intfiction.org

Well, command line interface works with such obscure commands as "dir /p", "cd", and such conventions as "X: to access drive X".

IF works with such commands as "look", "take", "inventory", "ask john about the fire".

Command line interface allows you to navigate and interact with the contents of your computer.

IF allows you to navigate and interact with a modelled virtual world.

You COULD try navigating a command line interface with "go north", but I'm not sure you'd go far.

Also: apparently there were some versions of DOS where "xyzzy" actually had a response.

I think you could argue either way; put succinctly: the command line interface is simpler in its implementation, while the IF interface is simpler in its ease-of-use. I'm hoping to use tab-completion and broad hints to make using my command line easier than dropping someone into a full command line...but, we shall see.

I seem to remember there was a game for the Amiga 500 (Yes I AM that old )
I believe it was called "The fourth protocol" It was a bit like the game you're creating.
In this game, as I remember, you had to hack your way through various system gathering information or something like that.
It was a challenging game and I remember spending countless hours on it.
If your game is anything like it, I'm looking forward to play it.
Good luck

_________________
Work in progress: D-Day (Adrift V.5)

I'm confortable with *nix shells, particularly bash. Are the puzzles meant to be solved via shell scripts?

Merlin, in a command line you type cryptic commands for the machine to do something, in a IF parser you type commands resembling natural language for the player character to do something.

_________________
IF without puzzles is simply like a novel where you turn pages by typing "go north" or other silly and useless nonsense.

They're not cryptic. They're well documented. They are only cryptic to someone who can't be bothered to RTFM

There are several games you may want to check out: Uplink and the Hacker Evolution series. Both have the model of hacking into a server, stealing info, selling it, etc. They may give ideas for what to do or not do in your engine.

Another game that definitely should be mentioned here is the classic old AGT text adventure, CosmoServe by Judith Pintar. I played a little of it a long time ago. It's sort of a simulation of the ancient command-line Internet (specifically CompuServe). And that description may be totally wrong, because the CosmoServe was written the same year I was born, and I don't claim to understand how online communities worked back then, before the web.

I'm almost sure it was (at least also) for the C64. There is where i bashed my head countless hours.

Supposedly, the Wachovskies' Matrix sequel tie-in game, Enter the Matrix, had a sequence like this, too. But I never got more than 5 minutes into that game.

I don't think the documentation, or lack thereof, is what makes them cryptic. Memorability makes a big difference. If you forget "z" you can type "wait" - but if you forget the switches and options you need for "zip," or the options available on the current platform's version of "find," you have to look them up again. Then there's the test command and loop syntaxes, which vary between bash and c-shell varieties...

Actually, I wouldn't call the Linux command line "well documented." Man pages are written for people who already know the commands and want to use them more effectively, not for people who don't even know which command to use.

I went for option #2 because OS X, FreeBSD and Linux are all pretty much the same OS from the perspective of the command line. I don't remember Apple II+, C64, Atari 800, Acorn or VMS command line, and I have to struggle with Windows, although I use it at work.