Help me with my CGDC Presentation?
Bill
At the next CGDC in May I'll be giving a talk titled:
"The Interface Is The Game" (Yeah, I did this in '94)
Which will cover the role of the interface in game design, history of game
interfaces, video game vs. computer interfaces, some stuff on Adventure
Games etc... blah, blah, blah ...
See http://www.cgdc.com/cgi-bin/main?x-a=v&x-id=357 for details.
BUT I figured that YOU might want me to cover some issues that are important
to YOUR game design efforts.
So let me know what you think... and what I should cover.
Bill Volk
--
``Perfection (in design) is achieved not when there is nothing more to add,
but rather when there is nothing more to take away.'' Antoine de
Saint-Exupery
Ola Sverre Bauge
Bill wrote in message <6cveqp$8ks$1...@ha2.rdc1.sdca.home.com>...
>At the next CGDC in May I'll be giving a talk titled:
What exactly is CGDC? Computer Game Developers' Con or something like
that? Cool.
>"The Interface Is The Game" (Yeah, I did this in '94)
>
>Which will cover the role of the interface in game design, history of
>game interfaces, video game vs. computer interfaces, some stuff
>on Adventure Games etc... blah, blah, blah ...
>
>BUT I figured that YOU might want me to cover some issues that are
>important to YOUR game design efforts.
>
>So let me know what you think... and what I should cover.
>
>Bill Volk
For the record, you asked. You've only got yourself to blame.
<prepares to launch two-megabyte rant on interfaces in adventure games>
"The Interface Is The Game". <smacks tongue> I like that. Consider,
for instance, the current graphical adventure vs. the current text
adventure. The sort of ideas you can express in both are linked
intimately with the interface. Most obvious is perhaps the way most
graphical adventures contain all possible actions within a set of icons
which is growing smaller by the processor cycle, sometimes being only a
cursor with some permutations of right and left buttons and single and
double clicking creating all possible input.
This has a tendency to create adventures where everything is heavily
object oriented. (not in the programming sense, and paleez let's not
start the OO thread again) Let me bring on an example from a game I
just played, Curse of Monkey Island; when you gain a book on
ventriloquism, you can play tricks on various characters with it as well
as doing the "useful" thing that advances the plot.
This is carried out by using the ventriloqution book on the object you
want to throw your voice at. When you lose the book later on, you are
unable to do any further ventriloqution. However, in a text adventure
it could have been carried out by typing "throw voice at object" or
"ventriloqute [1] object". In fact, I do know one text adventure has
featured ventriloquism, "Phlegm" by "Adjacent Drooler" from the 96
compo.
And that game illustrates another point on how interfaces shape games;
it features Leo, a lemming which the main character carries around in
his breast pocket. He's the one who knows how to ventriloqute, and at
the point where it's useful in the game, you have to ask him to do so,
in the form of "Leo, throw your voice at ...". This never occured to
me, probably for two reasons: I didn't think of using ventriloqution to
solve the problem, and I didn't know the expression "to throw your voice
at something" for ventriloqution, as I'm not a native English player.
In Curse of Monkey Island, I was well aware of the possibility of
ventriloqution, because I only had to drag'n'drop the ventriloquism book
on objects in order to do so, and since it often elicited an amusing
sequence, I did so frequently. Although in the situation where the
ventriloqution came in useful I definitely think I would have thought of
it if all the "useless" responses hadn't been there, it did speed up the
association process a bit. And it was fun, going around spoofing
everyone.
And this, ladies and gentlemen, demonstrates how a point&click interface
can make a game easier and more intuitive, and it demonstrates that all
adventure games should ideally reward ideas that are not strictly
necessary to the advancement of the plot, but which are logical (by the
game's own logic at least) or fun. And of course that it's a virtue to
include synonyms to an absurd degree.
You could argue that text adventures aren't *that* much better than
point&click ones in the simplicity department; once you figure out the
Inform library verbs and try them one after one on every object you can
find, where exactly is the difference from trying every icon and
inventory object on every other object? <- I think this point was made
some time earlier here in raif, sorry for ripping you off, whoever it
was.
Another thing about the Monkey Island games is the masterful way in
which they sometimes disrupt the regular working of the interface; the
sequence in Monkey 1 for instance, where Guybrush goes behind a wall the
player can't see behind and the action line begins to display all sorts
of insane commands like "hypnotize quarreling rhinoceros", to great
amusement and annoyance to the player (for not being able to see behind
the wall).
Or the effect that recurs all throughout the Monkey Island series: In
conversations, you choose lines from a collection at the bottom of the
screen. Usually, Guybrush speaks the line you choose. In most cases
that is, but not all: For instance in Monkey 2, when he is captured by
LeChuck and asked if he has anything to say for himself, you may choose
several witty replies, but all Guybrush can bring himself to say is
"(gulp)". Great use of the interface to surprise the player, not to
mention simulating the way you sometimes just want to say something but
can't quite bring yourself to saying it.
And video game interfaces... well, most of them have been a fraggin'
nightmare since consoles have typically only carried a basic joypad with
fire, select and start buttons in addition to the four directional ones.
But the interesting thing is that the interfaces of computer games have
moved towards the simplicity of that of the consoles, with everything
being covered by wiggling the mouse and clicking. What does that say of
us, the players?
Maybe I should construct a morale for IF authors from this... <begins
wagging moral index finger> OK, maybe we should all look to create text
adventures where abstract concepts are an integral part, and not just as
objects you can pick up either. (though that's fun too, as in Den's
poetic transcript back when Nelson announced his poetry book)
Geez, that's two rants on Monkey 3 in two days... (the other was in
comp.sys.ibm.pc.games.adventure) I really should get a life.
Ola Sverre Bauge
o...@bu.telia.no
http://w1.2327.telia.com/~u232700165/
My news server cannot be trusted; please CC me if you reply.
[1] I don't know if "ventriloqute" is a valid verb and I don't really
care. It sounds cool and to hell with the BBC English Dictionary.
He never promised me a new life
He never gave me second thoughts
Now they cut me up with their knives
And I'm bleeding for their cause
-Disco Judas, "Some Day"
David Brain
In article <34f49...@d2o201.telia.com>, o...@bu.telia.no (Ola Sverre
Bauge) wrote:
<excellent comments on Monkey Island snipped>
> And video game interfaces... well, most of them have been a fraggin'
> nightmare since consoles have typically only carried a basic joypad with
> fire, select and start buttons in addition to the four directional ones.
> But the interesting thing is that the interfaces of computer games have
> moved towards the simplicity of that of the consoles, with everything
> being covered by wiggling the mouse and clicking. What does that say of
> us, the players?
Actually, I think it tells us that video game companies can make a lot
more money by producing a product that can be sold across PC *and*
consoles than by producing a single platform product, rather than saying
anything about the players (the rants on newsgroups about Riven ought to
teach you that - a game with no discernable interface at all but some very
complex puzzles even so.)
It will be interesting to see how well "Starship Titanic" does,
considering that the text parser will presumably make it unworkable on
consoles (unless the system is going to use some form of icon sentence
contruction...)
I agree with your comments about the smart tricks that can be done with a
multiple-choice conversation system (the one used in the DiscWorld games
was quite clever as well), but sometimes you want to break the rules and
this sort of system doesn't know how to cope with that, whereas (in my
dreams) I had been expecting text games to have convincing NPCs by now.
David Brain
London, UK
Stephen Tjasink
>At the next CGDC in May I'll be giving a talk titled:
>"The Interface Is The Game" (Yeah, I did this in '94)
>Which will cover the role of the interface in game design, history of game
>interfaces, video game vs. computer interfaces, some stuff on Adventure
>Games etc... blah, blah, blah ...
>See http://www.cgdc.com/cgi-bin/main?x-a=v&x-id=357 for details.
>BUT I figured that YOU might want me to cover some issues that are important
>to YOUR game design efforts.
>So let me know what you think... and what I should cover.
This is something from playing games, not from designing them...
I've always thought of graphical adventures (in the Sierra mould, not
like Magnetic scrolls, etc) as totally different to text adventures.
Last year I bought Myst from a bargain-bin type sale because I
wanted to see the pretty pictures that people had been talking about
when it had come out. It took me a little while to get into it but
when I did it reminded me uncannily of bits of the Zork trilogy. It
was probably a combination of the deserted once-great landscape and
the way some of the puzzles were constructed.
As a result, I played it non-stop until I'd finished it. Since the
game reminded me of other games with raaaather different interfaces,
I'd venture that it's not entirely true that the interface makes
the game although that's what I've always thought in the past. Maybe
it's just an exception to the rule though, as it's the only graphical
adventure that I can remember having that effect on me.
Stephen
--
/--------- Stephen Tjasink aka stja...@cs.uct.ac.za ----------- Vermillion -\
| University of Cape Town, Cape Town, South Africa -==UDHISS==- |
| For Space Quest fans: http://www.cs.uct.ac.za/~stjasink/sq/ -==UDIC==- |
\ For Bubble Bobble fans: http://www.cs.uct.ac.za/~stjasink/bb/ ---- \/ -----/
Francis Irving
>[1] I don't know if "ventriloqute" is a valid verb and I don't really
>care. It sounds cool and to hell with the BBC English Dictionary.
And harrah for Esperanto!
Francis.
Home: fra...@pobox.co.uk Web: www.meta.demon.co.uk
Bill
Very good. The "hidden verb" nature of text adventures lends itself to a
better suspension of disbelief than the graphical interfaces. You don't
know what verbs are possible in what locations.
Thanks,
Bill
Ola Sverre Bauge wrote in message <34f49...@d2o201.telia.com>...
Andrew Plotkin
Bill (bv...@inetworld.net) wrote:
> Very good. The "hidden verb" nature of text adventures lends itself to a
> better suspension of disbelief than the graphical interfaces. You don't
> know what verbs are possible in what locations.
Except that one ideal of text games (not universal, but something I
strive for) is to make it immediately obvious what actions are possible
for any particular object.
You're right that there's a quality of open-endedness in text games, of
not having all the possibilities listed for immediate inspection. But
it's more subtle than just hiding verbs -- after all we use
"guess-the-verb" as a derogatory term.
I've felt the same open-endedness in several graphical games, although
more sporadically -- in particular scenes or puzzles rather than built
into the interface.
--Z
--
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the
borogoves..."
Bill
Andrew Plotkin wrote in message ...
>Bill (bv...@inetworld.net) wrote:
>> Very good. The "hidden verb" nature of text adventures lends itself to a
>> better suspension of disbelief than the graphical interfaces. You don't
>> know what verbs are possible in what locations.
>
>Except that one ideal of text games (not universal, but something I
>strive for) is to make it immediately obvious what actions are possible
>for any particular object.
BUT ... if the correct verb is "obvious" is it possible to create a puzzle
that takes a "sufficent" amount of time to solve? I'm not saying that
"guess the verb" should be the technique used to create difficult puzzles
... but hiding the verbs (as a text adventure does) allows for some novel
puzzle solutions that are impossible to create if you are showing all
possible actions.
I should add that I co-designed the interface (and designed the game engine)
for The Return to Zork (1993). So I understand the market reasons for a
graphical interface .... but in my talk I want to "tell it like it is" ...
and the way I see it, no graphical interface I've seen can provide the sense
of "unlimited possibilities" that text input creates.
Am I wrong?
Bill Volk
Ola Sverre Bauge
Francis Irving wrote...
>>care. It sounds cool and to hell with the BBC English Dictionary.
>
>
>Francis.
Is 'ventriloqute' or similiar a valid verb in Esperanto?
Erik Max Francis
Ola Sverre Bauge wrote:
> Is 'ventriloqute' or similiar a valid verb in Esperanto?
There is no _q_ in Esperanto, and _-e_ is an adverbial grammatical
suffix, not a verbal one. A ventriloquist is _ventroparolisto_
(literally, "person who belly speaks"), so "to do ventriloquism" would
be _ventroparoli_ ("to belly speak").
--
Erik Max Francis, &tSftDotIotE / mailto:m...@alcyone.com
Alcyone Systems / http://www.alcyone.com/max/
San Jose, California, United States / icbm://+37.20.07/-121.53.38
\
"I've got the fever for the / flavor of a cracker"
/ Ice Cube
Francis Irving
On Fri, 27 Feb 1998 14:04:42 +0100, "Ola Sverre Bauge"
<o...@bu.telia.no> wrote:
>Francis Irving wrote...
>
>>"Ola Sverre Bauge" <o...@bu.telia.no> wrote:
>>
>>>[1] I don't know if "ventriloqute" is a valid verb and I don't really
>>>care. It sounds cool and to hell with the BBC English Dictionary.
>>
>>And harrah for Esperanto!
>
>Is 'ventriloqute' or similiar a valid verb in Esperanto?
> Ola Sverre Bauge
The quick answer is, yes. The caveat to that is that I don't speak
Esperanto, so I can't be sure.
In Esperanto you are free to construct your own words, in a manner
very similar to the way you constructed "ventriloqute" above. In
fact, this is encouraged, and there are very precise and simple rules
for doing so.
I'm sure somebody reading the group who can speak Esperanto will now
step in with the Esperanto for this particular example...
Daniel Shiovitz
In article <erkyrathE...@netcom.com>,
Andrew Plotkin <erky...@netcom.com> wrote:
>Bill (bv...@inetworld.net) wrote:
>> Very good. The "hidden verb" nature of text adventures lends itself to a
>> better suspension of disbelief than the graphical interfaces. You don't
>> know what verbs are possible in what locations.
>
>strive for) is to make it immediately obvious what actions are possible
>for any particular object.
>
>not having all the possibilities listed for immediate inspection. But
>it's more subtle than just hiding verbs -- after all we use
>"guess-the-verb" as a derogatory term.
Yeah. Hmm. I think we really have two different ideals here. One is
the one that makes us dislike guess-the-verb: "Syntax should never
keep the player from trying an idea out". The other is the one that
makes us like open-endedness: "There should be lots of cool ideas that
do stuff if tried, not all immediately obvious". So I guess the moral
is you want the solutions to be not necessarily obvious at first
glance, but once you think of a solution it should be easy to see if
it works. Of course, that's easier said than done.
[..]
>--Z
--
(Dan Shiovitz) (d...@cs.wisc.edu) (look, I have a new e-mail address)
(http://www.cs.wisc.edu/~dbs) (and a new web page also)
(the content, of course, is the same)
Ola Sverre Bauge
Bill wrote...
>the way I see it, no graphical interface I've seen can provide the
>sense of "unlimited possibilities" that text input creates.
>
>Am I wrong?
>
>Bill Volk
Well, in LucasArts' interface, the permutations are not only finite,
they are quite limited, since there are only the combinations of the
various action icons (three in monkey 3) and the objects which can be
aquired, coupled with every object in the various rooms. Text
adventures do in theory have near-infinite possibilities if you count
unintelligible input, such as "asdfg" or "ventriloqute painting" (if the
parser doesen't understand "ventriloqute").
However, if you only count input that produces a non-standard response
from the game, is IF really that much more unlimited than most graphic
adventures? Or maybe... the limits in graphic adventures are usually
more inherent in the interface, while the limits in text adventures are
inherent in the coding? After all, a text adventure could have infinite
possibilities if an infinite amount of work went into it.
Or less theoretic, if the same amount of manhours what goes into a
modern graphic adventure went into a text adventure, it could have a
pretty good approximation of limitlessness... Imagine a 16MB text
adventure, where the size was taken up by sophistication rather than a
sprawling game world... But I'm being utopian. It wouldn't sell.
About the obvious actions thing, well... to me, puzzle-solving in
adventure games, graphical or not, are all about association, more than
logic. As a book on 8086/8088 assembly language said, while a computer
at the lowest level works purely mathematically, the human brain really
works associatively at all levels ~ if we get it to work mathematically
that is only as a function of association. Adventure games tap into
that association.
Actions are "revealed" to the player through association, for example we
will attempt to open a medicine bottle even if it's not hinted at in the
description because we're used to opening medicine bottles. Inobvious
actions need stronger hints, for instance a description in the chapel in
Zork I contains the word "pray", which is exactly what you're supposed
to do in the room. OK, so praying at an altar isn't exactly uncommon,
but it's not exactly expected from you in adventure games either.
And also from Bill Volk:
>if the correct verb is "obvious" is it possible to create a puzzle
>that takes a "sufficent" amount of time to solve?
It sounds a bit like you're confusing ease of expression with ease of
solving. The adventures I consider the best present the problem in a
way that makes it semi-obvious what to do, and the main challenge of the
puzzle is figuring out the practicalities. In these games, possible
solutions usually come about by building a picture in my mind of how I
perceive the interactive structure ("idea tree", innit called?) to be,
brainstorming[ugh] for possible solutions, then try them.
I become frustrated if I have a harder time figuring out what the puzzle
is than solving the actual puzzle, which was why I didn't enjoy So Far,
among others. And, of course, if you know exactly what to do but are
unable to express yourself (ie guess-the-verb) that's incredibly
annoying too.
Phew, that's enough IF theory out of me for at least 5.184x10^9
processor cycles, I'm off to play Normality...
Ola Sverre Bauge
o...@bu.telia.no
http://w1.2327.telia.com/~u232700165/
My news server cannot be trusted; please CC me if you reply.
"Again I was late. Or, more correctly: I would be the last to come to
school, and monsieur Bouze would tell me off. Too late, too early,
those kinds of words had no meaning in the Land of Eternal Present."
-Schuiten & Peeters, "Memories of the Eternal Present"
JC
On Thu, 26 Feb 1998 10:15:46 -0800, "Bill" <bv...@inetworld.net> wrote:
>Very good. The "hidden verb" nature of text adventures lends itself to a
>better suspension of disbelief than the graphical interfaces. You don't
>know what verbs are possible in what locations.
>
[...]
I find the "hidden verb" nature of text adventures one of the wosrt
candidates for destroying suspension of disbelief. From a HCI
(Human-Computer Interaction) perspective, it is very poor, and it can lead
to a lot of frustration, where either the player can't work out how to
express their meaning, or the game doesn't understand an acceptable
statement (i.e. one which comforms to the rules of the input language, but
the author hasn't thought of).
You can say all you like about the advantages of this input method, but
these problems do exist, and I think this is one of the worst offenders for
putting people off IF.
BTW, I'm saying the current method of input has the problem, not that this
necessarly has to be the case with any command line interface.
';';James';';
Bill
JC wrote in message <34f73538...@news.netspace.net.au>...
>I find the "hidden verb" nature of text adventures one of the wosrt
>candidates for destroying suspension of disbelief. From a HCI
>(Human-Computer Interaction) perspective, it is very poor, and it can lead
>to a lot of frustration, where either the player can't work out how to
>express their meaning, or the game doesn't understand an acceptable
>statement (i.e. one which comforms to the rules of the input language, but
>the author hasn't thought of).
Only if the verb that makes sense to the user doesn't produce a resonable
response.
Look, if you have a menu of actions ... you know your options. We expanded
upon this a bit with RTZ in that any object could be used on any object ...
and this menu might have some unique items depending on the object-object
pairing ... but it's still a fixed set of choices.
It's obvious that every adventure game really does have a fixed set of verbs
at any one given time. The hidden nature of these verbs in a text adventure
can be a pain ... if it isn't done RIGHT. BUT when done right, the hidden
verb can give the illusion that anything resonable might work anywhere.
Let me reveal my hand ... the best idea I can come up with for a graphical
interface that mimics this is to go to a real-time 3D world where you can
perform real physical manipulations with objects. Now that would be
something.
I'd still prefer a text parser interface for NPC conversations.
Bill
Lucian Paul Smith
Ola Sverre Bauge (o...@bu.telia.no) wrote:
: Or less theoretic, if the same amount of manhours what goes into a
: modern graphic adventure went into a text adventure, it could have a
: pretty good approximation of limitlessness... Imagine a 16MB text
: adventure, where the size was taken up by sophistication rather than a
: sprawling game world... But I'm being utopian. It wouldn't sell.
I'm not at all sure this is true.
Standard responses are a way of conditioning the player. If a player
tries something a few different ways, and each method produces the same
response, they will stop doing that thing, whether it be talking to an NPC
or pounding on a door. If pounding on a door is the wrong thing to do, a
good way to encourage the player in the correct direction is to *not* code
up a bajillion responses for hitting the door with the brick, the rock,
the bird, the candle,...
In general, I don't think we've run into this problem very often in IF.
But I do think it's something to keep in mind--more unique responses are
nice, but better placed along paths that will allow the player to
progress, rather than down dead ends. In fact, this could be a new class
of red herrings--many unique responses, none of which work. (Anyone play
'Callahans'? Remember the door at Pyotr's house?)
-Lucian
JC
On Fri, 27 Feb 1998 15:53:30 -0800, "Bill" <bv...@inetworld.net> wrote:
>
>JC wrote in message <34f73538...@news.netspace.net.au>...
>
>>I find the "hidden verb" nature of text adventures one of the wosrt
>>candidates for destroying suspension of disbelief. From a HCI
>>(Human-Computer Interaction) perspective, it is very poor, and it can lead
>>to a lot of frustration, where either the player can't work out how to
>>express their meaning, or the game doesn't understand an acceptable
>>statement (i.e. one which comforms to the rules of the input language, but
>>the author hasn't thought of).
>
>Only if the verb that makes sense to the user doesn't produce a resonable
>response.
>
>Look, if you have a menu of actions ... you know your options. We expanded
>upon this a bit with RTZ in that any object could be used on any object ...
>and this menu might have some unique items depending on the object-object
>pairing ... but it's still a fixed set of choices.
>
>It's obvious that every adventure game really does have a fixed set of verbs
>at any one given time. The hidden nature of these verbs in a text adventure
>can be a pain ... if it isn't done RIGHT. BUT when done right, the hidden
>verb can give the illusion that anything resonable might work anywhere.
If the user can't accurately figure out what is going to work (which a user
*can't* with a hidden verb parser) then they are going to run into trouble.
This is basic HCI. With a "hidden verb" parser, knowing the rules of the
input is not enough for "successful" interaction. In my experience, the
"hidden verb" parser puts a lot of people right off IF.
I'll re-iterate it, you can say all you like about the advantages of the
"hidden verb" parser, but it does have some serious problems. I believe
that if IF is going to reach a wider audience these problems need to be
addressed somehow.
Please take this at face value: The *only* thing I'm trying to say is that
I think the current "hidden verb" method is impeding a wider acceptance of
IF.
>Let me reveal my hand ... the best idea I can come up with for a graphical
>interface that mimics this is to go to a real-time 3D world where you can
>perform real physical manipulations with objects. Now that would be
>something.
>
>I'd still prefer a text parser interface for NPC conversations.
Who said I was advocating a graphical interaface? In my post, I explicitly
stated that I wasn't talking about a problem with a text parser, but with
the "hidden verb" text parser which is currently used -- "BTW, I'm saying
Erik Max Francis
Francis Irving wrote:
> In Esperanto you are free to construct your own words, in a manner
> very similar to the way you constructed "ventriloqute" above. In
> fact, this is encouraged, and there are very precise and simple rules
> for doing so.
Well, _q_ is not a letter in Esperanto, but yes . . .
In Esperanto, it is unwise to borrow words from other languages without
telling anyone, or at least not to borrow words that are not already
fairly widely used. The reason is because many readers, while they can
speak Esperanto, may not know the word in the language you borrowed
from!
Much better is to build new words from unambiguously chosen preexisting
word roots. So in Esperanto, this is _ventr/o/parol/i_, or literally,
"to belly-speak." (Granted, these are the same roots that the english
word _ventriloquism_ is built from.)
Ola Sverre Bauge
Lucian Paul Smith wrote...
>Ola Sverre Bauge (o...@bu.telia.no) wrote:
>: Or less theoretic, if the same amount of manhours what goes into a
>: modern graphic adventure went into a text adventure, it could have a
>: pretty good approximation of limitlessness... Imagine a 16MB text
>: adventure, where the size was taken up by sophistication rather than
>: a sprawling game world... But I'm being utopian. It wouldn't sell.
>
>I'm not at all sure this is true.
The "being near-equal to limitlessness" or the "it wouldn't sell" part?
>Standard responses are a way of conditioning the player. If a player
>tries something a few different ways, and each method produces the same
>response, they will stop doing that thing, whether it be talking to an
>NPC or pounding on a door. If pounding on a door is the wrong
>thing to do, a good way to encourage the player in the correct
>direction is to *not* code up a bajillion responses for hitting the
>door with the brick, the rock, the bird, the candle,...
*If* there is only one right thing to do with a given object. I suspect
most of us have grown so used to it that we take it for granted, but is
this really how things should be? I think that *ideally*, every
solution that is plausible should be rewarded with success. One game
which comes to mind is the graphic adventure Legend of Kyrandia 3, where
reportedly, there are 8 ways to solve the first problem in the game
(getting away from Kyrandia). I only ever found 3 or 4 of them (been a
while since I played it, my memory's a bit hazy).
When dealing with fairly linear games however, your comments hold true.
And I do realize that, especially considering the one-author efforts
modern IF mostly is, linearity is a necessary evil. And in graphic
adventures, linearity has prevailed because it limits the amount of FMV
that needs to rendered/acted out.
Just imagine if you could wear three different clothes prior to an FMV
sequence; the producers would have to make three sequences instead of
one. Then add permutations by being able to solve problems in different
ways, and the CDs (DVD disks, even) would fill up faster than you could
say "xyzzy", not to mention what it would to the budget. This is not
half as much of a problem in IF, which is why I think non-linear IF
could be rocking if a lot of work went into it. The problem is, of
course, that it's (probably) not commercially viable for anyone to work
on it full-time.
<utopia mode initiated>
IF lots of people were working full-time on a text adventure and put
lots of effort into making a bushy idea tree (ie one with lots of ways
to solve problems), it could be a truly interactive delight.
>In general, I don't think we've run into this problem very often in IF.
Or maybe rather, the problem has been there for so long, we don't
realize it's a problem.
>But I do think it's something to keep in mind--more unique responses
>are nice, but better placed along paths that will allow the player to
>progress, rather than down dead ends.
Of course, when speaking of linear adventures, which most are.
>In fact, this could be a new class of red herrings--many unique
>responses, none of which work. (Anyone play 'Callahans'?
>Remember the door at Pyotr's house?)
Must have been before my time. When was it released?
Ola Sverre Bauge
o...@bu.telia.no
http://w1.2327.telia.com/~u232700165/
My news server cannot be trusted; please CC me if you reply.
"Tell us your problems and we'll give you the solutions, take the
solutions and we'll give you more problems. What more can you do with
lemonade in your blood, with lemonade in your brain?"
-Tank Girl, "I've got friends at Bell's End"
Jeff Hatch
Ola Sverre Bauge wrote:
[snip]
My thoughts exactly. As long as a work of IF is logical, added depth is
good. If you have to perform seemingly senseless acts, like throwing
rocks at windows, then a game has to be small enough to encourage the
player to try every possible move.
-Rúmil
Kenneth Albanowski
In article <6d7jlv$gcm$1...@ha2.rdc1.sdca.home.com>,
Bill <bv...@inetworld.net> wrote:
>
>JC wrote in message <34f73538...@news.netspace.net.au>...
>
>>I find the "hidden verb" nature of text adventures one of the wosrt
>>candidates for destroying suspension of disbelief. From a HCI
>>(Human-Computer Interaction) perspective, it is very poor, and it can lead
>>to a lot of frustration, where either the player can't work out how to
>>express their meaning, or the game doesn't understand an acceptable
>>statement (i.e. one which comforms to the rules of the input language, but
>>the author hasn't thought of).
>
>Only if the verb that makes sense to the user doesn't produce a resonable
>response.
>
>Look, if you have a menu of actions ... you know your options. We expanded
>upon this a bit with RTZ in that any object could be used on any object ...
>and this menu might have some unique items depending on the object-object
>pairing ... but it's still a fixed set of choices.
>
>It's obvious that every adventure game really does have a fixed set of verbs
>at any one given time. The hidden nature of these verbs in a text adventure
>can be a pain ... if it isn't done RIGHT. BUT when done right, the hidden
>verb can give the illusion that anything resonable might work anywhere.
This makes me notice a connection I had missed before: "natural language"
programming languages share a lot with traditional text adventures. (The
most common modern example of such a language would be AppleScript, in which
a statement like 'if the third word of the fourth paragraph does not contain
"foo" then...' is possible.)
The problem that afflicts both is that a human can come up with far more
variants then the parser actually expects. The AppleScript designers were
really quite clever -- many common ways of describing relations, like "is
equal to" or "does not equal" work perfectly, and indeed they are very
flexible. Statements regarding objects ("the third word of the first
paragraph", or "the last sentance of the document 'foo'") have a similar
flexibility.
But as soon as you think of something the designers did not, some harmless
synonym or circumlocution, the language breaks, showing itself off to be a
mechanical parser. (Consider BASIC, where "x=3" can also be said "let x=3".
But don't try "let x be equal to 3", or "put 3 into x", or any other
variant.)
The cost of this is that humans _don't_ think in those terms. One phrase is
just as good as another if they mean the same thing. I remember well working
with HyperCard (another natural-language environment from Apple) and driving
myself crazy trying to remember _which particular_ English construction was
required to perform a specific task.
This is directly equivalent to the "guess the word" games we get into when
we think we know the correct action in a text game, but can't convince the
parser to understand us. (At least in the programming language, we have a
fair idea whether the task is actually possible.)
In contrast, a graphical game -- or a computer language that uses a fixed
set of uncommon symbols -- forces us to learn a small set of operations that
combine combinatorically to give us a reasonable (though not extravagant)
set of options for a game, or statements that can be expressed in a
programming language. There isn't going to be hunting for _how_ we say
"equals to" if it can only be said "==". (OTOH, more then one symbology can
confuse you about "=" vs. "==", and so on, which indicates that the
complexity is already rising.)
>Let me reveal my hand ... the best idea I can come up with for a graphical
>interface that mimics this is to go to a real-time 3D world where you can
>perform real physical manipulations with objects. Now that would be
>something.
I have hopes for Ultima 9...
>I'd still prefer a text parser interface for NPC conversations.
Depends on the conversations. Remember that if it weren't for LucasArt's
conversation system, we wouldn't have "How appropriate, you fight like a
cow." :-)
--
Kenneth Albanowski (kja...@kjahds.com)
Lucian Paul Smith
: >Ola Sverre Bauge (o...@bu.telia.no) wrote:
: >: Or less theoretic, if the same amount of manhours what goes into a
: >: modern graphic adventure went into a text adventure, it could have a
: >: pretty good approximation of limitlessness... Imagine a 16MB text
: >: adventure, where the size was taken up by sophistication rather than
: >: a sprawling game world... But I'm being utopian. It wouldn't sell.
: >
: >I'm not at all sure this is true.
: The "being near-equal to limitlessness" or the "it wouldn't sell" part?
What I meant was that filling up a CD with text responses to all possible
actions would not necessarily be 'utopian'--there are good reasons for
limitations.
: I think that *ideally*, every
: solution that is plausible should be rewarded with success.
Here I disagree as well. I *do* think that every solution that is
plausible should be rewarded with a *response*, though.
Let's go back to the example of 'pounding on a door'. Say there's a
locked door with no-one behind it. Somewhere in the game is the key to
the door. However, the player hasn't found the key yet, and thinks they
might be able to attract the attention of someone on the other side of the
door. Suppose they get responses like:
>HIT DOOR WITH TIN CUP.
The tin cup makes a faint mettalic sound as it rattles against the door,
but no-one responds.
>HIT DOOR WITH ROCK.
The rock makes a dull thump as it hits the door, but no-one responds.
>HIT DOOR WITH CAKE.
The cake splatters against the door, but makes no sound at all.
-----------
In this situation, the player might well assume that they need to find an
object which will make the correct sound when pounded on the door. But if
no-one's there, this will never work. This is what I'm trying to say--too
many responses down the wrong path can be detrimental.
: <utopia mode initiated>
: IF lots of people were working full-time on a text adventure and put
: lots of effort into making a bushy idea tree (ie one with lots of ways
: to solve problems), it could be a truly interactive delight.
Here I agree, at least in principle. I think.
: >In general, I don't think we've run into this problem very often in IF.
: Or maybe rather, the problem has been there for so long, we don't
: realize it's a problem.
What I meant to say was that I don't think authors generally overwhelm the
player with too many responses, so the point I'm arguing isn't really that
important. So I was arguing against my point, not yours.
But part of the reason I'm arguing here is so we can realize that 'ideal'
IF *is* within our reach. The work you describe would be nice, but it's
rather disheartening. I'm trying to say, "No, we don't need *all* of
that. Authors can use limitations to work for them."
: >In fact, this could be a new class of red herrings--many unique
: >responses, none of which work. (Anyone play 'Callahans'?
: >Remember the door at Pyotr's house?)
: Must have been before my time. When was it released?
*grin* Uh, about half a year ago. A graphical adventure, released by
Legend. Great work; lots of fun.
-Lucian
Ola Sverre Bauge
JC wrote...
>>verbs at any one given time. The hidden nature of these verbs in
>>a text adventure can be a pain ... if it isn't done RIGHT. BUT
>>when done right, the hidden verb can give the illusion that anything
>>resonable might work anywhere.
>
>user *can't* with a hidden verb parser) then they are going to run into
>trouble.
>
>the input is not enough for "successful" interaction.
Err... what? The difficulties most people have with IF would seem to me
to be mostly connected with learning the sort of phrases the parser
expects. Ken Williams mentions people trying "lift the lid of the box"
instead of "open box", for instance. Sierra's solution to the problem
was sticking an all-icons interface in their games, which sucked
horribly. The IF solution is to include examples of the sort of
instructions expected by the parser.
Yes, the current text parser means you can't know *in advance* what will
be successful or not. That's part of the charm of IF, isn't it? I
mean, you can have an idea as to what sort of input will produce
successful interaction, but you won't know for sure until you try them.
The verbs are hidden, *until they are discovered*. You figure out what
is going to work by trying things out, building and expanding on already
discovered conventions.
>In my experience, the
>"hidden verb" parser puts a lot of people right off IF.
Well yes, insofar as people expect it to understand plain English, or
think they have to ask it nicely(>please kill troll with chainsaw) and
so on. But does this remain a problem once they have learned the
conventions of IF? Is it learning the conventions that is the problem?
I would believe this is mostly a problem with adults: Back in the old
hoary days when I owned a 286 and got hold of my first text adventure
(McMurphy's Mansion I think it was too), learning the conventions was
one of the attractions with the game. Maybe it's the childhood
enthusiasm thing, but this apparently has happened to adults as well,
judging by the age of some r*if netizens.
>I believe that if IF is going to reach a wider audience these problems
>need to be addressed somehow.
Sure. Got a better solution? I'm sure everyone here would love to hear
it.
Wow, I managed to write a whole raif post without babbling excessively
about graphic adventures. Better stop before I do.
.
..
...
....
.....
AAARGH I JUST CAN'T DO IT! <graphic adventure babbling> Now, Normality
was an innovative game with that 3D environment... too bad the NPCs,
interface, story and puzzles were crap. </graphic adventure babbling>
There. It's done. I feel much better now.
Ola Sverre Bauge
o...@bu.telia.no
http://w1.2327.telia.com/~u232700165/
My news server cannot be trusted; please CC me if you reply.
All right / So maybe I dreamt it all / But the blisters on my skin /
They tell me different
-Motorpsycho, "Mountain"
Ola Sverre Bauge
Lucian Paul Smith wrote...
>Ola Sverre Bauge (o...@bu.telia.no) wrote:
>: Lucian Paul Smith wrote...
>: >Ola Sverre Bauge (o...@bu.telia.no) wrote:
>: I think that *ideally*, every
>: solution that is plausible should be rewarded with success.
>
>Here I disagree as well. I *do* think that every solution that is
>plausible should be rewarded with a *response*, though.
Erm. Yeah, that was sort of what I meant, at least when discussing
linear adventures. I sort of perceive any action which yields a
non-standard response successful, anyway, because I feel rewarded. And
we don't want to encourage players into pursuing loose ends, you're
absolutely right.
>Let's go back to the example of 'pounding on a door'. Say there's a
>locked door with no-one behind it. Somewhere in the game is the key to
>the door. However, the player hasn't found the key yet, and thinks
>they might be able to attract the attention of someone on the other
>side of the door. Suppose they get responses like:
[Unique responses snipped]
>In this situation, the player might well assume that they need to find
>an object which will make the correct sound when pounded on the
>door. But if no-one's there, this will never work. This is what I'm
>trying to say--too many responses down the wrong path can be
>detrimental.
Yes. But if there are many right paths, the situation changes. Say for
instance that you may open the door both by unlocking it, knocking on
it, breaking it down, sawing it down with a chainsaw etc. That was the
utopian vision I was trying to convey, a game where there are very few
wrong ways of doing things and a lot of right ones. But that would take
a LOT of work (hence 'utopian').
>But part of the reason I'm arguing here is so we can realize that
>'ideal' IF *is* within our reach. The work you describe would
>be nice, but it's rather disheartening.
To a single author writing games for the hell of it, ie the modern IF
community, yes. It could perhaps be realized as a commercial project,
but that's not very probable is it (hence, again, 'utopian').
>I'm trying to say, "No, we don't need *all* of
>that. Authors can use limitations to work for them."
Agreed; effort where effort is due.
>: >In fact, this could be a new class of red herrings--many unique
>: >responses, none of which work. (Anyone play 'Callahans'?
>: >Remember the door at Pyotr's house?)
>: Must have been before my time. When was it released?
>*grin* Uh, about half a year ago. A graphical adventure, released by
>Legend. Great work; lots of fun.
Huh. It wasn't reviewed in PC Review, and I've never seen it on the
shelves. A conspiracy, no doubt. I heard in comp.sys.ibm.pc. <wheeze>
games.adventure that it's hit budget though. I'll have to look out for
it. Life as a budget tramp has been extraordinarily good lately...
Bill
Kenneth Albanowski wrote in ......
>The cost of this is that humans _don't_ think in those terms. One phrase is
>just as good as another if they mean the same thing. I remember well
working
>with HyperCard (another natural-language environment from Apple) and
driving
>myself crazy trying to remember _which particular_ English construction was
>required to perform a specific task.
>
>This is directly equivalent to the "guess the word" games we get into when
>we think we know the correct action in a text game, but can't convince the
>parser to understand us. (At least in the programming language, we have a
>fair idea whether the task is actually possible.)
>
>In contrast, a graphical game -- or a computer language that uses a fixed
>set of uncommon symbols -- forces us to learn a small set of operations
that
>combine combinatorically to give us a reasonable (though not extravagant)
>set of options for a game, or statements that can be expressed in a
>programming language. There isn't going to be hunting for _how_ we say
>"equals to" if it can only be said "==". (OTOH, more then one symbology can
>confuse you about "=" vs. "==", and so on, which indicates that the
>complexity is already rising.)
So maybe the approach in RTZ was closer to optimal than I thought? No
"hidden verbs" but you do have to figure out what object to use on what
object. Hmmm....
Bill
Steve Young
Ola Sverre Bauge <o...@bu.telia.no> wrote in article
<34fc3...@d2o201.telia.com>...
> Lucian Paul Smith wrote...
; effort where effort is due.
> >: >Remember the door at Pyotr's house?)
>
> >: Must have been before my time. When was it released?
>
> >*grin* Uh, about half a year ago. A graphical adventure, released by
> >Legend. Great work; lots of fun.
>
> Huh. It wasn't reviewed in PC Review, and I've never seen it on the
> shelves. A conspiracy, no doubt. I heard in comp.sys.ibm.pc. <wheeze>
> games.adventure that it's hit budget though. I'll have to look out for
> it. Life as a budget tramp has been extraordinarily good lately...
come from Ola, but I presume it is around this way as you are reading PC
review, so I can't comment if it is available where you are living. This
was the guy who bought it for $1.88cents wasn't it. Don't expect we will
get it so cheap over here.
JC
On Mon, 2 Mar 1998 21:25:14 +0100, "Ola Sverre Bauge" <o...@bu.telia.no>
wrote:
>JC wrote...
>>On Fri, 27 Feb 1998 15:53:30 -0800, "Bill" <bv...@inetworld.net> wrote:
>>>It's obvious that every adventure game really does have a fixed set of
>>>verbs at any one given time. The hidden nature of these verbs in
>>>a text adventure can be a pain ... if it isn't done RIGHT. BUT
>>>when done right, the hidden verb can give the illusion that anything
>>>resonable might work anywhere.
>>
>>If the user can't accurately figure out what is going to work (which a
>>user *can't* with a hidden verb parser) then they are going to run into
>>trouble.
>>
>>This is basic HCI. With a "hidden verb" parser, knowing the rules of
>>the input is not enough for "successful" interaction.
>
>Err... what? The difficulties most people have with IF would seem to me
>to be mostly connected with learning the sort of phrases the parser
>expects. Ken Williams mentions people trying "lift the lid of the box"
>instead of "open box", for instance. Sierra's solution to the problem
>was sticking an all-icons interface in their games, which sucked
>horribly. The IF solution is to include examples of the sort of
>instructions expected by the parser.
>
>Yes, the current text parser means you can't know *in advance* what will
>be successful or not.
Yes, that's what I had said.
>That's part of the charm of IF, isn't it?
Not everyone thinks this. I'm a bit cautious when someone from a group of
enthusiasts of something some part of it is a "charm". Do people outside
this group think so?
>I mean, you can have an idea as to what sort of input will produce
>successful interaction, but you won't know for sure until you try them.
>The verbs are hidden, *until they are discovered*.
Hence "hidden verb".
>You figure out what is going to work by trying things out, building
>and expanding on already discovered conventions.
Yes, you can get more accurate as you learn the sorts of things. But you
can't accurately figure out what is going to work. And what about the
problem of not being able to express your meaning?
>>In my experience, the "hidden verb" parser puts a lot of people right off IF.
>
>Well yes, insofar as people expect it to understand plain English, or
>think they have to ask it nicely(>please kill troll with chainsaw) and
>so on.
That may be the case, but I'm talking about even if they know the
convetions.
>But does this remain a problem once they have learned the
>conventions of IF?
I think so. Thing is, how many of them are going to get to this stage?
And it'll be unlikely you'll hear from these people on this group, isn't
it? It's pretty self-selecting. If the "hidden-verb" style of input puts
you off your not going to bother with IF.
>Is it learning the conventions that is the problem?
I do think that this could be improved.
>I would believe this is mostly a problem with adults: Back in the old
>hoary days when I owned a 286 and got hold of my first text adventure
>(McMurphy's Mansion I think it was too), learning the conventions was
>one of the attractions with the game. Maybe it's the childhood
>enthusiasm thing, but this apparently has happened to adults as well,
>judging by the age of some r*if netizens.
As I said, this is not simply a problem with learning the conventions.
>>I believe that if IF is going to reach a wider audience these problems
>>need to be addressed somehow.
>
>Sure. Got a better solution? I'm sure everyone here would love to hear
>it.
Here's one: Not everyone likes the "hidden-verb" input method, but perhaps
just not until they get used to it. If there was some IF which didn't have
a "hidden-verb" input, with instead a fixed number of verbs, then perhaps
this would attract the more of the type of person who doesn't like the
"hidden-verb" into the IF fold? I'm thinking these people are more likely
to be non-technical people, adults who might not like computer games, but
wouldn't mind something they could read. And who knows, perhaps once
they're "in" they might end up liking the traditional IF with a
"hidden-verb" parser?
Just to be clear, I'm *not* saying that anyone *should* be making this type
of IF. Also, making such IF would in no way invalidate, or stop people
making, the traditional "hidden-verb" IF.
';';James';';
Giles Boutel
Bill <bv...@inetworld.net> wrote in article
<6dhglg$40c$1...@ha2.rdc1.sdca.home.com>...
>
> So maybe the approach in RTZ was closer to optimal than I thought? No
> "hidden verbs" but you do have to figure out what object to use on what
> object. Hmmm....
>
I have to admit - it's the best I've seen. Not because you have to figure
out what object to use with what object (ala LucasArts) but because you
could use an object on another object in different ways (eg, cut the plant
with the knife, or dig out the plant with the knife, or something else)
each with its own explanatory icon. Removes guess the verb, allows
individual objects to have special abilities and interactions, appears to
be as extensible as the designer needs and really does feel like a
graphical text adventure without actually using text.
That being said - parts of the game itself make me want to vomit. But I
blame the writers for most of those (Dwarven Sword of Zork anybody?)
-Giles
Joe Mason
In article <6dhglg$40c$1...@ha2.rdc1.sdca.home.com>,
Bill <bv...@inetworld.net> wrote:
>
>So maybe the approach in RTZ was closer to optimal than I thought? No
>"hidden verbs" but you do have to figure out what object to use on what
>object. Hmmm....
I thought the interface to RTZ was very good - possibly the best I've seen in
a graphic adventure. No, DEFINITELY the best I've seen in a graphic adventure.
I was very disappointed when I'd heard that it had gone to a more Myst-like
system for Nemesis, although I haven't played it yet so I shouldn't really
comment.
I especially liked the first objects, such as the plant, where there were many
options. I can't remember if the number of options stayed that high throughout
the game, either because I stopped noticing the interface (a good thing - the
most successful mechanics are those that are unnoticeable) or because I stopped
enjoying the game. A little of both, I think.
For an RTZ-type interface, though, I think one thing that could improve the
quality is to have a lot of "useless" or "red herring" verbs. It did
degenerate occasionally to "use everything on everything", and if there had
been enough default verbs that this was unfeasable I wouldn't have the urge
to do this. Also, when something like "warm hands" showed up, it was pretty
obvious that this was the action to take. Having one-shot verbs show up on
some other objects, with no results or minor setbacks resulting, would have
helped make these stand out as much.
Joe
Joe
Bill
Thank you for the kind words re: RTZ Interface.
I guess I'll just have to find a way to get back into adventure gaming....
Bill
Joe Mason wrote in message ...
Kenneth Albanowski
In article <6dhglg$40c$1...@ha2.rdc1.sdca.home.com>,
Bill <bv...@inetworld.net> wrote:
>
>So maybe the approach in RTZ was closer to optimal than I thought? No
>"hidden verbs" but you do have to figure out what object to use on what
>object. Hmmm....
If "optimal" means "having some idea what the options are", then sure. One
could consider the graphical games as giving you only a few verbs, and no
more nouns then there are pixels in the game. (And always far fewer then
that). In a text adventure, there are no more nouns then there are words in
the game (and usually considerably fewer), but the number and types of verbs
are unspecified. Thus a text adventure is less constrained then a graphical
RTZ-style one, since the verbs are unbounded.
Whether that is a good thing is another matter.
(Hmm, this also implies that the older Sierra style game with graphics and
text is the worst of both worlds, since you don't know the nouns for the
things on the screen, or the verbs. We'll have to see how Startship Titanic
fares.)
'Course, you can still have guess-what-to-click-on, which throws the noun
boundry sky-high. LucasArts game work around this by giving a name to
everything clickable. I don't recall whether RTZ used that approach.
--
Kenneth Albanowski (kja...@kjahds.com)
J. Holder
Thus spake Steve Young <steve...@eclipse.co.uk>:
: I don't think it has been released in the UK yet. I'm not sure where you
: come from Ola, but I presume it is around this way as you are reading PC
: review, so I can't comment if it is available where you are living. This
: was the guy who bought it for $1.88cents wasn't it. Don't expect we will
: get it so cheap over here.
Actually, it was Ultima VIII I got for US$1.88, and the original
Battletech trilogy also. (The ones that were under the Infocom label:
Battletecch:Crescenthawk's inception BT:CR Revenge, and Mechwarrior (EGA!))
--
John Holder (jho...@frii.com) http://www.frii.com/~jholder/
Sr. Programmer Analyst, J.D.Edwards World Source Company, Denver, CO
http://www.jdedwards.com/