When should "all" be allowed?

20 views
Skip to first unread message

David Baggett

unread,
Mar 6, 1994, 4:55:01 PM3/6/94
to
An issue that frequently comes up in playtesting is when to allow multiple
direct objects. The problem is that allowing too much makes some puzzles
trivial:

-------------------------------------------------------------------------------
The troll eyes you hungrily. He begins drooling on your leg. "Me hungry,"
he growls.

>give all to troll

...
Kewpie doll: the troll, fearing the ancient legends, flees in terror
from the Kewpie doll.
...
-------------------------------------------------------------------------------

The player need not understand the reason behind the successful solution
here to get it right. Certainly, prohibiting the use of "all" with give-to
doesn't fix the problem -- it merely forces the player to use brute force.
Hopefully, however, this is enough of a deterrent if you have dozens of
objects that players will prefer to *solve* the puzzle properly (in this
case, figuring out what the connection between Kewpie dolls and trolls is).

The question is, what verbs should "all" (and multiple direct objects in
general) be allowed with? Unnkulia Zero doesn't allow all with anything
but basic inventory manipulation verbs like "get," "drop," "put," etc. You
can't, for example, say "turn all" or "press all" to twiddle all the
buttons or dials in a room.

Infocom (in later games, at least) seems inconsistent in this regard.
Multiple direct objects aren't allowed for some verbs (eg: press) but are
allowed for others (eg: turn).

Anybody have any thoughts on this? My initial temptation is to do as Leary
has done with Zero, and prohibit multiple direct objects for everything but
inventory manipulation. Are there any "zinger" cases I'm missing?

Dave Baggett
__
d...@ai.mit.edu Boot up, log in, drop out. MIT AI Lab
ADVENTIONS: We make Kuul text adventures! Email for a catalog of releases.
PO Box 851 Columbia, MD 21044 USA / CIS: 76440,2671 / GEnie: ADVENTIONS

Michael Van Biesbrouck

unread,
Mar 6, 1994, 11:24:44 PM3/6/94
to
In article <2ldjfl...@life.ai.mit.edu>,

David Baggett <d...@ai.mit.edu> wrote:
>An issue that frequently comes up in playtesting is when to allow multiple
>direct objects. The problem is that allowing too much makes some puzzles
>trivial:
>
>-----------------------------------------------------------------------------
>The troll eyes you hungrily. He begins drooling on your leg. "Me hungry,"
>he growls.
>
>>give all to troll
>
>...
>Kewpie doll: the troll, fearing the ancient legends, flees in terror
>from the Kewpie doll.
>...
>-----------------------------------------------------------------------------

The best solution to this sort of problem is:

* >give all to troll
*
* Book: the troll, not being much of a bookworm, decides to eat you.
* Kewpie doll: the troll is not in the habit of accepting strange things
* from food.
* :

If someone starts fiddling with all of the knobs for Zork's Flood
Control Gate, the result is catastrophic. Any situation for which
cycling through objects in your inventory will provide a solution is not
a good puzzle. Planetfall sort of solved this problem by giving you
more objects than you could carry. Trying objects without understanding
why they would be useful might take hours.

Killing the player on a wrong choice (like my example might be construed
to be) is probably excessive, and doesn't stop people from undoing or
restoring until they get it right. If your puzzles require players to
realize that they need an object and then search for it ("The store is
just to big to go wandering about in. Perhaps you should ask one of the
staff members."), then this problem doesn't really occur.

Sometimes you really do want to feed `all' to the dragon. If stuffing
the dragon will keep it from eating you, then good. Another possiblity
is requiring things to be given with some sort of order so that random
guessing will be too slow. When feeding a dragon, maybe

* The dragon looks hungry, and licks her lips when she looks at you.
*
* > feed dragon flamethrower.
*
* The dragon sees through your ruse and blocks the cave entrance with
* her tail. She begins sharpening her teeth.
*
* > feed dragon sandwich
*
* The dragon is on to your ways, and refuses the food. You're lunch.
*
* You have died: restore
*
* > feed dragon sandwich
*
* The Earl of Sandwich is screams in terror as the dragon swallows him
* whole. The dragon is eyeing you warily, and still seems very hungry.
*
* > feed dragon princess
*
* What do you want to feed to the dragon Princess?
*
* > steak
*
* The dragon licks her lips after eating the steak. She's still hungry
* and looks at you expectantly.
*
* > feed flamethrower to dragon
*
* Trustingly, the dragon swallows the flamethrower whole. You barely
* make it to cover before she explodes.

--
Michael Van Biesbrouck, UW CSC Librarian
gopher://descartes.uwaterloo.ca/h0/mathSOC/.csc/.www/.mlvanbie/homepage.html
Spin-doctors are like evil anti-librarians; they're
the Dark Side of the Force. -- Bruce Sterling

David Baggett

unread,
Mar 7, 1994, 11:40:02 AM3/7/94
to
In article <JAMIE.94M...@kauri.vuw.ac.nz>,
Jamieson Norrish <ja...@kauri.vuw.ac.nz> wrote:

>Giving everything to the troll might mean that you get none of it back,
>including that wonderful screwdriver which would mean escape later on in
>the game...

Right, which is why your players will do:

>save "foo"
>give all to troll
>restore "foo"
>give kewpie to troll

OK, from the posts and from email I've gotten, I see two problems with my
original question:

1) Bad example
2) Lack of clarity

So I'll make some things more clear. First, puzzles that involve giving
the correct object to some actor are not inherently bad puzzles (this is a
reasonable, but separate argument, in any case), but the example I gave was
obviously silly. Nervermind the pros and cons of such ridiculous puzzles
-- the point is that there are lots of cases where "all" will allow the
player to dodge the designer's intentions.

If you type "examine all" in the first room of my current game, you get
literally a dozen screenfulls of description. You *do not want* to do
this. Furthermore, I *do not want* this to happen, because it destroys any
feeling of exploration the player would normally get from examing things in
turn. Players have no qualms destroying their own game experience along
the way to solving the puzzles! As the designer I want players to slow
down and actually *read* the text, *feel* what's going on, and *understand*
the point of the work. (This is not the same as saying that I want the
experience to be tedious -- surely there is a happy medium between no
immersion and frustratingly realistic immersion.)

There seems to be an "anything should be allowed" sentiment here lately.
As I've tried to show before, this is in direct conflict with having an
actual plot (in the literary sense). For there to be a real plot, the
designer must occassionally impose certain constraints on the player's
behavior. "All" tends to screw this up: having "all" work with "examine,"
for example, allows players to wander around typing "examine all" in every
location, thereby instantly overturning every rock, looking behind every
mirror -- automatically zeroing in on exactly the important details the
designer wants the player to *discover*.

The lack of realism here isn't bad in and of itself, but the consequence is
that the player is not immersed in the game. "Examine all" destroys the
feeling that you, the player, are actually *there*, and not just solving
little puzzles.

Again, I'm interested in hearing whether anyone can make a case that "all"
should be allowed for something other than inventory mainpulation.

Phil Goetz

unread,
Mar 7, 1994, 2:46:04 PM3/7/94
to
In article <2lfld...@life.ai.mit.edu>, David Baggett <d...@ai.mit.edu> wrote:
>If you type "examine all" in the first room of my current game, you get
>literally a dozen screenfulls of description. You *do not want* to do
>this. Furthermore, I *do not want* this to happen, because it destroys any
>feeling of exploration the player would normally get from examing things in
>turn. Players have no qualms destroying their own game experience along
>the way to solving the puzzles! As the designer I want players to slow
>down and actually *read* the text, *feel* what's going on, and *understand*
>the point of the work. (This is not the same as saying that I want the
>experience to be tedious -- surely there is a happy medium between no
>immersion and frustratingly realistic immersion.)

This is the fascist school of literature, exemplified by educational
stories which intersperse plot with informative stuff that the
kids skim over. If your game is basically puzzle solving, I'm going
to want to solve the puzzles, and anything that slows me down is a
nuisance.

>There seems to be an "anything should be allowed" sentiment here lately.
>As I've tried to show before, this is in direct conflict with having an
>actual plot (in the literary sense). For there to be a real plot, the
>designer must occassionally impose certain constraints on the player's
>behavior. "All" tends to screw this up: having "all" work with "examine,"
>for example, allows players to wander around typing "examine all" in every
>location, thereby instantly overturning every rock, looking behind every
>mirror -- automatically zeroing in on exactly the important details the
>designer wants the player to *discover*.

I have yet to find a game where we are supposed to "discover" things
through reasoning rather than through simply examining everything
we encounter. This is such deeply-ingrained behavior that if you
ever do write a game where I don't have to examine everything,
you should state that at the start of the game.

>The lack of realism here isn't bad in and of itself, but the consequence is
>that the player is not immersed in the game. "Examine all" destroys the
>feeling that you, the player, are actually *there*, and not just solving
>little puzzles.

The only thing "examine all" destroys for me is the tedium of typing
"x door. x cheez door. x stamp. x postcard. read postcard. x bed.
look under bed. look behind bed. move bed. push bed." If that doesn't
destroy the feeling of being there, I don't know what does.

If you _are_ just solving little puzzles (which we are), then
you shouldn't try to cover that up by making it tedious.

>Again, I'm interested in hearing whether anyone can make a case that "all"
>should be allowed for something other than inventory mainpulation.
>
>Dave Baggett

The customer is always right. If I want to use "all", and you can
provide it, then do so.

<<< smiley mode ON >>>

In fact, I think ALL should be used as a verb, so "all rock"
will result in

ask rock: The rock has nothing to say.
eat rock: You would chip a tooth.
examine rock: It's a rock.
feel rock: It has sharp edges.
hit rock: Ouch!
lick rock: It feels rough to your tongue.
weigh rock: It's about a pound.
drop rock: OK.
throw rock: You don't have it.

Then, of course, we would solve the adventure via repeated use of

ALL ALL

;)

Phil go...@cs.buffalo.edu

Chaos is the source of complexity.

Gerry Kevin Wilson

unread,
Mar 7, 1994, 5:56:15 PM3/7/94
to
Hmm, wonder if 'ask <person> about all' works in TADS. I'll have to make
sure it doesn't. It would be awfully unrealistic. All of you sudden you
know everything to ask this NPC you've run across. Sheesh. As for examine
all, you don't want to try that in my game unless you want to page through
a heap of stuff. Most things aren't important, they're just scenery. In any
event, I'll probably turn 'examine all' off after betatesting is done. I
want people to take a little time with my game and enjoy it rather than
barreling through at full speed. I mean, I don't see how you can enjoy
something when you do that. Well, I guess some people just like the puzzles
and don't care about the rest of the game. I wouldn't bother with Avalon if
that's all you care about. You have to do a bit of exploring and talking to
people before you ever get to any of the hard puzzles. It may be considered
fascist, but then, what do you think of games that use GUIs and restrict you
to only a few actions, reducing the puzzles to try everything on everything
else, just as you jokingly mentioned. If you had an 'all' command in the
LucasArts games, you could beat them as fast as you could move the mouse.
So hey, look at the bright side, at least you can try wacky things in text
adventures besides "use hamster with microwave".
--
<~~~~~E~~~G~~~~~~~~~~~HEINLEIN~~~~~~~~~~~DOYLE~~~~~~~~~~~~~SPAM~~|~~~~~~~>
< V R I O Software. We bring words to life! | ~~\ >
< T | /~\ | >
<_WATCH for Avalon in early MAY!____wh...@uclink.berkeley.edu_|_\__/__>

Conrad Wong

unread,
Mar 7, 1994, 8:23:31 PM3/7/94
to
In article <CMB88...@acsu.buffalo.edu> go...@cs.buffalo.edu (Phil Goetz) writes:
>I have yet to find a game where we are supposed to "discover" things
>through reasoning rather than through simply examining everything
>we encounter. This is such deeply-ingrained behavior that if you
>ever do write a game where I don't have to examine everything,
>you should state that at the start of the game.

Infocom's A Mind Forever Voyaging? Granted, the endgame requires some
puzzle-like thought, but nothing unduly sneaky. But I tend to agree,
games centered on puzzles will have players investigating *everything*.

><<< smiley mode ON >>>
>
>In fact, I think ALL should be used as a verb, so "all rock"
>will result in
>
>ask rock: The rock has nothing to say.
>eat rock: You would chip a tooth.
>examine rock: It's a rock.

...


>drop rock: OK.
>throw rock: You don't have it.

Actually, why not? ('gryn)

If a game is going to lay claim to the title 'interactive fiction',
in my opinion, it should make things as convenient as possible for
the player to pass up all the trifling things in favor of the
game experience. This means that it shouldn't be about solving
puzzles, it should be about what happens next in the story and how
the player can affect that.

So all the silly things to find out properties of objects could be
combined into, say, 'search'. If you want to prevent people from
habitually doing 'search' on everything, then you might make search
take more time:

It's ten minutes to midnight.
> examine dresser

A lovely example of modern woodcrafting, it boasts exquisite carvings
machine-engraved into its simulated wood veneer surface, with imitation
bone scrimshaw handles for each of its three drawers. Centuries ago, a
bit of furniture as good looking as this one would have cost a year's
salary for a bourgeois. Now it's Sears $399.95.

It's nine minutes to midnight.
> search dresser

You riffle through each drawers, finding nothing but dainties in the
topmost, blouses and sweaters in the middle, and socks in the lowest.
True to form, the socks don't match.

A peek behind the dresser shows nothing but dust.

You kneel to look under the dresser and discover a tube of lipstick.
It must have rolled underneath some time ago.

It's three minutes to midnight. ...


Obviously if you WANTED to do the open, look behind, look under, etc.
separately, in the interest of saving time if you knew you only had to
do one of those. And similarly, if you did 'search room', the game
might very well do a categorical examination of everything that seemed
promising, prompting you if you wished to continue searching at
ten minute intervals.


My pet peeve about these games is the constant need to put 'X is direction
Y' and 'To the Y, you see X' in the descriptions. A number of MUDs offer
'Obvious exits' which means that you can simply list the exits thusly
available, and use the room description to describe the room itself,
thus allowing the user to see at a glance where she can go. Reading
through descriptions to make the user find the available exits may *SEEM*
to 'immerse the user in the game', but IMHO, it really is another form
of tedium that makes the user look for hinted-at passages. It also
creates strained descriptions that must thoroughly describe all available
exits that would be easily noticed.


-- Lynx

--
__ ___ ___ _/' Name: Conrad "Lynx" Wong
/ \ _/ \----' \-' O`-g Address: 28368 Christopher's Lane
| | / > __/_ / __/_`, _| Los Altos Hills, CA 94022
\__/ \____\`--\____\ ;/' E-mail: ly...@netcom.com

Lynx is "AL" Go B Y++ L++ C++++ T++ A-- H++ S++ V+ F- Q+ P+ B PA+ PL++
(see rec.pets.cats for code explanation or E-mail me and ask)

Jamieson Norrish

unread,
Mar 7, 1994, 2:15:03 PM3/7/94
to
In article <2ldjfl...@life.ai.mit.edu> d...@min.ai.mit.edu (David
Baggett) writes:

An issue that frequently comes up in playtesting is when to allow
multiple direct objects. The problem is that allowing too much
makes some puzzles trivial:

[Puzzle deleted.]

The player need not understand the reason behind the successful
solution here to get it right.

No, the player doesn't, and I don't see that as a bad thing. The
situation you presented seems perfectly reasonable to me. Indeed,
while going through the backpack (or whatever), the character might
well show the doll for a moment, and the troll might run off then. It
should present little difficulty to invent puzzles which cannot be
solved in such a way, thus eliminating the problem. And what better
way is there to find out the connection between the doll and the troll
than stumbling upon it such a fashion - perhaps that event would then
allow the player to make other inferences about the relationship which
could be used in puzzles later on. I don't think chance is something
that should be ruled out of the game; but this is not of course saying
that chance must be on the player's side in order for the game to be
solved.

I think you are quite right to say that confining the use of "all" to
inventory manipulations (although I would define your puzzle example
to be in that category) simply requires the player to use brute force
- so why bother? A simple enough deterrent would be for the designer
to ensure that reasonable, logical consequences of using "all" happen,
which will not always be in the character's best interests. Giving


everything to the troll might mean that you get none of it back,
including that wonderful screwdriver which would mean escape later on
in the game...

Jamie

smeg...@castlebbs.com

unread,
Mar 7, 1994, 11:37:13 PM3/7/94
to

whizzard writes:

>So hey, look at the bright side, at least you can try wacky things in
text
>adventures besides "use hamster with microwave".

Hey! I happen to LIKE Maniac Mansion! ;)

-K. "And I finished it five ways, too" C.
---
"Can't you exist without crapping on people?" -Joel Furr

+----------------------------------------------------------------------+
| THE CASTLE bbs - Los Angeles, CA, USA - 50 Lines! +1 213.953.0040 |
| 30 CD ROMs - 40+ Multi-Player Games Internet: in...@castlebbs.com |
+----------------------------------------------------------------------+

Phil Goetz

unread,
Mar 9, 1994, 10:59:44 PM3/9/94
to

I'm sorry already, Dave! Don't take it personally.
(I'm sure those rumors about your swastika collection are untrue. ;)

>>I have yet to find a game where we are supposed to "discover" things
>>through reasoning rather than through simply examining everything
>>we encounter.
>

>I don't know what you mean by this -- there are plenty of puzzles in
>existing games that you have to solve by reasoning, and not just my
>examination aand object manipulation. How about riddles, where you "say"
>the right answer?

I mean that in any game we have to examine everything, and cannot afford
to leave a wad of paper in the garbage can ignored like we would in
real life. Hence we examine all.

If we're not given the convenience of X ALL, then we shouldn't have to
examine everything. If a wad of paper is important, we should be led
to know that someone may have thrown away a map in that room. Figuring
out that a makeup kit may have something important in it is a valid
puzzle. Discovering something useful about the makeup kit by
examining it is not. I want to back-chain from obstacles, not
forward-chain from objects.

Phil

Mark Phillips

unread,
Mar 10, 1994, 5:40:24 AM3/10/94
to
In article <CMFJv...@acsu.buffalo.edu> go...@cs.buffalo.edu (Phil Goetz) writes:


In article <JAMIE.94M...@kauri.vuw.ac.nz>,
Jamieson Norrish <ja...@kauri.vuw.ac.nz> wrote:

>In article <lynxCMB...@netcom.com> ly...@netcom.com (Conrad Wong)
>writes:


> My pet peeve about these games is the constant need to put 'X is
> direction Y' and 'To the Y, you see X' in the descriptions.
>

>Here I am in complete disagreement. Obvious exit lists are ugly in the
>extreme, and hugely detract from any atmosphere that might be garnered
>by reading a nice description.

At least twice I've been hopelessly stuck and almost given up an
adventure, and on reading the hints found that I had missed an
exit in a room description somewhere.

Phil

I agree!!!!

I recently finished Zork Zero and thought one of its best features was
that it displayed the obvious exits (and the hidden ones once they had
been used). I do ALWAYS read the entire room desription, but I hate
mapping - I managed to do a lot of Zork Zero without contantly
refering to my maps (ie. when I was just trying to get back to
somewhere I had been before).

Also many adventure games do not mention all the exits in the room
descriptions - Trying n,ne,e,se,s,sw,w,nw,u,d everywhere (and then
finding you go somewhere you can't get back from) just is pain.

I find getting completely stuck because I forgot where an exit was, or
forgot to map it :-) very annoying and it doesn't add any enjoyment to
the game at all (for me:-)).

Mark Phillips

Jamieson Norrish

unread,
Mar 10, 1994, 6:49:14 AM3/10/94
to
In article <2lfld...@life.ai.mit.edu> d...@min.ai.mit.edu (David
Baggett) writes:

>Giving everything to the troll might mean that you get none of it
>back, including that wonderful screwdriver which would mean escape
>later on in the game...

Right, which is why your players will do:

>save "foo"
>give all to troll
>restore "foo"
>give kewpie to troll

Whereas if you disallow "all" in this context, your players will do:

>save "foo"
>give <x..z> to troll


>restore "foo"
>give kewpie to troll

Why make them go through the bother of that?

If you type "examine all" in the first room of my current game, you
get literally a dozen screenfulls of description. You *do not
want* to do this. Furthermore, I *do not want* this to happen,
because it destroys any feeling of exploration the player would
normally get from examing things in turn. Players have no qualms
destroying their own game experience along the way to solving the
puzzles!

I think that if players wish to destroy their experience of the game,
they are entitled to, and in any case there is nothing anyone, as a
designer, can do to stop them. You might like to tell prospective
players some things about the game, so that they don't ruin the game
for themselves, but I don't see disallowing all outside of inventory
manipulation as a part of this. Players who want to enjoy the game
will do so, and not abuse the system. I would never type "examine
all", or for that matter "give all".

As the designer I want players to slow down and actually *read* the
text, *feel* what's going on, and *understand* the point of the
work. (This is not the same as saying that I want the experience
to be tedious -- surely there is a happy medium between no
immersion and frustratingly realistic immersion.)

I think it is a useless task to try and force players to do the things
you mention. Allow them to do that if they wish, but that's all you
can do.

There seems to be an "anything should be allowed" sentiment here
lately. As I've tried to show before, this is in direct conflict
with having an actual plot (in the literary sense). For there to
be a real plot, the designer must occassionally impose certain
constraints on the player's behavior. "All" tends to screw this
up: having "all" work with "examine," for example, allows players
to wander around typing "examine all" in every location, thereby
instantly overturning every rock, looking behind every mirror --
automatically zeroing in on exactly the important details the
designer wants the player to *discover*.

I don't think it has to be this way. As I've said, I wouldn't ever
type "examine all", and I'm sure I'm not the only one, particularly if
you made a note to the effect that the intention is for the player to
discover things slowly. Also, I think that it is possible and
preferably to limit "all" in such cases. For example, only objects
which can be sensed by the character at the time of the command should
be examined. Objects behind/inside/whatever other objects should not
be examined in this instance. I'm sure this is extensible to other
areas, and should address at least some of your complaints with the
use of "all".

The lack of realism here isn't bad in and of itself, but the
consequence is that the player is not immersed in the game.
"Examine all" destroys the feeling that you, the player, are
actually *there*, and not just solving little puzzles.

Perhaps. Although if done well, I'm sure that getting all the data at
once would be just the same as having all that data split up by little
commands such as "examine <x>". If the player reads through it all
faithfully, where is the difference?

Jamie

Jamieson Norrish

unread,
Mar 10, 1994, 7:00:21 AM3/10/94
to
In article <lynxCMB...@netcom.com> ly...@netcom.com (Conrad Wong)
writes:

So all the silly things to find out properties of objects could be


combined into, say, 'search'. If you want to prevent people from
habitually doing 'search' on everything, then you might make search
take more time:

Thank you for pointing this out; I think that this would indeed make
players more cautious of "examine all", and the like. And if a
character has plenty of time, then the player can safely do it. I
don't see that many screens of text automatically means that the
player does not read them - unless of course they scroll of the screen.

My pet peeve about these games is the constant need to put 'X is
direction Y' and 'To the Y, you see X' in the descriptions. A
number of MUDs offer 'Obvious exits' which means that you can
simply list the exits thusly available, and use the room
description to describe the room itself, thus allowing the user to
see at a glance where she can go. Reading through descriptions to
make the user find the available exits may *SEEM* to 'immerse the
user in the game', but IMHO, it really is another form of tedium
that makes the user look for hinted-at passages. It also creates
strained descriptions that must thoroughly describe all available
exits that would be easily noticed.

Here I am in complete disagreement. Obvious exit lists are ugly in the


extreme, and hugely detract from any atmosphere that might be garnered

by reading a nice description. I personally don't mind rereading a
description in order to find an exit - it simply helps to clarify in
my mind what the room "looks" like. Once I've created a picture in my
head, I don't much need to read the description again unless something
changes. Nor would I need to look for exits.

Jamie

Jamieson Norrish

unread,
Mar 10, 1994, 7:05:54 AM3/10/94
to
In article <2liepc...@life.ai.mit.edu> d...@min.ai.mit.edu (David
Baggett) writes:

I don't think you *need* "all" to solve any of the puzzles. You
have to figure out *which* is the right dial to turn, *what* to
give the troll, etc. Simply because brute force works doesn't mean
that it's the right solution. As a game designer I would prefer to
eliminate brute force approaches entirely, but this isn't so easy
in some cases.

And simply because the brute force option exists does not mean that
players will use that in every case over other methods. However, in
many cases the use of "all" would be very useful.

Jamie

Jamieson Norrish

unread,
Mar 10, 1994, 7:10:08 AM3/10/94
to
In article <2lgbef$1...@agate.berkeley.edu>

whiz...@uclink.berkeley.edu (Gerry Kevin Wilson) writes:

I want people to take a little time with my game and enjoy it
rather than barreling through at full speed. I mean, I don't see
how you can enjoy something when you do that.

I think that the point that is being missed here is that using
"examine all" does not necessarily mean that the player goes through
the resulting information "at full speed". "All" can be used as a
convenience device, to save the player typing in all the different
object names after "examine". It in no way forces the player to skip
large chunks of prose.

Jamie

Gerry Kevin Wilson

unread,
Mar 10, 1994, 5:37:26 PM3/10/94
to
Well, I for one got the distinct impression from you that you didn't care to
read the text, and instead preferred to skim past to get to the puzzles.
Sorry if that's not the case, but that's what it sounded like to me. And no,
examine x, examine y is not just increased tedium. There is often a method
to madness. 1) It simulates real life, in that it takes a certain amount of
time to examine an object thoroughly. 2) examine all can often turn up items
that were meant to be hidden inside another object or what have you. You
might think that this is a good thing, but what about a mystery game, where
clues are often hidden from the player, forcing him to read carefully and pay
attention, the same as a real detective. examine all seems to me to be an
overly simplified way to solve things that are dependent on paying attention.
I guess that one person's tedium is another person's hobby. I like to look
through the various items and take my time with everything. The scenery is
usually one of my favorite parts of any text adventure, and I enjoy the
challenge of watching for important things. How about this, those who care
one way or another on the subject of all send me e-mail on examine all, etc,
telling me whether or not you support it, and your reasons for this. I'll
bend to the will of the customers, and Avalon will follow the majority's
opinion on the matter. I'll post the results two weeks from this Sat. How's
that grab you?

Jon Drukman

unread,
Mar 10, 1994, 7:00:10 PM3/10/94
to
ja...@kauri.vuw.ac.nz (Jamieson Norrish) writes:
>By "more effort", you presumably mean forcing the player to type,
>"examine x", "examine y", and so on, rather than just typing "examine
>all" and reading all the descriptions? That sounds very much like more
>inconvenience and tedium to me.

i would never allow "examine all" in my game because examining one
object might lead you to examine something else contained within, and
i don't WANT the player to see all the hidden items in one go. i
structure things in layers. looking at everything in the top layer in
one pass might be useful to save some typing, but the amount of extra
typing involved is, i feel, negligible.

Jon Drukman jdrukman%dls...@oracle.com
-------------------------------------------------------------------------------
This calls for a very special blend of psychology and extreme violence.

Darin Johnson

unread,
Mar 10, 1994, 9:10:14 PM3/10/94
to
> I do ALWAYS read the entire room desription, but I hate
> mapping

Same here. And I HATE rereading it over and over again to find
the exits. Ie, I want to get from A to B, and a list of obvious
exits in the intermediate rooms makes doing this very fast, and
alerts me quicker when I've taken a wrong turn.

--
Darin Johnson
djoh...@ucsd.edu
"Particle Man, Particle Man, doing the things a particle can"

Michael Booth

unread,
Mar 11, 1994, 12:17:46 PM3/11/94
to
Conrad Wong (ly...@netcom.com) wrote:
: So all the silly things to find out properties of objects could be

: combined into, say, 'search'. If you want to prevent people from
: habitually doing 'search' on everything, then you might make search
: take more time:

: It's ten minutes to midnight.
: > examine dresser

: A lovely example of modern woodcrafting, it boasts exquisite carvings
: machine-engraved into its simulated wood veneer surface, with imitation
: bone scrimshaw handles for each of its three drawers. Centuries ago, a
: bit of furniture as good looking as this one would have cost a year's
: salary for a bourgeois. Now it's Sears $399.95.

: It's nine minutes to midnight.
: > search dresser

: You riffle through each drawers, finding nothing but dainties in the
: topmost, blouses and sweaters in the middle, and socks in the lowest.
: True to form, the socks don't match.

: A peek behind the dresser shows nothing but dust.

: You kneel to look under the dresser and discover a tube of lipstick.
: It must have rolled underneath some time ago.

: It's three minutes to midnight. ...

: Obviously if you WANTED to do the open, look behind, look under, etc.
: separately, in the interest of saving time if you knew you only had to
: do one of those. And similarly, if you did 'search room', the game
: might very well do a categorical examination of everything that seemed
: promising, prompting you if you wished to continue searching at
: ten minute intervals.

I tend to agree with this auto-searching, thinking of it as a kind of
high-level command to your alter ego, assuming that the results of
the command were well documented so the player would know what to expect.

A similar line of reasoning extends to the notion of how combat is handled
in IF. That is, whether to allow 'kill Moriarty' and let the system
fight for you, or have the user specify each strike.

However, the area of IF in which I am interested is network multi-user
systems (along the line of MUDs). The notion of time becomes a problem,
in that the system is 'real-time', since you are not the only 'player'.
Hence, typing 'search all' could result in 10 minutes of real time
waiting, in order to simulate the action, with descriptions being output
every now and then, as the player's character 'discovered' them.

This is not necessarily a bad thing, but I wanted to point it out.

Mike

--
| Michael S. Booth (bo...@cs.uiowa.edu) |
| Computer Science Graduate Student, Univ. of Iowa, Iowa City, IA |
| (An Engineer pretending to be a Computer Scientist) |

Jamieson Norrish

unread,
Mar 11, 1994, 6:03:19 AM3/11/94
to
In article <2ljfm0$3...@news.u.washington.edu> scy...@u.washington.edu
(The Grim Reaper) writes:

Ug. This is like going through an art gallery by ignoring the
actual art and just reading the little cards underneath each
painting ;P. I want people to read the text I write. I'll try and
keep it short (or fairly so; Maybe putting long bits in footnotes),
but I want you to read what I do write! It has either relavance to
the point of the story, or is their to set the scene. If you skim
the text, you're missing something important.

Mini-argh. As I'm sure I've tried to point out before, there are
advantages to using "examine all" which are entirely unrelated to
being able to skim pages of text. Using "examine all" does *not* mean
that the output is skimmed, and not read thoroughly. Why is everyone
assuming that it does?

We're working on it, really. It doesn't have to be just little
puzzles, however. The "fiction" part of interactive fiction is
being worked on. IMHO, all doesn't just make it less tedious, it's
like playing outside the rules of the game. With "get" or "put" or
whatever, I can see using all is helpful. That's just pushing
objects around. But looking at text, looking/trying to figure out
things, should force the player to put a bit more effort into
things.

By "more effort", you presumably mean forcing the player to type,
"examine x", "examine y", and so on, rather than just typing "examine
all" and reading all the descriptions? That sounds very much like more
inconvenience and tedium to me.

Jamie

David Baggett

unread,
Mar 12, 1994, 1:09:21 PM3/12/94
to
In article <JAMIE.94M...@kauri.vuw.ac.nz>,
Jamieson Norrish <ja...@kauri.vuw.ac.nz> wrote:

>Using "examine all" does *not* mean that the output is skimmed, and not
>read thoroughly. Why is everyone assuming that it does?

I guess another (related) problem I have with it is that is removes the
hierachical exploration of the scene -- it flattens the whole room into one
big hunk of text. Normally you wouldn't look at the type ball on the
Selectric typewriter until you'd looked at the typewriter itself, but
by doing "examine all" (as it's typically implemented) you'd see:

-------------------------------------------------------------------------------
typewriter: It's a souped-up Selectric -- a star in its day, but now a mere
dinosaur. It has one of those funny-looking "type balls" in it.
type ball: It's just an ordinary type ball. The letter Q seems a bit worn.
letter Q: It's so worn from overuse that it now looks like an O.
-------------------------------------------------------------------------------

If you have this level of detail (simply to make the game world seem more
real, not for puzzle reasons), the player will be totally overwhelmed by
"examine all", and will feel more like he's playing a computer simulation
that can mechnically spew out reams of text than like he's actually in the
game world looking over things. If someone wants to take the time to
examine everything down to this level (the letter Q on the type ball), it
really seems like it should take them longer than a quarter second to case
the whole area.

It also seems to me that "examine all" ruins any dramatic value that
exploring hierarchically may have. I try to think about my descriptions in
terms of camera positions and pans. In movies, the camera explores things
along a pan, and there is considerable dramatic value to a good pan. In
the example above, where the player explores the typewriter, the camera is
zooming in -- first the whole typewriter, then the type ball, then the
letter Q. When you "examine all" you lose the panning or zooming effect
envisioned by the designer.

>By "more effort", you presumably mean forcing the player to type, "examine
>x", "examine y", and so on, rather than just typing "examine all" and
>reading all the descriptions? That sounds very much like more inconvenience
>and tedium to me.

If someone finds examining things in turn and reading the descriptions
tedious, what the heck are they playing a text adventure for? Anyway, the
fact is that we're not talking about 100 objects per room here -- most
games have at most a dozen main objects at in a location.

We've seen a desire on the part of some r.a.i-f readers to shift the focus
of IF away from the "every detail is important for a puzzle, and the
purpose of the game is to solve puzzles" mode to a more balanced approach.
Doing this changes the effect of commands like "examine all" from speeding
up an otherwise tedious task to mechanically bombarding the player with
unnecessary details. The letter Q on the Selectric type ball, friends, is
probably not important. :) (And, I should point out, a game that makes it
important without calling attention to the typewriter in the top level of
description in the game is, IMHO, a badly designed game.)

In a game where every detail is important, "examine all" is, I admit, a
good thing. But in a work where the focus is not entirely on solving
puzzles, it can destroy the author's intentions -- which, if the game has
any literary merit, may be more important than the whims of a subset of the
readers/players. (OK, so I *am* a fascist!) :)

David Baggett

unread,
Mar 12, 1994, 1:57:27 PM3/12/94
to
In article <CMFKF...@acsu.buffalo.edu>, Phil Goetz <go...@cs.buffalo.edu> wrote:

>I mean that in any game we have to examine everything, and cannot afford
>to leave a wad of paper in the garbage can ignored like we would in
>real life. Hence we examine all.

In real life we aren't detectives, Valley Warriors, vampires, and so forth.
If you're a detective searching through a suspected murderer's room, you
might very well have to pay attention to that piece of paper in the trash
can. Real life is boring -- we don't want either literature or games to be
like real life. I realize this may sound contradictory for me, given that
I've appealed to arguments of realism elsewhere. The difference is this:
realism is important insofar as it makes the player feel immersed. Beyond
that, it detracts, because real life is generally not very fun or
interesting. (What a maudlin thought!)

Perhaps this difference of opinion over "examine all" is related to where
we each draw the line between realism as a benefit versus realism as a
source of tedium.

>If we're not given the convenience of X ALL, then we shouldn't have to
>examine everything. If a wad of paper is important, we should be led
>to know that someone may have thrown away a map in that room.

Perhaps Graham should add this to his players' bill of rights:

The player has the right to know that something is important
from the top level of description. E.g., if no particular
attention is called to an item in a room description, the
designer should not assume that the player will examine the item.

Jamieson Norrish

unread,
Mar 12, 1994, 7:31:30 AM3/12/94
to
In article <jdrukman.763344010@dlsun87> jdrukman%dls...@oracle.com
(Jon Drukman) writes:

i would never allow "examine all" in my game because examining one
object might lead you to examine something else contained within,
and i don't WANT the player to see all the hidden items in one go.
i structure things in layers. looking at everything in the top
layer in one pass might be useful to save some typing, but the
amount of extra typing involved is, i feel, negligible.

If the players realised that you did things in layers, then maybe they
wouldn't use "examine all". Or perhaps they would. Surely the choice
should be up to them?

As for extra typing, one thing that I loathe is typing "examine x",
then having to go "look" in order to reread the description and find
out what the next object I want to examine is. Extra typing and
tedium? I think so.

Jamie

Conrad Wong

unread,
Mar 12, 1994, 8:11:10 PM3/12/94
to
In article <2lq93q$i...@nexus.uiowa.edu> bo...@grant.cs.uiowa.edu (Michael Booth) writes:
>I tend to agree with this auto-searching, thinking of it as a kind of
>high-level command to your alter ego, assuming that the results of
>the command were well documented so the player would know what to expect.
> ...

>
>However, the area of IF in which I am interested is network multi-user
>systems (along the line of MUDs). The notion of time becomes a problem,
>in that the system is 'real-time', since you are not the only 'player'.
>Hence, typing 'search all' could result in 10 minutes of real time
>waiting, in order to simulate the action, with descriptions being output
>every now and then, as the player's character 'discovered' them.
>
>This is not necessarily a bad thing, but I wanted to point it out.

Bear in mind that in MUDs, one doesn't necessarily want users to be able to
macro "kill troll with sword" "kill troll with sword" etc. within mere
seconds. So they might well wind up waiting to get a reply, to make things
more "fair" for others, or the program might prevent their doing another
kill action until a certain amount of time had passed. Annoying, perhaps,
but more 'fair' to people.

Similarly, a 'search all' might take some time, and you might be described as
searching to others who happened to be in the room. But you could perform
'non-preemptive' commands such as talking and inventory manipulation while
your character went from one place to another looking at things.

> search room

Peregrine casts his eagle-eyed look about the room for valuables, beginning
with the most promising, the ironworked brazier that sat at the center of
the runes, a flame flickering within. It billowed forth an eerie green-lit
smoke.

>
[looking in brazier]
Coals burn orange within the brazier's cup. But surely that could not be
the cause of the green lights...

[looking under brazier]
Complicated runes line the floor. The brazier itself rests squarely upon
the 'anchor' sigil.

[looking behind brazier]
There's nothing particularly interesting back there.

Sigoro, High Priest of Fractals, bursts into the room!

Peregrine is searching the brazier.

Done with the brazier, at least for the moment, Peregrine begins looting the
jars that stand racked against the far wall of the runes.

Sigoro protests, "Cease, desist, abandon your foolery!"

>say Make my day, pateface.

Peregrine doth utter, "Make my day, pateface."

Sigoro begins casting a complex spell, drawing forth slender bone wands!

>kill sigoro
[with the Elvish longsword of great antiquity]

do you want to stop searching?
>

...

Jamieson Norrish

unread,
Mar 13, 1994, 7:38:18 AM3/13/94
to
In article <2lo7f6$t...@agate.berkeley.edu>

whiz...@uclink.berkeley.edu (Gerry Kevin Wilson) writes:

Well, I for one got the distinct impression from you that you
didn't care to read the text, and instead preferred to skim past to
get to the puzzles. Sorry if that's not the case, but that's what
it sounded like to me.

I am assuming that I am the "you" mentioned here; sorry if that's what
it sounded like - it certainly isn't what I intended.

And no, examine x, examine y is not just increased tedium. There
is often a method to madness. 1) It simulates real life, in that
it takes a certain amount of time to examine an object thoroughly.

Except that the passing of time can be simulated in much easier ways
than simply by forcing the player to type in extra commands.

2) examine all can often turn up items that were meant to be hidden
inside another object or what have you. You might think that this
is a good thing, but what about a mystery game, where clues are
often hidden from the player, forcing him to read carefully and pay
attention, the same as a real detective. examine all seems to me
to be an overly simplified way to solve things that are dependent
on paying attention.

Again, I don't think this is a valid criticism. It should be no
problem to make "examine all" act on only those objects which are
obvious from what is given in the room description. That way, hidden
items would not automatically be looked at.

What other arguments are there against games having the option of
"examine all"?

I guess that one person's tedium is another person's hobby. I like
to look through the various items and take my time with everything.
The scenery is usually one of my favorite parts of any text
adventure, and I enjoy the challenge of watching for important
things.

I agree with you; however, I don't think that "examine all" would stop
anyone from doing this. It is, after all, only an option, and if a
player does not want to use it, then so be it. And if a player does
wish to use it, I don't think that automatically means that that
player won't take the time to look at everything and read it
thoroughly.

Jamie

David Baggett

unread,
Mar 13, 1994, 11:56:08 PM3/13/94
to
Jamieson Norrish <ja...@kauri.kauri> writes:

>However, as I have pointed out before, you can write "examine all" to
>not delve into hierarchical descriptions, and simply give the first
>level of description for all the obvious items in the room.

I'll agree that this is fine, but I've been tacitly omitting this option
from the discussion because it's a colossal pain to implement -- you have
to tag every single decoration with its level in the description hierarchy.
(The description hierarchy is not the same as the containment hierarchy --
the "camera" may see the computer disk drive before it sees the brand
label, but both are on the computer at equal depth.)

>In which case attach time values for examining. It shouldn't be so
>hard to advance time by more than passes in real time.

Yes, but what if there are processes going on in the background ("daemons
running" in TADS parlance) while your examine is running? Suppose you're
holding a bomb that goes off in five turns, but warns you two turns before
exploding that it "seems to be making noise" so you can toss it away. Then
you "examine all" and three turns in you're told that the bomb's making
noise, but are helpless because you can't interrupt "examine all". Two
more turns and you're killed. Less extreme cases are problematic too.

Then we also have the problem of interrupting "examine all," which, in TADS
at least, wouldn't be trivial. Making commands take more than one turn has
repercussions. I can certainly imagine a system handling such things well,
but doing so is far from trivial.

It would, on the other hand, be easy to make "examine all" apply only to
takeable items. But somehow I think this is a bit pointless.

Michael Booth

unread,
Mar 14, 1994, 2:29:09 PM3/14/94
to
Conrad Wong (ly...@netcom.com) wrote:
: I wrote:
: >I tend to agree with this auto-searching, thinking of it as a kind of

: >high-level command to your alter ego, assuming that the results of
: >the command were well documented so the player would know what to expect.
: > ...
: >
: >However, the area of IF in which I am interested is network multi-user
: >systems (along the line of MUDs). The notion of time becomes a problem,
: >in that the system is 'real-time', since you are not the only 'player'.
: >Hence, typing 'search all' could result in 10 minutes of real time
: >waiting, in order to simulate the action, with descriptions being output
: >every now and then, as the player's character 'discovered' them.
: >
: >This is not necessarily a bad thing, but I wanted to point it out.

: Bear in mind that in MUDs, one doesn't necessarily want users to be able to
: macro "kill troll with sword" "kill troll with sword" etc. within mere
: seconds. So they might well wind up waiting to get a reply, to make things
: more "fair" for others, or the program might prevent their doing another
: kill action until a certain amount of time had passed. Annoying, perhaps,
: but more 'fair' to people.

I don't believe I said it would take seconds. In fact, my whole point
was to mention that, in a MUD-type system, these search/kill/meta commands
would take _real_ time, not just jumping the game time ahead. So, of
course the user would have to wait some amount of time, proportional
to the amount of time the action would 'really' take.

: Similarly, a 'search all' might take some time,

See above.

: and you might be described as


: searching to others who happened to be in the room.

Yes, a valuable thing.

[...rest of response deleted...]

Just wanted to clarify my original point (although it seemed pretty
clear to me).

Jamieson Norrish

unread,
Mar 14, 1994, 7:09:56 AM3/14/94
to
In article <2lt0gh...@life.ai.mit.edu> d...@min.ai.mit.edu (David
Baggett) writes:

I guess another (related) problem I have with it is that is removes
the hierachical exploration of the scene -- it flattens the whole
room into one big hunk of text. Normally you wouldn't look at the
type ball on the Selectric typewriter until you'd looked at the
typewriter itself, but by doing "examine all" (as it's typically
implemented) you'd see:

[Example deleted.]

However, as I have pointed out before, you can write "examine all" to

not delve into heirarchical descriptions, and simply give the first


level of description for all the obvious items in the room.

If you have this level of detail (simply to make the game world


seem more real, not for puzzle reasons), the player will be totally
overwhelmed by "examine all", and will feel more like he's playing
a computer simulation that can mechnically spew out reams of text
than like he's actually in the game world looking over things. If
someone wants to take the time to examine everything down to this
level (the letter Q on the type ball), it really seems like it
should take them longer than a quarter second to case the whole
area.

In which case attach time values for examining. It shouldn't be so


hard to advance time by more than passes in real time.

It also seems to me that "examine all" ruins any dramatic value


that exploring hierarchically may have. I try to think about my
descriptions in terms of camera positions and pans. In movies, the
camera explores things along a pan, and there is considerable
dramatic value to a good pan. In the example above, where the
player explores the typewriter, the camera is zooming in -- first
the whole typewriter, then the type ball, then the letter Q. When
you "examine all" you lose the panning or zooming effect envisioned
by the designer.

Except that this is the same point you raised above, and which I
answer in the same way. Write "examine all" to "pan" across the first
level of descriptions, and not delve heirarchically.

If someone finds examining things in turn and reading the
descriptions tedious, what the heck are they playing a text
adventure for? Anyway, the fact is that we're not talking about
100 objects per room here -- most games have at most a dozen main
objects at in a location.

It is not the reading of descriptions which is tedious, at least not
the first few times. What is tedious is having to check and recheck a
single description in order to know which noun to put after "examine",
for however many objects there are. This is the tedium which "examine
all" solves.

Jamie

Oliver Rothe

unread,
Mar 17, 1994, 4:18:12 AM3/17/94
to
Jason Noble (jno...@laurel.ocs.mq.edu.au) wrote:
> How might this be implemented? If anyone else has done it already (you must
> be out there) I'd love to hear from you. I'm thinking about doing it one of
> two ways (or maybe both combined):

> 1) the game has a way of recognising happenings that the player will almost
> certainly be interested in responding to, eg. being attacked, having
> something stolen, the entry of someone obviously hostile, becoming hungry /
> thirsty, bomb fuse mesages and the like. Each cycle of a "wait until
> midnight"-type command, a check is made to see if anything like this has
> occurred. If so, the waiting (or searching or whatever) is interrupted and
> the player can enter a new command.

> AND / OR

> 2) whenever the game is cycling through player turns for something like
> "wait until midnight", the computer checks each cycle to see if the player
> has pressed a key. The player is notified in the documentation that
> [...]

In a couple of Infocom games ("Suspect" comes to my mind), a third approach
is implemented:
Whenever an important event happens, the player is asked:

"Do you want to go on with <whatever you are doing>? (Y/N)"

This is the best approach because:
a) it's from Infocom ;-)
b) if you deliberately want to go on, you can do that easily (just type
Y<CR>)
c) you can't miss an opportunity to interrupt what you're doing
(your reaction time should not be important for solving an Infocom game;
"Borderzone" being the only exception).

Oliver

-------------------------------------------------------------------
Oliver Rothe is...@ztivax.zfe.siemens.de
also reachable via: 10026...@compuserve.com


David Michael Tuller

unread,
Mar 27, 1994, 2:06:43 PM3/27/94
to
In article <2milnh$q...@sunb.ocs.mq.edu.au>, jno...@laurel.ocs.mq.edu.au (Jason Noble) writes:
|> My original problem was this:
|> what if the player types "wait until midnight"? Let's say that one game
|> step represents a minute of game time, and the player types this at 6:00pm
|> game time. Now, that would mean 360 steps of game time. Obviously, no-one
|> wants to press enter 360 times. Now, assuming the game really needed to
|> simulate the passing of time accurately (eg. there was a sunrise-sunset
|> daemon, and daemons for character actions that might do things like `at
|> 11:30 pm Bob arises from the dead and leaves his crypt to go to the bank')
|> how can we best allow the player to wait?

I have been working on dealing with this problem for the mystery.t file I
hope to put in the if-archive soon. If the player types "wait until 12:00",
he/she will not be interrupted unless something important happened (what
"important" means is left to the programmer). At that point, you would be asked
"Do you want to stop waiting? <Y/N>". I also allow the program to force you to
stop waiting with the command intsay(nil) [GKW - It's not in the copy you have].
This could be useful if, say, you are captured by aliens who proceed to give you
something that makes you sleep. I don't think people would like it if, after you
wake up, you are asked if you want to wait just because you said to wait before
you fell asleep. In case I'm not clear..

> WAIT UNTIL 12:00
Time passes...

Suddenly, you are beamed aboard an alien spacecraft which resembles a giant
ham sandwich. The aliens then place you on what appears to be a very high-tech
operating table and inject you with some kind of drug. It seems to be a sedative
of some sort. You try to fight off sleep, but to no avail. The drug kicks in
and you are transported to la-la-land.

You wake up some time later in a cell aboard the spaceship.

Do you want to stop waiting? <Y/N> [insert screams here]

|> Maybe Phil would argue that if the
|> player has to wait 6 hours it's a bad game in the first place, and there's a
|> strong case for that view. But I (foolishly) envision a game where the
|> action takes place in short bursts (like in many novels) with periods of
|> inaction between. If the game dynamics were tuned just so, and there were
|> sophisticated methods of waiting, it might make for an interesting
|> experience and not just a frustrating one. Such a game would allow many
|> more elements to come into play, like you can only go to the shop between 9
|> and 5, or the politician you're supposed to assassinate only goes to the
|> seedy gay bar on Friday nights, etc.

Another possibility is that you may want to give the player some time when they
don't have to do anything except figure out how that weird-looking machine
down in the basement works.

|> ---------------------------------------------------------------------------
|> Jason Noble |
|> National Centre for HIV Social Research | jno...@bunyip.bhs.mq.edu.au
|> Macquarie University, Sydney, Australia |
|> ---------------------------------------------------------------------------

David M. Tuller
tul...@rpi.edu

Reply all
Reply to author
Forward
0 new messages