New Development Language. (READ THIS)

4 views
Skip to first unread message

Thomas J. Palmer

unread,
Feb 17, 1995, 2:50:26 PM2/17/95
to

How many of you would be interested in a C++ library to handle
parsing and other aspects of IF?
Because I'm interested in making one. I would probably
distribute it freely (though copyrighted) as source code that could be
compiled on any system and used as a link library.

Furthermore, I'm thinking about new ideas for game construction.
It would be a development in the basic structure of IF. Think, instead
of rooms actually storing physical data that would be interpreted into
text. This could be used in conjunction with the old rooms thing too.
Imagine storing a map of the countryside where coordinates are kept of
everything and user input could be more freeform, or kept simplified to
old compass directions. Also, imagine a property simulator that kept
track of objects' durability, flammability, sizes, and so on for the sake
of less linear story control and improved input flexibility. (Realize
that this could coexist with old rooms/hardwired control which has a
number of benefits.)
Now put good data editing with that, to make world construction
easier, and put it with the C++ library. I'd like to implement this.

Any opinions?

- Tom Palmer

----------------------
And what a .sig it is!

Jadrian Zun

unread,
Feb 19, 1995, 1:22:29 PM2/19/95
to
In article <Pine.ULT.3.91.950217...@bert.cs.byu.edu>,

Thomas J. Palmer <le...@bert.cs.byu.edu> wrote:

> How many of you would be interested in a C++ library to handle
>parsing and other aspects of IF?

We are attempting to do almost the same thing, in our boogum virtual world
project. It will actually end up being a mix of MUD and IF technology, we
hope.

The boogum driver is the game engine (like z-machines etc) on top of which
is built the world. The first experimental world being built is the Cafe,
a simulation of a coffee shop.

Eventually boogum aims to simulate worlds as realistically as is feasible,
with a framework that allows multiple players to interact simultaneously.
The focus will mainly be on role-playing and puzzle-solving gaming worlds,
but other applications are surely possible.

Currently it is in pre-alpha, version 0.8, and is nothing more than
a simple driver which accepts telnet connections. The following classes
are slated for the next version: World, Domain, Place, Thing, Container,
Body.

> Furthermore, I'm thinking about new ideas for game construction.
>It would be a development in the basic structure of IF. Think, instead
>of rooms actually storing physical data that would be interpreted into
>text. This could be used in conjunction with the old rooms thing too.
>Imagine storing a map of the countryside where coordinates are kept of
>everything and user input could be more freeform, or kept simplified to
>old compass directions.

In addition to scripted, pre-written rooms, we will have rooms whose
descriptions are constructed on the fly from, as you term it, 'physical data'.
This would be very useful in large areas such as the outdoors.

Also, all rooms have a universal coordinate. Something we'd like to do in
the future is include line-of-sight revelation of things happening in other
rooms that you can presumably see.

>Also, imagine a property simulator that kept
>track of objects' durability, flammability, sizes, and so on for the sake
>of less linear story control and improved input flexibility. (Realize
>that this could coexist with old rooms/hardwired control which has a
>number of benefits.)

All Things in boogum have physical characteristics, size, weight, bulk, etc.
This includes your own Body :) (Body is derived from Thing, in c++ terms)

There are many many more ideas and concepts in the cooker, so to speak..
for instance, sense propogation, where sounds and scents are propogated
through the world so you can smell and hear things :) AI-driven non-player
characters are another biggy.

To sum it up, boogum aims to provide a realistic world framework for both
scripted storylines and unscripted, user-driven action. In IF, it seems that
a lot of effort is expended to round out worlds, making sure all the objects
respond in kind to all verbs. Note the many releases of Curses, all of which
make the world complete, however, there are _still_ things you can't do (trying
to climb the mast in the ship or look at it comes readily to mind).

Anyone interested in discussing this further please e-mail me at
z...@foop.mit.edu. One topic of especial interest is parsing.. we have a rather
simple-minded parser right now :p.

. . . Zun.

Tim Middleton

unread,
Feb 22, 1995, 6:19:00 AM2/22/95
to
I have been reading about Zun's text based virtual reality aspirations, and
it does sound good. But it also raises a question in my mind that I have yet
to resolve for myself--and that is, does too much detail (as much as I
personally am drawn to it) detract from the STORY.

If I tie all the posts I've been reading here together I then think about
someone's post suggesting that one fill in the details after the basics have
been accomplished. This makes a certain amount of sense and I find that as
I am playing with Inform that is what is naturally evolving for me (I
think). I find if I keep dwelling on room details i get bogged down by it
and the story does not progress.

But of course, i wouldn't want to say my experience is the only experience.
As with most creative processes, people have wildly different approaches
that seem to work for them.

The thing that I find most curious though is the difference between text
adventures that are really mostly a string of rooms with some objects placed
in such a way as you have to collect them and use them in a certain order
and place to get to the next area. There is a plot of course, to thread it
along, but (and I'm not saying this is bad) really it's a puzzle game more
than a work of literature (for example).

There is a place in the post suggesting to leave the details to the end (and
in fact Zun referred to this as well) where it is suggested a rich victorian
england environment could be created for sherlock holmes adventures.

I'm sure gas-light objects would not be very hard to code in stunning
realism (ha!), but that aside I then think of sherlock holmes stories and
think... but the only common setting in any of them is 221b Baker street!
The rest of the settings are always original and created just as the STORY
needs. In fact Doyle's stories are very exacting, there is not a lot of
detail or flowery description given except what is essential (even if it
doesn't seem essential when someone is describing something).

I'm also not against creation just for the sake of creation. For example I
think it would be great fun if someone coded a startlingly detailed 221b
Baker street with no plot or story at all! It would be interesting to just
wander around and see what's there (if it's well written) I think. However
that might just be an eccentricity of mine.

The thing that really fascinates me about this whole I-F thing is the
theoretical possibilities of something more panoramic though. I wonder if I
am capable of write an I-F adventure that it's not the puzzles that draw you
along but the plot, and the story, and the characters. And for this to
happen I then wonder if too much superfluous detail isn't in the end doing a
disservice to the story and distracting the reader/adventurer from what as a
writer the author wants to direct them to. Perhaps this is slightly more of
a deterministic vision than I-F should be (if anyone can say what it
`should' be), but the question remains as how effective a vehicle for a
LITERARY endeavour (at least experimental) it could be.

However for now, in my Inform diapers, I too, trying to grasp the concepts
of the tool, go for detail object and environmental! My favorite object so
far being my "microwave oven-like" object. Heheh. And there is beauty in
it. But I don't think it's the direction that I want to go in for myself if
I should actually go in any direction, eventually. (As much as I love
Origin's ULTIMA games (except ultima 8!) which try also very much for as
much environmental realism as possible, in a graphical way--)

I think though, as an example, of the photograph on the wall in the dark
room in `Curses'. I was struck reading it's description (my god, what DRY
wit Graham has!) about it looking as if the gentleman in it was being held
up momentarily, while the picture was taken, by some sort of metal
clamps--and he was (it goes on to add!). That's the sort of excellent and
subtle commentative description only text makes possible. No matter how
funny or strange that picture might have appeared if `Curses' happened to be
a graphical adventure there is no way that the player would have gleaned
that whimsical detail.

It's with examples like that that I'm inspired that perhaps I-F could be for
me a (minor?) vehicle for literary expression. Adventures are fine, and I
even read some occasionally--but the fiction I'm most drawn to is usually
of a more psychological nature and based on character rather over
plot--exploration over problem solving--and above all they are usually
trying to express something profound about the author's vision of life and
reality. Now can such a thing be made interactive... or is it best left to
the disguised soap box of a book?

I do not exactly know, all I know is it's all very interesting--VERY
interesting!

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Tim Middleton -=-=-=-=<when sense makes no sense>=-=-=-=- 02/22/95 06:20
<Internet: tim.mi...@canrem.com>=-=(Fidonet: 1:229/15)=-={Rime:->118}
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
I believe in the Kingdom Come
Then all the colours will bleed into one
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
I Still Haven't Found What I'm Looking For -U2


---
d CREEPERmail d It's cold outside, but brightly lit...

David Gilbert

unread,
Feb 23, 1995, 9:14:54 AM2/23/95
to
[SNIP]

[Very minor "Shade's of Grey" spoilers]

: The thing that really fascinates me about this whole I-F thing is the


: theoretical possibilities of something more panoramic though. I wonder if I
: am capable of write an I-F adventure that it's not the puzzles that draw you
: along but the plot, and the story, and the characters. And for this to
: happen I then wonder if too much superfluous detail isn't in the end doing a
: disservice to the story and distracting the reader/adventurer from what as a
: writer the author wants to direct them to. Perhaps this is slightly more of
: a deterministic vision than I-F should be (if anyone can say what it
: `should' be), but the question remains as how effective a vehicle for a
: LITERARY endeavour (at least experimental) it could be.

I agree. A prime example of this is "Shades of Grey." The puzzles in this
game aren't too extensive or mind-boggling, but the detail and originality
in the writing still keeps me interested a year after solving it.

Think of "A Mind Forever Voyaging," the landscape of Rockville was HUGE!! And
most of it might have seemed superfluous, but to remove it would remove a
vital aspect of the game.If it wasn't written well, it just would have been
plain annoying to play. But it's interesting to explore this world as it moves
forward in time; to see how it changes. AMFV could have been made a heck of a
lot smaller, but the game just wouldn't have been too interesting, and the
player wouldn't have gotten the full impact of what the author was trying to
say. There aren't any read "puzzles" until the third part or so, and even
those aren't too mind-bending to solve (at least, compared to others I've
seen), but it's the plotline that keeps this game going, not the puzzles.

I finished both of these games with a feeling of being told a good story. And
the good thing about the so-called "superfluous" details just give you more
things to look out for the next time you play. I can play AMVF a hundred
times and still find something different. Nothing's WRONG with that, at least
in my opinion. I enjoyed solving SOG and
AMVF just as much as I enjoyed the mental puzzles of "Curses" and
"Zork."
And in the end, it's just as satisfying.

Only from the warped mind of,

David L. Gilbert

SueLong

unread,
Feb 24, 1995, 11:29:24 AM2/24/95
to
Add my vote to those who would like to wander around in a rich, detailed
environment, interacting with non-playing characters and getting
information from them, solving a mystery instead of moving objects around.
I have been working on an Arabian-Nights-like city for some time. The
hard part is to build in a plot, in my case a conspiracy and a murder to
be solved, while still allowing the player free choice to move and
explore. It is like a movie composed of tiny individual scenes, but where
the player might not encounter them in the order in which I plan them. I
think this is the trickiest part of designing hypermedia - the author
loses control over the flow of information to the reader. The individual
scenes have to stand on their own, and there has to be a lot of redundancy
built into the information you can get from different characters. Any
advice or comments?

Sue Long

Tim Middleton

unread,
Feb 28, 1995, 8:20:00 PM2/28/95
to
Hello David.

(I apologize for taking so long in my replies. The mail is very slow at my
site and sometimes can take up to four days coming and going! For bizarre
technical reasons that actually make no sense at all--hrm!)

Okay, now that that parenthetical disclaimer is done, on to bigger and
better disclaimers! Yahoo! There is no stopping me now.

I have to unburden myself, at this point, with a startling confession. I
would throw myself on the interactive (though hopefully not fictitious)
mercy of the group! My confession may not even come as a surprise to many,
but here it is: I am ignorant.

Sigh.

The fact of the matter is that though I am a fan of interactive fiction/text
adventure i don't have a really wide exposure to a lot of it. And in fact
many of the things I mentioned in my previous article may exist--and things
i have not even imagined may exist. I of course have Zork in by shady past,
and Adventure, and a few others... Hitchhiker's Guide comes to mind. But
still, I have come to realize that there is quite a lot out there I've not
been exposed to, and i'm endeavoring to expose myself even as I type.

So what I'm saying (will I ever get to the point?!) is, for one thing, I
appreciate in your response your mentioning a few titles in relation to your
perspectives. It gives me something to check out, as time permits. I already
tracked down "shades of gray" and *think* "A mind forever voyaging" is in
Lost Treasures volume 2... anyhow I will keep these in mind.

DE= Think of "A Mind Forever Voyaging," the landscape of Rockville was
DE= HUGE!!

I should like to make myself clearer on this concept of size; as I have a
feeling you are talking size in a sort of physical dimension here (though i
could be wrong). When I spoke of bigness (and I believe "panoramic" was the
work I used), I was speaking more figuratively. I was thinking of lushness
of description, as one might find, for example in a Joseph Conrad novel.
(Sheesh, why am I using Conrad for an example?? I don't much like the guy!
<G>) Or perhaps certain Arthur C. Clarke writings. For example in his book
"Rondezvous With Rama" (I'm not picking great examples tonight, but oh
well!) the novel is not very big, but you get the feeling due to the
description that you've been somewhere entirely and mind bogglingly immense.

Um, well, I'm not saying all adventures should take place in large
cylindrical abandoned spacecraft--i'm just trying to get at (and not very
well) a certain quality of description and prose.

DE= I finished both of these games with a feeling of being told a good
DE= story.

Yes, that's good. The author has certainly succeeded if that is the case.

DE= And the good thing about the so-called "superfluous" details just give
DE= you more things to look out for the next time you play.

Heheh, I hope you didn't take "superfluous" in a bad way. I was only
pondering the danger of certain types of details or clutter to take away
from the plot. Clutter or detail certainly need not take away from a plot.
And also you make a very good point, even if it does, it has certain charms.
Especially for replayability.

I also don't want to be dogmatic. I'm sure lots of people have different
perspectives for different reasons. I'm not (really) trying to claim
anything is -better- than anything else (at this time, at least <G>). I'm
just musing and trying to centre in on a particular style I have a vague
notion of ruminating in my brain.

---
d CREEPERmail d Please don't confuse rationality with empiricism.

Jim Edwards-Hewitt

unread,
Mar 1, 1995, 5:27:29 PM3/1/95
to

> I have been working on an Arabian-Nights-like city for some time. The
>hard part is to build in a plot, in my case a conspiracy and a murder to
>be solved, while still allowing the player free choice to move and
>explore. It is like a movie composed of tiny individual scenes, but where
>the player might not encounter them in the order in which I plan them. I
>think this is the trickiest part of designing hypermedia - the author
>loses control over the flow of information to the reader. The individual
>scenes have to stand on their own, and there has to be a lot of redundancy
>built into the information you can get from different characters. Any
>advice or comments?

First, forgive me if this discussion has long since passed me by; my
newsfeed seems to be about a week behind right now. Anyway...

A possible solution comes to mind based on some experiences with standard
(non-computer) role-playing games. First, let me tell briefly what I'll
call the "Tale of Two Gamemasters".

A gaming group I was in had two gamemasters, who took turns running games.
They both created detailed maps and background for their adventures, but in
execution there was a noticeable difference. With one of them (call him
"A"), we never seemed to go too long without encountering some clue or
adversary, though he didn't give things away unless the characters
legitimately discovered them, and we could usually count on having some
sort of dramatic progression to each evening of the game (which ran one
night a week for several weeks.) With the other (call him "B"), we would
sometimes have long stretches where we were exploring places that turned
out to be unimportant, and sometimes found our objectives only by
brute-force searching.

The "secret" of gamemaster "A" was that he didn't actually have everything
in a fixed location. If the characters were in an area where there seemed
to be several choices, one of which led to further adventure, and if we'd
made anything like a reasonable effort to find the right way, the one we
chose would lead there. We had the enjoyable illusion of having a
completely free choice, but without the meaningless wandering that too
often characterizes simulations.

I think a similar concept might work for fitting plot into a rich simulated
environment. Basically, I'm thinking that you would have a "plot
landscape" that would not have to connect in a fixed way to the simulated
physical landscape. As a simple example, to get the player-character
involved in a murder investigation, if you wandered past a dark alley, an
informant might whisper to you out of the dark, but if you wandered into
the Casbah, he might tug at your sleeve in a crowd.

Now obviously, you'd want to have some flexibility, so the player does not
feel that his actions don't make any difference. But I think that if you
think of the plot elements as sequences of things that can happen to the
player, rather than things that relate to a "location" in the traditional
interactive fiction sense, it should be possible to construct something
that is neither a walk-through nor a random collection of events.

I would imagine that some of the basic elements of a plot landscape would
be:

1) Sequence - there are certain events that can't happen until a trigger
event occurs. This can be used to help keep the plot from running away
without the player. For dramatic purposes, it is probably worth
artificially constraining some events until the player is involved in them,
even if real people might be inclined to progress on their own.

2) Time - it should take different amounts of time to do different things,
such as travel different places, have a fistfight, etc., rather than the
usual one-action, one-turn setup. This may be less necessary for some
plot types, but action and suspense plots frequently involve needing to do
something "before it's too late", while dealing with obstacles that cause
delay.

3) Characters - The characters involved in an event or scene are more
important than the location.

4) Information - Rather than progressing from "room" to "room", or
physically searching areas, the player character should be guided to
possible plot events by information from other characters, papers, etc.
Most important, these pieces of information do not have to be fixed at the
beginning of the game/story; they can be determined in relation to the
player's actions.

I'm speaking in fairly vague theoretical terms here, of course. I'm not
saying any of this will be easy to implement, but I think it will be a lot
easier to create plot elements that interact to create a reasonable,
interesting story that it would be to create characters that interact to
create a reasonable, interesting story. The first step is to stop thinking
of the options as either a fairly fixed story that the player "solves"
(classic IF) or a fairly fixed world that the character blunders through
(pure simulation), and instead create a flexible story that interacts with
the player-character's actions.

-- Jim

---------------------------------------------------------------------------
Jim Edwards-Hewitt j...@visix.com Visix Software Inc.
---------------------------------------------------------------------------

Reply all
Reply to author
Forward
0 new messages