic model for processing data into information is Input → Process → Output; even complex programs break down into IPO steps.

Trace the five computer generations and separate “computing” from “computers.”

• Miniaturization trend — computers started large and shrank as technology advanced; what once filled a university gymnasium now fits in a pocket.
• Five generations — First (1940–1956, vacuum tubes); Second (1956–1963, transistors); Third (1964–1971, integrated circuits + operating systems); Fourth (1971–2010, microprocessor, IBM PC, Apple Macintosh, GUI); Fifth (2010–present, artificial intelligence).
• Computing vs computers — computing is a processing operation (like the IPO model); computers are the people, machines, and devices that perform the processing.
• “Computers” as a job title — the term once described people who “computed” accounts into ledgers.
• Birth of the modern computer — historians generally place its relevant birth around 1940, though earlier developments also contributed.

Connect the 1957 Sputnik launch to the Cold War need for a survivable communications network.

• Sputnik trigger — in 1957 the USSR launched Sputnik (about 23 in. in diameter, ~190 lb), energizing the technology boom and putting space, satellites, and science on the public’s mind.
• Cold War fear — if the Soviets could orbit a satellite, they might launch a missile that could take down U.S. telephone communications; what was needed was a system that could withstand an attack and stay operational.
• Agencies formed — the U.S. government began research grants and formed agencies such as NASA and DARPA to investigate advanced technologies (rockets, weapons, computers).
• Licklider’s “galactic network” — in 1962 MIT scientist J.C.R. Licklider proposed a network of interconnected computers that could communicate should the phone system be destroyed.
• Survivability via packet switching — Paul Baran and Donald Davies independently devised transmitting “packets” across a switched network; Davies’s term “packet switching” stuck, and it was the key to a network that could not be taken down by a single site’s destruction.

Follow ARPANET through TCP/IP to the networks that became today’s internet.

• ARPANET — designed by Licklider’s team in 1966 and ready for testing in October 1969; the first message “LOGIN” (UCLA → Stanford) failed after “L” and “O,” but it proved the Interface Message Processor (IMP), the early precursor to a router.
• From internetwork to “internet” — as ARPAnet added nodes (going international in 1971), it became an internetwork — a network of networks interconnected for communication; “internetwork” was shortened to “internet.”
• TCP/IP — Vint Cerf and Bob Kahn developed TCP (originally including IP’s functions) in the early 1970s; in the early 1980s TCP/IP replaced the Network Control Protocol (NCP) and made a worldwide network possible.
• TCP/IP model layers — the four-layer reference model, top to bottom: Application, Transport, Internet (Network), and Link (Network Interface/Access).
• Key protocols — TCP (connection-oriented, guarantees delivery), IP (addressing), HTTP (transfers webpage content), UDP (loss-tolerant, no guarantee), POP3/SMTP (email), and FTP (file transfer).
• NSFnet and others — NSFnet (1986) grew and replaced the older, slower ARPAnet in the late 1980s; NSFnet shut down in 1995 and control was privatized, opening the network to commercial use.

Name the people and pieces that turned hypertext into the World Wide Web.

• Hypertext pioneers — Vannevar Bush conceived “Memex” (“As We May Think,” 1945); Ted Nelson coined “hypertext” in 1965 (part of “Xanadu”); Doug Engelbart demonstrated NLS in the 1968 “Mother of All Demonstrations” and is credited as the “inventor” of hypertext (also the mouse and GUI); Brown University (Andries van Dam) built HES (1967) and FRES (1968).
• Tim Berners-Lee & CERN — frustrated by fragmented information, Berners-Lee (an engineer at CERN) proposed “Mesh,” then with Robert Cailliau a formal “WorldWideWeb” (W3) project: a web of hypermedia documents viewed with a browser on a client/server architecture.
• Three central web elements — HTML (a web markup language based on SGML), the Universal Document Identifier (UDI, foundation for the URL/URI), and HTTP (with four design principles: client/server, request, response, termination).
• Free to everyone — CERN announced the web would be free to everyone, just as the University of Minnesota announced Gopher would start charging fees — spiking interest in the web.
• Browsers — Nicola Pellow built the Line Mode Browser (libWWW), the first generally available browser; Mosaic (Marc Andreessen & Eric Bina, NCSA, 1993) was the first generally available browser to display text and graphics together; Andreessen then founded Netscape (Navigator, 1994), and Microsoft licensed Mosaic code to build Internet Explorer (1995).
• Web phases — Web 1.0 (Static/Read-Only), Web 2.0 (Social/Read-Write — sharing, blogs, wikis, social networking), Web 3.0 (Semantic Web — one large database, blockchain, decentralization), Web 4.0 (Intelligent/pervasive/ubiquitous Web).

Define the client/server relationship and how virtualization simulates hardware in software.

• Client/server — like ordering in a diner: a client (a network host) requests a resource, and a server (specialized software on a centralized system) retrieves it and sends it back.
• A server is software — it performs, manages, or controls special functions on a network; one physical computer can run multiple servers, and a server can also act as a client. Types include application, mail, proxy, virtual, and web servers.
• Client types — thin clients have very few on-board resources and run apps on servers; thick (fat/rich) clients have all resources on-board; hybrid clients lean thick but rely on servers for centralized services.
• Virtualization — a virtual object (machine, server, router) is the simulation of its functions through software; creating virtual devices via software is called virtualization.
• Hypervisor — the key element of a virtual environment, also called the virtual machine manager (VMM); it is essentially the OS for the virtual environment. Type 1 runs “bare metal” directly on hardware; Type 2 runs on top of a host OS.
• Network virtualization — virtualizing a network splits bandwidth into manageable channels assigned in real time; storage virtualization makes resources on multiple servers appear as one entity; server virtualization makes multiple servers appear as one.

Separate cloud service delivery models (the “what”) from deployment models (the “who”).

• Cloud computing — access to resources or processing located beyond the local firewall, provided by dedicated (typically virtualized) internet-based servers; it shifts local processing to remote providers over the internet.
• Why “the cloud” — telephone diagrams used a cloud image for switching between central offices (POTS/PSTN); the cloud came to represent the randomized switching of the internet.
• Delivery models (the “what”) — NIST defines three: SaaS (subscribers access applications on a CSP’s servers via the browser), PaaS (customized programming/testing/modeling environments), and IaaS (servers, storage, networking, and virtualization delivered over the internet, pay-per-use).
• Other “aaS” offerings — AIaaS, BaaS (backup), CaaS (a “full meal” combining SaaS/PaaS/IaaS), DaaS (desktop), NaaS (network), UaaS (unified communications), and VaaS (video-conferencing).
• Deployment models (the “who”) — NIST defines three: private (restricted access, in-house or via a CSP), public (a private cloud offered over a public network; membership is the constant), and hybrid (public deployment with private-cloud restrictions). A community cloud is a common variation.