Imagine a game that begins with a printout that reads: You are standing at the end of a road before a small brick building. Around you is a forest. A small stream flows out of the building and down a gully. In the distance there is a tall gleaming white tower. Now imagine typing some information into a teletype and then reading the next printout. And then another. A trail of paper lists your every move. This is interactive gaming in the 1970s. Later versions had a monitor so a screen could just show a cursor and the player needed to know what to type. Type N and hit enter and the player travels north. “Search” doesn’t work but “look” does. “Take water” works as does “Drink water” but it takes hours to find dwarves and dragons and figure out how to battle or escape. This is one of the earliest games we played and it was marvelous. The game was called Colossal Cave Adventure and it was one of the first conversational adventure games. Many came after it in the 70s and 80s, in an era before good graphics were feasible. But the imagination was strong.  The Oregon Trail was written before it, in 1971 and Trek73 came in 1973, both written for HP minicomputers. Dungeon was written in 1975 for a PDP-10. The author, Don Daglow, went on the work on games like Utopia and Neverwinter Nights Another game called Dungeon showed up in 1975 as well, on the PLATO network at the University of Illinois Champagne-Urbana. As the computer monitor spread, so spread games. William Crowther got his degree in physics at MIT and then went to work at Bolt Baranek and Newman during the early days of the ARPANET. He was on the IMP team, or the people who developed the Interface Message Processor, the first nodes of the packet switching ARPANET, the ancestor of the Internet. They were long hours, but when he wasn’t working, he and his wife Pat explored caves. She was a programmer as well. Or he played the new Dungeons & Dragons game that was popular with other programmers. The two got divorced in 1975 and like many suddenly single fathers he searched for something for his daughters to do when they were at the house. Crowther combined exploring caves, Dungeons & Dragons, and FORTRAN to get Colossal Cave Adventure, often just called Adventure. And since he worked on the ARPANET, the game found its way out onto the growing computer network. Crowther moved to Palo Alto and went to work for Xerox PARC in 1976 before going back to BBN and eventually retiring from Cisco. Crowther loosely based the game mechanics on the ELIZA natural language processing work done by Joseph Weizenbaum at the MIT Artificial Intelligence Laboratory in the 1960s. That had been a project to show how computers could be shown to understand text provided to computers. It was most notably used in tests to have a computer provide therapy sessions. And writing software for the kids or gaming can be therapeutic as well. As can replaying happier times.  Crowther explored Mammoth Cave National Park in Kentucky in the early 1970s. The characters in the game follow along his notes about the caves, exploring the area around it using natural language while the computer looked for commands in what was entered. It took about 700 lines to do the original Fortran code for the PDP-10 he had at his disposal at BBN. When he was done he went off on vacation, and the game spread. Programmers in that era just shared code. Source needed to be recompiled for different computers, so they had to. Another programmer was Don Woods, who also used a PDP-10. He went to Princeton in the 1970s and was working at the Stanford AI Lab, or SAIL, at the time. He came across the game and asked Crowther if it would be OK to add a few features and did. His version got distributed through DECUS, or the Digital Equipment Computer Users Society. A lot of people went there for software at the time. The game was up to 3,000 lines of code when it left Woods. The adventurer could now enter the mysteriou