Sorry JackN, my bad. I assmed to much. (I didn't make "an ass out of u and me," just me, thus "assmed." [img]http://216.15.145.59/mainforums/smile.gif[/img] )
[quote]Quite frankly, what I have heard so far from almost every TTS program or hardware I have tried still sounds too machine like.[/quote]
True, and I agree.
I should rephrase what I meant by "robotic" though.
Putting quality of tone aside, my disappointment in TTS was with the robotic quality of emphasis and emotion. (Tone quality is like graphics quality - it's being worked on and will probably slowly get better over time.)
MBROLA was the only TTS system that I found (from about ten different systems) that actually let the user define how syllables and words should be stressed. One of the samples on its site stresses the word "think" in a line similar to "I think that is not the case," to good effect. Most TTS don't seem to allow any freedom in this aspect, even though the idea has been around for decades, leading to that monotonous robotic reading.
The downside to MBROLA is that it can't just read text - input has to be in the form of stresses and phoenemes (or something like that). That's no excuse for true text-to-speech though - even the ancient Amigas could stress syllables with crude markup language: Insert a number before a syllable to represent its stress. If the default number is 5, you could write "incr9edible" to stress the middle.
Speech quality has to include the human aspects as well as the tonal. It's all about interface - speech won't be considered user-friendly to the general populace without at least trying to have human inflections.
Grrrrrrrrrrr I should stop ranting. [img]http://216.15.145.59/mainforums/frown.gif[/img]
[quote]Originally posted by Biggles:
[b]Where's Randy gone?
[/b][/quote]
I'm lurking. I went about as far as I could, regarding general ways to bring drama, storytelling and interactivity together. I was looking forward to trying to jointly work out some of the details - strategies for implementing the ideas, like story-embedded environments, three act structure, information passing between NPC's and how that can changer their behavior/ personality profiles, triggers for moving to the next act, ways of sharing narrative information with the experiencer, etc., but the conversation forked off into three different specific "game" ideas, and then just silliness. Figured it was a good time to start paying attention to writing that actually pays something.
I tried to get Chris Crawford interested in our on-going discussion, btw. But I never got a reply – for whatever reason.
[This message has been edited by Randy (edited 11-23-2001).]
Randy, I'm willing to be lead. I've ranted so much hereabouts that I've got a clean slate - there's nothing lurking on the edges of my mind that will cause me to drive the conversation anywhere. I've reached one of those plateus where I am removed of conflict. So I'll talk about anything. [img]http://216.15.145.59/mainforums/biggrin.gif[/img]
Is there a particular aspect of the mix that you'd like to explore? Or do you feel that it's best to leave the topic for a while and earn some money? Or hopefully best of both worlds? [img]http://216.15.145.59/mainforums/smile.gif[/img]
Throw us an issue and a context; I'm interested.
FWIW, I ran into two or three examples of true art recently that made me realise that my system couldn't produce anything of that quality without a lot of cheating. (That cheating would consist of a [i]huge[/i] soup of artistic/dramatic/whatever ideas and knowledge to choose from, and some brilliant algorithms to do the choosing.) I still beleive we can get satisfactorily close though. (Yeah, that's subjective.) I've realised that I must channel my desire to create art into creating a system that can create art, by identifying the processes I went through and trying to recreate them.
Well, that's how I'm sitting now anyway. [img]http://216.15.145.59/mainforums/smile.gif[/img]
My few emails to Chris Crawford were replied to after a space of several weeks, so you'll probably get a response soon. [img]http://216.15.145.59/mainforums/wink.gif[/img] He seems to be a busy guy.
Biggles<font color=#AAFFAA>The Man Without a Face</font>
Care to share those examples, Grant? [img]http://216.15.145.59/mainforums/smile.gif[/img]
Oh, and I would just like to say that my first game passed a major hurdle today: Basic Direct3D device enumeration. GUI, here I come!
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
[quote]Care to share those examples, Grant?[/quote]
If ya want! Examples, with categorisation and results? [img]http://216.15.145.59/mainforums/wink.gif[/img]
One example was from Baldur's Gate 2. The storyline took a couple of twists in quick succession, and presented them in an emotional way.
The twists were of a huge magnitude (as they often are), but I eventually convinced myself that the twists were driven by (a) the plexor being emotionally involved in the situation, and (b) the nature of the hidden information was such that by definition it could not be exposed to most characters in the game world. Thus, a hidden and focused twist, whilst appearing to be huge, involves only a small portion of the game world.
A drama engine could recreate twists, but it needs to be careful about it. (Obviously.) It would need a variety of rules about how twists work - keep all information hidden or incorrect before revelation, with possible foreshadowing (foreshadowing would have another set of rules), reveal twist, present aftermath of twist. Each step has further rules - revealing the twist could be accomplished by sudden revelation, or long winded explanation, etc. Sudden revelation could occur by presenting a character in a new light, or having a character say something, or having a character appear in an unexpected location, or revealing a known object in an unexpected location, etc etc etc etc.
Yes, my drama system will have a lot of rulesets. [img]http://216.15.145.59/mainforums/smile.gif[/img]
The other example that jumps to mind was music and lyrics. While interactive music will be nearly essential to the drama engine, I haven't really thought much about how this affects my views so I won't say anything about it. Although I'd be keen to hear anyone's ideas on interactive music.
Another thought was of references to things within movies. (There's a fancy word for this but I can't remember it.) Basically, the idea is referencing external influences within the experience for emotional affect.
We could add a ruleset that would allow the drama engine to spoof Titanic. But each possibility would have to be coded by the drama designer. (By whom I mean the person who designs the dramatic rules. There's no reason we can't have drama devices that include "call a nasty place a 'hive of scum and villainy.'")
Oh and congrats, Biggles, even though we've already talked about that in ICQ. [img]http://216.15.145.59/mainforums/smile.gif[/img]
I'm gonna have to work hard to catch up now...
Biggles<font color=#AAFFAA>The Man Without a Face</font>
Considering I've been doing Direct3D for about 3 months and you for much longer, you should be far ahead of me by now. [img]http://216.15.145.59/mainforums/smile.gif[/img]
Any drama engine is going to have a lot of sets of rules. That's the best way to keep it in line with the creator's vision for the experience without making it scripted. Rules can be far more easily adapted to the current situation, scripts tend to be specific. So you would have rule sets for twists, rule sets for language, rule sets for actions, etc. The engine could then take the rules it needs to generate a situation, then create something that meets those rules.
You've heard my views on interactive music many a time in ICQ. [img]http://216.15.145.59/mainforums/smile.gif[/img]
For the third one, do you mean referencing things that have been done in the real world and that the plexor is likely to know about but which have no relation to the experience in order to cause an emotional response in the plexor, such as laughter because of a spoof or a similar emotional response?
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
[quote]For the third one, do you mean referencing things that have been done in the real world and that the plexor is likely to know about but which have no relation to the experience in order to cause an emotional response in the plexor, such as laughter because of a spoof or a similar emotional response?[/quote]
Yeah. Spoofs are classic. So are references to biblical stories. Other examples can be more abstract, for example, in The Phantom Menace, jet engines were used as the base for spacecraft noises because people already associate large flying vehicles with jet noises.
I guess a database could be designed with the rules for each instance of reference, with tags detailing style, type of emotions targetted, etc. The engine could pick a few that match and have the same general goals.
Biggles<font color=#AAFFAA>The Man Without a Face</font>
Classic example: Wandering around in Fallout 2, I came across a wooden bridge over a gorge. At the end of the bridge stood a man. He wouldn't let me cross without first answering three questions. The first two were complex things from the manual and any answer seemed to do. The third question was "What is the average cruising speed of a pigeon carrying a coconut?". Guess what I answered. The man promptly exploded. [img]http://216.15.145.59/mainforums/biggrin.gif[/img]
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
Another. (Although my memory is hazy on this one.)
Wandering around in the desert in Fallout (the original), I came across a crashed UFO. It had "made in Taiwan" etched into it. There were two dead alien bodies lying around. One had a (very nice) ray gun, the other had an autographed picture of Elvis.
Anyway, the technique is more useful when it can instil emotions [i]other[/i] than humour. (Unless you're trying to create a humour-drama engine.)
Biggles (or?), I'm wondering if there are any plans to archive this thread. I continue to talk with folks who may want to read what we've got so far. I'm noticing that everyone seems to be taking a breather for now, and I'm concerned that it will fall off the bottom of the screen and get lost beyond the rim.
Hey, someone's alive in here! [img]http://216.15.145.59/mainforums/smile.gif[/img]
Randy: Better yet, convince those people to drop in and say hi. [img]http://216.15.145.59/mainforums/smile.gif[/img]
Biggles<font color=#AAFFAA>The Man Without a Face</font>
Yeah, that'll keep it up here where all can see it!
Randy: I'll have a look at creating an archive on the boards for important threads such as this.
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
wow... Finally I get to the end of it all after three days (or rather nights after work) of trying.....
While I dont have anything constructive to add yet as Ive had trouble following most of it... I will say this.... I think that the ideas that you guys have all brought up are very impressive... and I hope that there will be more to follow soon.... [img]http://216.15.145.59/mainforums/smile.gif[/img]
------------------
Idiocy is not a learnt thing, but neither is it a genetic thing. It is simply something that jumps from genration to generation as it chooses and latches itself onto certain individuals.
Aint life great :) :) :)
Biggles<font color=#AAFFAA>The Man Without a Face</font>
Just give us a little time and we'll be sure to come up with plenty. [img]http://216.15.145.59/mainforums/smile.gif[/img]
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
Just because things are dead, I'm going to throw in a question and see what happens.
Anyone can play now, don't be shy.
[b]What do you envision as the best interface for character interaction in interactive entertainment?[/b]
If anyone suggests pre-scripted branching conversations ala any RPG in existance, I'll scream.
IMO, we can rule out natural language understanding (whether from voice recognition or typed), as any system with enough knowledge and understanding to comprehend natural language is better suited to taking over the world than providing interactive games.
There's plenty of room for lateral design. Perhaps a plexor shouldn't have direct control over conversation, but set a goal at the beginning of an interaction and let the computer play out the maneuvring.
Ideas?
Biggles<font color=#AAFFAA>The Man Without a Face</font>
Definitely prescripted branching conversations ala any RPG.
[img]http://216.15.145.59/mainforums/biggrin.gif[/img]
I'm not entirely sure what would be best. It would be good if the plexor could say anything they want.
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
I'm not sure either, though I am very, very interested in a voice-driven interface. I understand that a complete natural language system would be unrealistic. But we must keep in mind that in designing interactive entertainment, as in other forms of entertainment, it is okay to use smoke and mirrors – as long as the resulting experience is convincing, given the conventions set up at the outset.
At least the interface should be very simple, none intrusive in terms of experience so that it doesn't scream that "This is a game!" and thus take one out of the experience, and it should be very intuitive in order to bring the masses to the project. Non-gamers would be unfamiliar with, perhaps confused by, and I'll bet uninterested in, the standard game conventions.
[This message has been edited by Randy (edited 12-02-2001).]
just as an embroidery on the actual game engine itself in ITF... it would be cool to be able to issue some simple voice commands to your ships computer...
"target X", full burn, etc
perhaps this is one way of explaining away the fact that your actual machine cant cope with language... the 'ships computer' only understands a limited vocab...
[quote]Originally posted by GrantNZ:
[b]What do you envision as the best interface for character interaction in interactive entertainment?
If anyone suggests pre-scripted branching conversations ala any RPG in existance, I'll scream.
IMO, we can rule out natural language understanding (whether from voice recognition or typed), as any system with enough knowledge and understanding to comprehend natural language is better suited to taking over the world than providing interactive games.[/b][/quote]
I think Pre-scripted branching conversations ala any RPG in existance is a good suggestion here...
hah hah! Just kiding! [img]http://216.15.145.59/mainforums/biggrin.gif[/img]
Alright... Seriously...
If we can't have a machine with real understanding and exchange of language, we have to do all the work ahead of time to account and cover for almost any situation that comes up in a conversation. This is a [b]LOT[/b] of work. People do some strange things that QA people sometimes don't even expect!
I think Shadow Boxer is on the right track, as far as limiting the "language between real and virtual". I believe this can still be done to a point where the mind takes over and makes the leap to interactivity.
I still believe we are years away from totally immersive, believable interactivty. There's still too much work to be done to get there.
An interesting pair of opposites has arisen - the comparison between the necessity of simple interfaces and complex interactions.
Perhaps a game could be focussed more towards interaction, making interaction [i]the[/i] game? I suppose similar to business simulation games, where rather than having a simple economic system as a feature of a larger game, making the economic system the basis for the game.
Would limiting the language limit interactions to the point that drama becomes much harder to achieve? Much drama hinges on emotion. If language was reduced to the point where a plexor couldn't express emotion, would this kill a major factor of drama?
Conversely, if a simple language was created that concentrated on emotion but omitted the communication of concepts, would this create empty feeling? Could this be avoided by creating a system that allows expression of concepts through actions rather than direct communication? (A lot of games already do this to a tiny extent. "Shoot me and we're enemies, shoot my enemy and we're friends.")
I agree that we are years away from total immersive interactivity. Physicists haven't even got complete physics formulae, let alone simulating psychology and spirituality. So, enjoyment and communication must be the key goals. (By communication, I mean communication of theme, of the "message" behind the art.)
What audience should we aim at? The general gamer who is used to clicking on things, selecting from ranges of options and learning proceedures, or a more general audience who needs a free-flowing natural interface? Bearing in mind that presenting a limited natural interface is like asking someone to walk for an hour with metre-long iron bars strapped to their legs. A novelty, but one that gets tiresome and wears off unless there is a [b]lot[/b] of depth and positive feedback from the situation. Also bearing in mind that selecting from icons and menus is complex and slow and can be quite unintuitive to a non-logical thinker, unless (again) there is fantastic feedback about what is going on.
Or should a system be designed that can handle as much detail as the plexor wants to throw at it, but can cope with as little as the plexor can be bothered with?
And even after resolving that issue, no matter the choice, the problem nevertheless remains - how would you acheive it? What system? What interface? What presentation? What goal?
Biggles<font color=#AAFFAA>The Man Without a Face</font>
You would achieve it with a holodeck. That's about the only thing I can think of that could do all that you mention.
I think the best aim would be to go for providing an experience with a natural-feeling interface and a natural flow to the gameplay. That would make it accessible to almost everyone.
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
GrantNZ, if I understand your question, I think a icon based system is best, since the definition of the icons implies a limited vocabulary. At the same time, the rules used to combine icons open up a large amount of flexibility. I'm thinking of a "Sims" type language. (Not having played the game, I not sure how it works in reality.)
Look at the web. The use of emoticons was created to add a "dramatic", i.e., emotive, aspect to a pretty flat environment. Only a small set of characters are used, yet they demonstrate a wide range of emotions by representing the eyes and mouth.
Map the icons to both a "graphical" menu on screen and character combinations on the keyboard, then you can support the novice and the expert.
Use of the iconsto create a new language falls back into my earlier discussion about "alien" vs simplistic: if you can't represent a realistic system, use an "alien" environment, where the user is expecting to learn new things.
Does that make any sense?
An interesting pair of opposites has arisen - the comparison between the necessity of simple interfaces and complex interactions.[/b] [/quote]
This alone is a worthy area of discussion. We could start by defining what our expectations are for interactions. It would depend on the project. My expectations would be communication with characters, the ability to move around in 3D space, including tilting up and tilting down, the ability to examine the environment in detail, including the handling of 3D objects, and a limited ability to defend myself without going into a “fight-mode”. In other words, an interface that gives me the basics and is an unobtrusive as possible, while encouraging the illusion of a “real” experience as much as possible.
The reason that I have been thinking about a sim where the plexor encounters characters who live in a parallel universe on the other side of the monitor glass is that it immediately sets the convention that the computer is the device by which the plexor is allowed a window into the parallel reality. By setting this convention immediately it allows the plexor to immediately adjust to the idea of using the computer as a method of interacting with that realm. Microphones, keyboards, and joysticks become “okay”. At the same time, if the characters on the other side of the screen are presented as “real” characters living in a parallel universe, it gets the plexor away from the idea that this is an ordinary “game”. For instance, say the potential plexor is surfing the Net and “stumbles across” this alternate reality. The character(s) in the alternate reality tell the plexor that they are real and that their world is real, although only represented by the digital images the plexor sees – and they tell the plexor of their plight – and they ask for help. At this point the potential plexor has the choice to join in an adventure with at least potentially “real” alternate reality “people”. What if they say that they’re aliens living on a planet light years from Earth, and this is their method for trying to find an Earther to communicate with. Why they must have an Earther’s help is up to the storyteller.
[quote][b]
Would limiting the language limit interactions to the point that drama becomes much harder to achieve? Much drama hinges on emotion. If language was reduced to the point where a plexor couldn't express emotion, would this kill a major factor of drama?
Conversely, if a simple language was created that concentrated on emotion but omitted the communication of concepts, would this create empty feeling? Could this be avoided by creating a system that allows expression of concepts through actions rather than direct communication? (A lot of games already do this to a tiny extent. "Shoot me and we're enemies, shoot my enemy and we're friends.")[/b][/quote]
Since I am coming from a evocative storytelling/ development of drama point of view, I’ll start with a drama tool.
I would tackle the problem in terms of the creation of “moments”. Each moment would be dealt with separately, in terms of using the best dramatic/storytelling tools to solve the problems of the needs of a specific moment – this as apposed to making general, broad-scope decisions about interactions.
Some moments would best be served by overt dramatic action. Others would best be served by action as it occurs in dialogue and blocking (the purposeful movement and stage business of actors used to communicate subtext and underscore theme). Some moments would find the plexor alone and would best be served by communications from the environment (footprints, a note left behind, blood dripped across a desktop). And some moments would benefit from a combination of tools arranged in an escalating arch of dramatic tension.
Since we’re talking interactive entertainment, we would speak of “potential” moments. Further, each moment could be designed to have dynamic potential.
[quote][b]I agree that we are years away from total immersive interactivity. Physicists haven't even got complete physics formulae, let alone simulating psychology and spirituality. So, enjoyment and communication must be the key goals. (By communication, I mean communication of theme, of the "message" behind the art.)[/b][/quote]
I disagree. I think that we can indeed create total immersive interactivity. We can not create total reality. But there is a difference between the two. If a person watching a great old WB cartoon on TV is totally absorbed in the animation, then it is a totally immersive experience for that person. The living room goes away, consciousness of body goes away, outside noises go away – it’s totally immersive because of a kind of hypnosis – a kind of altered state of consciousness that good storytelling encourages. This process is helped along by the plexor, who comes to the experience with a certain willingness to put aside disbelief. If we add interactivity to what movies and TV have already achieved in their abilities to encourage this “hypnosis”, which is an agreement between entertainers and audience, then surely the resulting experience will be even more immersive.
The contract that audience agrees to with entertainer is that in the beginning of the experience the “conventions” of the world that will be presented will be set. The convention can be that the world will be made of 2D animation. The convention can be that the people we watch (and in the case of interactive entertainment, interact with) will be buffoons, and thus and agreement is made that the convention of “farce” will be set. Or the convention of “reality” can be set up at the beginning. In this case the audience would expect to take part in an experience that seems just like reality (but more fun, of course). Because of this well-known (if subliminal) process of setting the convention of the entertainment up front, there is no need to create “real reality”. The audience (plexor) will agree to become immersed in whatever reality we set as the convention at the outset. We will offer an interactive reality (of whatever sort), and the plexor will disengage disbelief willingly in order to take part in the experience. Regardless of the reality we set as the convention, as long as we maintain internal logic, the plexor will gladly go along for the ride and become immersed (hypnotized).
[This message has been edited by Randy (edited 12-03-2001).]
[quote]You would achieve it with a holodeck. That's about the only thing I can think of that could do all that you mention.[/quote]
That is because I continuously mentioned opposed goals (with the exclusion of creating a [i]completely[/i] interactive system ala holodeck).
The point was to explore the choices available and consider the ups and downs of each, hopefully leading to various implementations of the varied possibilities...
[quote]I think the best aim would be to go for providing an experience with a natural-feeling interface and a natural flow to the gameplay. That would make it accessible to almost everyone.[/quote]
So what to you would be a natural-feeling interface with naturally flowing gameplay? (Given today's technology, of course.) And what would be the costs of using your interface? It's the age-old debate about mass-market appeal...
bobo:
I agree with the icon system, and I'll be using something along those lines when (if I ever get to it!) I develop my system.
The Sims was a bit simplistic, dealing wholly with emotion communication. (Options available were things like "tease," "hit," "hug," "talk," "give gift," "tell joke." There were certainly no information communications. Of course, the game has no storyline so information wasn't a necessity. Nevertheless, it tended to denegrate to "use the most effective befriending action for person X" rather than true interaction.)
[quote]Look at the web. The use of emoticons was created to add a "dramatic", i.e., emotive, aspect to a pretty flat environment.[/quote]
The drama and information there combine to create nice communication - could we recreate this by letting the plexor build or choose a sentence of information, and intertwine it with emotion icons? Perhaps using several emotions per sentence (when required) to build complex messages?
One advantage of this system comes from the fact that displayed emotions are often responded to by a small number of ways. Often either admitted or rebutted. A smile meets a reciprocation, or meets a scowl from a grumpy person. An oppressor inspires either submission or agression. Exceptions occur when the receiver has overwheling emotions of their own - a smile may be met by an outburst of tears from a sad person. Fuzzy logic could be a perfect candidate for actioning all this.
Hmm, I should take notes of that, I like it. [img]http://216.15.145.59/mainforums/smile.gif[/img]
The idea of using an alien environment makes good sense, as long as you're happy to accept an alien environment. Primitive environments (cavemen) would be another contendor. Limiting the number of emotions and messages that can be communicated would ease the problem substantially.
Unfortunately for me, I need a human system. [img]http://216.15.145.59/mainforums/frown.gif[/img]
Randy:
[quote]We could start by defining what our expectations are for interactions. It would depend on the project.[/quote]
My project requires NPCs that can communicate feelings (avoiding the possibility of the plexor treating NPCs as pawns (and yes, I do have a contingency for difficult plexors. Disrespect of lifeforms is the Dark Side of the Force)) and NPCs that can organise behavior to meet goals. Of course, this introduces crossover, where NPCs need to discuss the reasons behind emotions, combining information and feeling. The same conversely, NPCs should be able to anticipate the emotions caused by information. Quite a complex problem really.
Moving around 3D space isn't so important for me. I'm happy to accept the stage/actors paradigm. (Plus, I feel 3D movement hurts the cinematic quality of a game in certain ways.)
The rest of the game is pure action and is thus somewhat irrelevant from a character interaction discussion.
[quote]By setting this convention immediately it allows the plexor to immediately adjust to the idea of using the computer as a method of interacting with that realm.[/quote]
A good idea. There's two good design choices for interface design (IMO), the "explain the reasons" system that you are promoting, and the "mimic reality" system that I seek in action games.
By "mimic reality" (which is probably a bad description), I mean a system where the interface matches what occurs as closely as possible. A sword fighting game where the sword matches the joystic position is far preferred to me than a system where an abstract motion activates a certain maneuvre. Similarly, natural language speech recognition would be obviously preferable to branching conversation.
The difficulty is finding the middle ground.
Anyway, the "explain the reasons" system is all good, as long as the game supports the reasons. It can be lots of fun to learn a system and interact using it.
[quote]I disagree. I think that we can indeed create total immersive interactivity.
...
If we add interactivity to what movies and TV have already achieved in their abilities to encourage this “hypnosis”, which is an agreement between entertainers and audience, then surely the resulting experience will be even more immersive.[/quote]
**Warning: I'm still in philosophy mode so I'm going to talk about the concepts rather than the implementation. I know the ideals of immersivity and interactivity are far beyond us.**
I contend (yeah here I go again, sorry) that interactivity can and will damage immersivity over time unless done perfectly.
The contract between entertainer and viewer for a TV show is partially based on the fact that the entertainer can present his work uninterrupted. Interactivity requires that both parties respond to each other. Unless the computer and plexor can explain their actions to the stage where there are few misunderstandings, immersiveness is interrupted. Also, responses to interactions have to be as real as possible to avoid the plexor discovering exploitable/annoying patterns. And this is impossible without interactions in a very real environment.
Before anyone complains that they aren't writing games for those that would exploit holes, note that there are exploiters [i]in real life.[/i] They're often the loud-yet-quiet type; the type that explosively acts then watches the results carefully. If they're good, by the time they're thirty or so they've picked patterns in reality to the point that they can manipulate a lot of things, even managing to manipulate whole societies [i]without appearing to be the source of the manipulation[/i]. So it isn't an ignorable artifact of game players.
Anyway. The more interactive a game is, the more open to misunderstandings are likely to occur. (Don't you hate when you select what appears to be a friendly option in a game where the designer had intended it to be a hostile one?) Also, the more distant from reality the system is - the less complex - the less interesting and more predictable are the responses.
EA Sport's FIFA (a soccer game) was one example of a game whose responses seemed to be so varied and lifelike I believed I would never tire of playing. Then all of a sudden, the patterns clicked into place and I could pick interactions with an 80% chance of them succeeding as intended, with failures attributed mostly to bad luck. The drop in enjoyability was huge. While we can say that we are happy with a certain lifetime of smoke and mirrors before the secrets are shown, it would be preferable that we don't have to massively rewrite/expand the drama engine between games.
Anyway, I know I probably sound like I'm being pedantic and difficult, but I just don't believe immersivity is solely a factor of setting the conventions of interaction - the system inherently dissuades immersivity due to the communication problems introduced through interactivity, and due to possible simple responses to plexor input.
FWIW, "Abe's Exoddus" does a perfect job of setting the conventions of interactivity. It is the best designed game I have ever seen from the point of view of introducing the interactions and purposes. Unfortunately, game response to the interactions is so predetermined that the game is nothing more than a puzzle. (Of course, that's what the game was [i]made[/i] for, I'm not disputing that.) Being a puzzle rather than an interestingly interactive world reduced the immersiveness considerably. All I was doing was running, jumping, and farting. (Odd game.) It didn't [i]involve[/i] me past a certain level.
For these reasons, I don't believe a drama system would be a lasting one unless it involves some form of "true" AI - whether implemented by neural networks, genetic programs, whatever. Unless someone can come up with a [i]fabulous[/i] algorithm for understanding user input (adjusting understanding until the plexor agrees with the computer's deductions) and response.
shadow boxer:
[quote]F'rinstance, the computer can't make out the instruction you are giving it via the mic... so the character/onboard Fury computer says...
"unable to comply with that request"
"Plexor, say again, you're breaking up, I repeat say again."
In patches it would wear thin but my word would it help...[/quote]
It's a start. Preferable is a system that tries to [i]discuss[/i] input with the plexor, until the two can settle on a system that suits both.
Asking the plexor to rephrase would help - followed by comparing the computer's deduction of the original to the rephrase. Guessing and confirmation is another tool, leading to the computer eventually learning the plexor's communication protocol.
To be honest, having an NPC ask me to repeat because communications are breaking up every time I say something unusual would get rather boring and tiresome. Interactivity needs to go both ways. I have to guess what the computer means, and discover the truth through experimentation and verification - why shouldn't the computer do this too?
The biggest trap in interaction system design seems to be the assumption that the computer must know exactly what the player means all the time, and is therefore given no tools to ascertain the truth. This clashes [i]horribly[/i] with the player's job - determining what's going on in the game world with little information, little idea of the rules going on behind the scenes, little concept of the true attributes of various entities. The player is forced to continually experiment, guess, pray, and buy hint books to determine just what is happening, and [i]then[/i] resolve the situation presented. It can get rather tiresome. And it could be one of the things that dissuade the general public from gaming.
I never can shut up, can I. [img]http://216.15.145.59/mainforums/smile.gif[/img]
Just to prove I can never shut up, I'm going to add one more thing. [img]http://216.15.145.59/mainforums/smile.gif[/img]
I am [i]hellbent[/i] on having NPCs ask the plexor as many questions as the plexor traditionally asks NPCs.
This would play a large role in determining that the plexor means by his actions. I'm not too sure what this could mean for system abusers... at the very least it could result in a lot of surprised NPCs. But perhaps that can be just another part of the game.
Which leads me to say that for a truly interactive experience, the computer should be playing as much as the plexor is...
Biggles<font color=#AAFFAA>The Man Without a Face</font>
Rick: It's definitely the longest sensible thread ever. I don't think it's beaten the ******* Wars yet though.
As for the rest of you, I'll have to read your posts at work. [img]http://216.15.145.59/mainforums/smile.gif[/img]
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
Comments
[quote]Quite frankly, what I have heard so far from almost every TTS program or hardware I have tried still sounds too machine like.[/quote]
True, and I agree.
I should rephrase what I meant by "robotic" though.
Putting quality of tone aside, my disappointment in TTS was with the robotic quality of emphasis and emotion. (Tone quality is like graphics quality - it's being worked on and will probably slowly get better over time.)
MBROLA was the only TTS system that I found (from about ten different systems) that actually let the user define how syllables and words should be stressed. One of the samples on its site stresses the word "think" in a line similar to "I think that is not the case," to good effect. Most TTS don't seem to allow any freedom in this aspect, even though the idea has been around for decades, leading to that monotonous robotic reading.
The downside to MBROLA is that it can't just read text - input has to be in the form of stresses and phoenemes (or something like that). That's no excuse for true text-to-speech though - even the ancient Amigas could stress syllables with crude markup language: Insert a number before a syllable to represent its stress. If the default number is 5, you could write "incr9edible" to stress the middle.
Speech quality has to include the human aspects as well as the tonal. It's all about interface - speech won't be considered user-friendly to the general populace without at least trying to have human inflections.
Grrrrrrrrrrr I should stop ranting. [img]http://216.15.145.59/mainforums/frown.gif[/img]
[b]Where's Randy gone?
[/b][/quote]
I'm lurking. I went about as far as I could, regarding general ways to bring drama, storytelling and interactivity together. I was looking forward to trying to jointly work out some of the details - strategies for implementing the ideas, like story-embedded environments, three act structure, information passing between NPC's and how that can changer their behavior/ personality profiles, triggers for moving to the next act, ways of sharing narrative information with the experiencer, etc., but the conversation forked off into three different specific "game" ideas, and then just silliness. Figured it was a good time to start paying attention to writing that actually pays something.
I tried to get Chris Crawford interested in our on-going discussion, btw. But I never got a reply – for whatever reason.
[This message has been edited by Randy (edited 11-23-2001).]
Is there a particular aspect of the mix that you'd like to explore? Or do you feel that it's best to leave the topic for a while and earn some money? Or hopefully best of both worlds? [img]http://216.15.145.59/mainforums/smile.gif[/img]
Throw us an issue and a context; I'm interested.
FWIW, I ran into two or three examples of true art recently that made me realise that my system couldn't produce anything of that quality without a lot of cheating. (That cheating would consist of a [i]huge[/i] soup of artistic/dramatic/whatever ideas and knowledge to choose from, and some brilliant algorithms to do the choosing.) I still beleive we can get satisfactorily close though. (Yeah, that's subjective.) I've realised that I must channel my desire to create art into creating a system that can create art, by identifying the processes I went through and trying to recreate them.
Well, that's how I'm sitting now anyway. [img]http://216.15.145.59/mainforums/smile.gif[/img]
My few emails to Chris Crawford were replied to after a space of several weeks, so you'll probably get a response soon. [img]http://216.15.145.59/mainforums/wink.gif[/img] He seems to be a busy guy.
Oh, and I would just like to say that my first game passed a major hurdle today: Basic Direct3D device enumeration. GUI, here I come!
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
If ya want! Examples, with categorisation and results? [img]http://216.15.145.59/mainforums/wink.gif[/img]
One example was from Baldur's Gate 2. The storyline took a couple of twists in quick succession, and presented them in an emotional way.
The twists were of a huge magnitude (as they often are), but I eventually convinced myself that the twists were driven by (a) the plexor being emotionally involved in the situation, and (b) the nature of the hidden information was such that by definition it could not be exposed to most characters in the game world. Thus, a hidden and focused twist, whilst appearing to be huge, involves only a small portion of the game world.
A drama engine could recreate twists, but it needs to be careful about it. (Obviously.) It would need a variety of rules about how twists work - keep all information hidden or incorrect before revelation, with possible foreshadowing (foreshadowing would have another set of rules), reveal twist, present aftermath of twist. Each step has further rules - revealing the twist could be accomplished by sudden revelation, or long winded explanation, etc. Sudden revelation could occur by presenting a character in a new light, or having a character say something, or having a character appear in an unexpected location, or revealing a known object in an unexpected location, etc etc etc etc.
Yes, my drama system will have a lot of rulesets. [img]http://216.15.145.59/mainforums/smile.gif[/img]
The other example that jumps to mind was music and lyrics. While interactive music will be nearly essential to the drama engine, I haven't really thought much about how this affects my views so I won't say anything about it. Although I'd be keen to hear anyone's ideas on interactive music.
Another thought was of references to things within movies. (There's a fancy word for this but I can't remember it.) Basically, the idea is referencing external influences within the experience for emotional affect.
We could add a ruleset that would allow the drama engine to spoof Titanic. But each possibility would have to be coded by the drama designer. (By whom I mean the person who designs the dramatic rules. There's no reason we can't have drama devices that include "call a nasty place a 'hive of scum and villainy.'")
Anyway, that'll do. [img]http://216.15.145.59/mainforums/smile.gif[/img]
I'm gonna have to work hard to catch up now...
Any drama engine is going to have a lot of sets of rules. That's the best way to keep it in line with the creator's vision for the experience without making it scripted. Rules can be far more easily adapted to the current situation, scripts tend to be specific. So you would have rule sets for twists, rule sets for language, rule sets for actions, etc. The engine could then take the rules it needs to generate a situation, then create something that meets those rules.
You've heard my views on interactive music many a time in ICQ. [img]http://216.15.145.59/mainforums/smile.gif[/img]
For the third one, do you mean referencing things that have been done in the real world and that the plexor is likely to know about but which have no relation to the experience in order to cause an emotional response in the plexor, such as laughter because of a spoof or a similar emotional response?
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
Yeah. Spoofs are classic. So are references to biblical stories. Other examples can be more abstract, for example, in The Phantom Menace, jet engines were used as the base for spacecraft noises because people already associate large flying vehicles with jet noises.
I guess a database could be designed with the rules for each instance of reference, with tags detailing style, type of emotions targetted, etc. The engine could pick a few that match and have the same general goals.
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
Wandering around in the desert in Fallout (the original), I came across a crashed UFO. It had "made in Taiwan" etched into it. There were two dead alien bodies lying around. One had a (very nice) ray gun, the other had an autographed picture of Elvis.
Anyway, the technique is more useful when it can instil emotions [i]other[/i] than humour. (Unless you're trying to create a humour-drama engine.)
Randy: Better yet, convince those people to drop in and say hi. [img]http://216.15.145.59/mainforums/smile.gif[/img]
Randy: I'll have a look at creating an archive on the boards for important threads such as this.
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
While I dont have anything constructive to add yet as Ive had trouble following most of it... I will say this.... I think that the ideas that you guys have all brought up are very impressive... and I hope that there will be more to follow soon.... [img]http://216.15.145.59/mainforums/smile.gif[/img]
------------------
Idiocy is not a learnt thing, but neither is it a genetic thing. It is simply something that jumps from genration to generation as it chooses and latches itself onto certain individuals.
Aint life great :) :) :)
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
Anyone can play now, don't be shy.
[b]What do you envision as the best interface for character interaction in interactive entertainment?[/b]
If anyone suggests pre-scripted branching conversations ala any RPG in existance, I'll scream.
IMO, we can rule out natural language understanding (whether from voice recognition or typed), as any system with enough knowledge and understanding to comprehend natural language is better suited to taking over the world than providing interactive games.
There's plenty of room for lateral design. Perhaps a plexor shouldn't have direct control over conversation, but set a goal at the beginning of an interaction and let the computer play out the maneuvring.
Ideas?
[img]http://216.15.145.59/mainforums/biggrin.gif[/img]
I'm not entirely sure what would be best. It would be good if the plexor could say anything they want.
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
At least the interface should be very simple, none intrusive in terms of experience so that it doesn't scream that "This is a game!" and thus take one out of the experience, and it should be very intuitive in order to bring the masses to the project. Non-gamers would be unfamiliar with, perhaps confused by, and I'll bet uninterested in, the standard game conventions.
[This message has been edited by Randy (edited 12-02-2001).]
"target X", full burn, etc
perhaps this is one way of explaining away the fact that your actual machine cant cope with language... the 'ships computer' only understands a limited vocab...
[b]What do you envision as the best interface for character interaction in interactive entertainment?
If anyone suggests pre-scripted branching conversations ala any RPG in existance, I'll scream.
IMO, we can rule out natural language understanding (whether from voice recognition or typed), as any system with enough knowledge and understanding to comprehend natural language is better suited to taking over the world than providing interactive games.[/b][/quote]
I think Pre-scripted branching conversations ala any RPG in existance is a good suggestion here...
hah hah! Just kiding! [img]http://216.15.145.59/mainforums/biggrin.gif[/img]
Alright... Seriously...
If we can't have a machine with real understanding and exchange of language, we have to do all the work ahead of time to account and cover for almost any situation that comes up in a conversation. This is a [b]LOT[/b] of work. People do some strange things that QA people sometimes don't even expect!
I think Shadow Boxer is on the right track, as far as limiting the "language between real and virtual". I believe this can still be done to a point where the mind takes over and makes the leap to interactivity.
I still believe we are years away from totally immersive, believable interactivty. There's still too much work to be done to get there.
[img]http://216.15.145.59/mainforums/smile.gif[/img]
[img]http://216.15.145.59/mainforums/smile.gif[/img]
Philosophy time.
An interesting pair of opposites has arisen - the comparison between the necessity of simple interfaces and complex interactions.
Perhaps a game could be focussed more towards interaction, making interaction [i]the[/i] game? I suppose similar to business simulation games, where rather than having a simple economic system as a feature of a larger game, making the economic system the basis for the game.
Would limiting the language limit interactions to the point that drama becomes much harder to achieve? Much drama hinges on emotion. If language was reduced to the point where a plexor couldn't express emotion, would this kill a major factor of drama?
Conversely, if a simple language was created that concentrated on emotion but omitted the communication of concepts, would this create empty feeling? Could this be avoided by creating a system that allows expression of concepts through actions rather than direct communication? (A lot of games already do this to a tiny extent. "Shoot me and we're enemies, shoot my enemy and we're friends.")
I agree that we are years away from total immersive interactivity. Physicists haven't even got complete physics formulae, let alone simulating psychology and spirituality. So, enjoyment and communication must be the key goals. (By communication, I mean communication of theme, of the "message" behind the art.)
What audience should we aim at? The general gamer who is used to clicking on things, selecting from ranges of options and learning proceedures, or a more general audience who needs a free-flowing natural interface? Bearing in mind that presenting a limited natural interface is like asking someone to walk for an hour with metre-long iron bars strapped to their legs. A novelty, but one that gets tiresome and wears off unless there is a [b]lot[/b] of depth and positive feedback from the situation. Also bearing in mind that selecting from icons and menus is complex and slow and can be quite unintuitive to a non-logical thinker, unless (again) there is fantastic feedback about what is going on.
Or should a system be designed that can handle as much detail as the plexor wants to throw at it, but can cope with as little as the plexor can be bothered with?
And even after resolving that issue, no matter the choice, the problem nevertheless remains - how would you acheive it? What system? What interface? What presentation? What goal?
I think the best aim would be to go for providing an experience with a natural-feeling interface and a natural flow to the gameplay. That would make it accessible to almost everyone.
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
-R.
------------------
[i]"...Never start a fight...but [b]always[/b] finish it."[/i]
Look at the web. The use of emoticons was created to add a "dramatic", i.e., emotive, aspect to a pretty flat environment. Only a small set of characters are used, yet they demonstrate a wide range of emotions by representing the eyes and mouth.
Map the icons to both a "graphical" menu on screen and character combinations on the keyboard, then you can support the novice and the expert.
Use of the iconsto create a new language falls back into my earlier discussion about "alien" vs simplistic: if you can't represent a realistic system, use an "alien" environment, where the user is expecting to learn new things.
Does that make any sense?
------------------
bobo
<*>
B5:ITF
[b]
Philosophy time.
An interesting pair of opposites has arisen - the comparison between the necessity of simple interfaces and complex interactions.[/b] [/quote]
This alone is a worthy area of discussion. We could start by defining what our expectations are for interactions. It would depend on the project. My expectations would be communication with characters, the ability to move around in 3D space, including tilting up and tilting down, the ability to examine the environment in detail, including the handling of 3D objects, and a limited ability to defend myself without going into a “fight-mode”. In other words, an interface that gives me the basics and is an unobtrusive as possible, while encouraging the illusion of a “real” experience as much as possible.
The reason that I have been thinking about a sim where the plexor encounters characters who live in a parallel universe on the other side of the monitor glass is that it immediately sets the convention that the computer is the device by which the plexor is allowed a window into the parallel reality. By setting this convention immediately it allows the plexor to immediately adjust to the idea of using the computer as a method of interacting with that realm. Microphones, keyboards, and joysticks become “okay”. At the same time, if the characters on the other side of the screen are presented as “real” characters living in a parallel universe, it gets the plexor away from the idea that this is an ordinary “game”. For instance, say the potential plexor is surfing the Net and “stumbles across” this alternate reality. The character(s) in the alternate reality tell the plexor that they are real and that their world is real, although only represented by the digital images the plexor sees – and they tell the plexor of their plight – and they ask for help. At this point the potential plexor has the choice to join in an adventure with at least potentially “real” alternate reality “people”. What if they say that they’re aliens living on a planet light years from Earth, and this is their method for trying to find an Earther to communicate with. Why they must have an Earther’s help is up to the storyteller.
[quote][b]
Would limiting the language limit interactions to the point that drama becomes much harder to achieve? Much drama hinges on emotion. If language was reduced to the point where a plexor couldn't express emotion, would this kill a major factor of drama?
Conversely, if a simple language was created that concentrated on emotion but omitted the communication of concepts, would this create empty feeling? Could this be avoided by creating a system that allows expression of concepts through actions rather than direct communication? (A lot of games already do this to a tiny extent. "Shoot me and we're enemies, shoot my enemy and we're friends.")[/b][/quote]
Since I am coming from a evocative storytelling/ development of drama point of view, I’ll start with a drama tool.
I would tackle the problem in terms of the creation of “moments”. Each moment would be dealt with separately, in terms of using the best dramatic/storytelling tools to solve the problems of the needs of a specific moment – this as apposed to making general, broad-scope decisions about interactions.
Some moments would best be served by overt dramatic action. Others would best be served by action as it occurs in dialogue and blocking (the purposeful movement and stage business of actors used to communicate subtext and underscore theme). Some moments would find the plexor alone and would best be served by communications from the environment (footprints, a note left behind, blood dripped across a desktop). And some moments would benefit from a combination of tools arranged in an escalating arch of dramatic tension.
Since we’re talking interactive entertainment, we would speak of “potential” moments. Further, each moment could be designed to have dynamic potential.
[quote][b]I agree that we are years away from total immersive interactivity. Physicists haven't even got complete physics formulae, let alone simulating psychology and spirituality. So, enjoyment and communication must be the key goals. (By communication, I mean communication of theme, of the "message" behind the art.)[/b][/quote]
I disagree. I think that we can indeed create total immersive interactivity. We can not create total reality. But there is a difference between the two. If a person watching a great old WB cartoon on TV is totally absorbed in the animation, then it is a totally immersive experience for that person. The living room goes away, consciousness of body goes away, outside noises go away – it’s totally immersive because of a kind of hypnosis – a kind of altered state of consciousness that good storytelling encourages. This process is helped along by the plexor, who comes to the experience with a certain willingness to put aside disbelief. If we add interactivity to what movies and TV have already achieved in their abilities to encourage this “hypnosis”, which is an agreement between entertainers and audience, then surely the resulting experience will be even more immersive.
The contract that audience agrees to with entertainer is that in the beginning of the experience the “conventions” of the world that will be presented will be set. The convention can be that the world will be made of 2D animation. The convention can be that the people we watch (and in the case of interactive entertainment, interact with) will be buffoons, and thus and agreement is made that the convention of “farce” will be set. Or the convention of “reality” can be set up at the beginning. In this case the audience would expect to take part in an experience that seems just like reality (but more fun, of course). Because of this well-known (if subliminal) process of setting the convention of the entertainment up front, there is no need to create “real reality”. The audience (plexor) will agree to become immersed in whatever reality we set as the convention at the outset. We will offer an interactive reality (of whatever sort), and the plexor will disengage disbelief willingly in order to take part in the experience. Regardless of the reality we set as the convention, as long as we maintain internal logic, the plexor will gladly go along for the ride and become immersed (hypnotized).
[This message has been edited by Randy (edited 12-03-2001).]
Spot on as usual Randy... finding precisely the right tools to suspend disbelief is all you need.
F'rinstance, the computer can't make out the instruction you are giving it via the mic... so the character/onboard Fury computer says...
"unable to comply with that request"
"Plexor, say again, you're breaking up, I repeat say again."
In patches it would wear thin but my word would it help...
[quote]You would achieve it with a holodeck. That's about the only thing I can think of that could do all that you mention.[/quote]
That is because I continuously mentioned opposed goals (with the exclusion of creating a [i]completely[/i] interactive system ala holodeck).
The point was to explore the choices available and consider the ups and downs of each, hopefully leading to various implementations of the varied possibilities...
[quote]I think the best aim would be to go for providing an experience with a natural-feeling interface and a natural flow to the gameplay. That would make it accessible to almost everyone.[/quote]
So what to you would be a natural-feeling interface with naturally flowing gameplay? (Given today's technology, of course.) And what would be the costs of using your interface? It's the age-old debate about mass-market appeal...
bobo:
I agree with the icon system, and I'll be using something along those lines when (if I ever get to it!) I develop my system.
The Sims was a bit simplistic, dealing wholly with emotion communication. (Options available were things like "tease," "hit," "hug," "talk," "give gift," "tell joke." There were certainly no information communications. Of course, the game has no storyline so information wasn't a necessity. Nevertheless, it tended to denegrate to "use the most effective befriending action for person X" rather than true interaction.)
[quote]Look at the web. The use of emoticons was created to add a "dramatic", i.e., emotive, aspect to a pretty flat environment.[/quote]
The drama and information there combine to create nice communication - could we recreate this by letting the plexor build or choose a sentence of information, and intertwine it with emotion icons? Perhaps using several emotions per sentence (when required) to build complex messages?
One advantage of this system comes from the fact that displayed emotions are often responded to by a small number of ways. Often either admitted or rebutted. A smile meets a reciprocation, or meets a scowl from a grumpy person. An oppressor inspires either submission or agression. Exceptions occur when the receiver has overwheling emotions of their own - a smile may be met by an outburst of tears from a sad person. Fuzzy logic could be a perfect candidate for actioning all this.
Hmm, I should take notes of that, I like it. [img]http://216.15.145.59/mainforums/smile.gif[/img]
The idea of using an alien environment makes good sense, as long as you're happy to accept an alien environment. Primitive environments (cavemen) would be another contendor. Limiting the number of emotions and messages that can be communicated would ease the problem substantially.
Unfortunately for me, I need a human system. [img]http://216.15.145.59/mainforums/frown.gif[/img]
Randy:
[quote]We could start by defining what our expectations are for interactions. It would depend on the project.[/quote]
My project requires NPCs that can communicate feelings (avoiding the possibility of the plexor treating NPCs as pawns (and yes, I do have a contingency for difficult plexors. Disrespect of lifeforms is the Dark Side of the Force)) and NPCs that can organise behavior to meet goals. Of course, this introduces crossover, where NPCs need to discuss the reasons behind emotions, combining information and feeling. The same conversely, NPCs should be able to anticipate the emotions caused by information. Quite a complex problem really.
Moving around 3D space isn't so important for me. I'm happy to accept the stage/actors paradigm. (Plus, I feel 3D movement hurts the cinematic quality of a game in certain ways.)
The rest of the game is pure action and is thus somewhat irrelevant from a character interaction discussion.
[quote]By setting this convention immediately it allows the plexor to immediately adjust to the idea of using the computer as a method of interacting with that realm.[/quote]
A good idea. There's two good design choices for interface design (IMO), the "explain the reasons" system that you are promoting, and the "mimic reality" system that I seek in action games.
By "mimic reality" (which is probably a bad description), I mean a system where the interface matches what occurs as closely as possible. A sword fighting game where the sword matches the joystic position is far preferred to me than a system where an abstract motion activates a certain maneuvre. Similarly, natural language speech recognition would be obviously preferable to branching conversation.
The difficulty is finding the middle ground.
Anyway, the "explain the reasons" system is all good, as long as the game supports the reasons. It can be lots of fun to learn a system and interact using it.
[quote]I disagree. I think that we can indeed create total immersive interactivity.
...
If we add interactivity to what movies and TV have already achieved in their abilities to encourage this “hypnosis”, which is an agreement between entertainers and audience, then surely the resulting experience will be even more immersive.[/quote]
**Warning: I'm still in philosophy mode so I'm going to talk about the concepts rather than the implementation. I know the ideals of immersivity and interactivity are far beyond us.**
I contend (yeah here I go again, sorry) that interactivity can and will damage immersivity over time unless done perfectly.
The contract between entertainer and viewer for a TV show is partially based on the fact that the entertainer can present his work uninterrupted. Interactivity requires that both parties respond to each other. Unless the computer and plexor can explain their actions to the stage where there are few misunderstandings, immersiveness is interrupted. Also, responses to interactions have to be as real as possible to avoid the plexor discovering exploitable/annoying patterns. And this is impossible without interactions in a very real environment.
Before anyone complains that they aren't writing games for those that would exploit holes, note that there are exploiters [i]in real life.[/i] They're often the loud-yet-quiet type; the type that explosively acts then watches the results carefully. If they're good, by the time they're thirty or so they've picked patterns in reality to the point that they can manipulate a lot of things, even managing to manipulate whole societies [i]without appearing to be the source of the manipulation[/i]. So it isn't an ignorable artifact of game players.
Anyway. The more interactive a game is, the more open to misunderstandings are likely to occur. (Don't you hate when you select what appears to be a friendly option in a game where the designer had intended it to be a hostile one?) Also, the more distant from reality the system is - the less complex - the less interesting and more predictable are the responses.
EA Sport's FIFA (a soccer game) was one example of a game whose responses seemed to be so varied and lifelike I believed I would never tire of playing. Then all of a sudden, the patterns clicked into place and I could pick interactions with an 80% chance of them succeeding as intended, with failures attributed mostly to bad luck. The drop in enjoyability was huge. While we can say that we are happy with a certain lifetime of smoke and mirrors before the secrets are shown, it would be preferable that we don't have to massively rewrite/expand the drama engine between games.
Anyway, I know I probably sound like I'm being pedantic and difficult, but I just don't believe immersivity is solely a factor of setting the conventions of interaction - the system inherently dissuades immersivity due to the communication problems introduced through interactivity, and due to possible simple responses to plexor input.
FWIW, "Abe's Exoddus" does a perfect job of setting the conventions of interactivity. It is the best designed game I have ever seen from the point of view of introducing the interactions and purposes. Unfortunately, game response to the interactions is so predetermined that the game is nothing more than a puzzle. (Of course, that's what the game was [i]made[/i] for, I'm not disputing that.) Being a puzzle rather than an interestingly interactive world reduced the immersiveness considerably. All I was doing was running, jumping, and farting. (Odd game.) It didn't [i]involve[/i] me past a certain level.
For these reasons, I don't believe a drama system would be a lasting one unless it involves some form of "true" AI - whether implemented by neural networks, genetic programs, whatever. Unless someone can come up with a [i]fabulous[/i] algorithm for understanding user input (adjusting understanding until the plexor agrees with the computer's deductions) and response.
shadow boxer:
[quote]F'rinstance, the computer can't make out the instruction you are giving it via the mic... so the character/onboard Fury computer says...
"unable to comply with that request"
"Plexor, say again, you're breaking up, I repeat say again."
In patches it would wear thin but my word would it help...[/quote]
It's a start. Preferable is a system that tries to [i]discuss[/i] input with the plexor, until the two can settle on a system that suits both.
Asking the plexor to rephrase would help - followed by comparing the computer's deduction of the original to the rephrase. Guessing and confirmation is another tool, leading to the computer eventually learning the plexor's communication protocol.
To be honest, having an NPC ask me to repeat because communications are breaking up every time I say something unusual would get rather boring and tiresome. Interactivity needs to go both ways. I have to guess what the computer means, and discover the truth through experimentation and verification - why shouldn't the computer do this too?
The biggest trap in interaction system design seems to be the assumption that the computer must know exactly what the player means all the time, and is therefore given no tools to ascertain the truth. This clashes [i]horribly[/i] with the player's job - determining what's going on in the game world with little information, little idea of the rules going on behind the scenes, little concept of the true attributes of various entities. The player is forced to continually experiment, guess, pray, and buy hint books to determine just what is happening, and [i]then[/i] resolve the situation presented. It can get rather tiresome. And it could be one of the things that dissuade the general public from gaming.
I never can shut up, can I. [img]http://216.15.145.59/mainforums/smile.gif[/img]
I am [i]hellbent[/i] on having NPCs ask the plexor as many questions as the plexor traditionally asks NPCs.
This would play a large role in determining that the plexor means by his actions. I'm not too sure what this could mean for system abusers... at the very least it could result in a lot of surprised NPCs. But perhaps that can be just another part of the game.
Which leads me to say that for a truly interactive experience, the computer should be playing as much as the plexor is...
As for the rest of you, I'll have to read your posts at work. [img]http://216.15.145.59/mainforums/smile.gif[/img]
------------------
[b][url="http://www.minbari.co.uk/log12.2263/"]Required reading[/url][/b]
Never eat anything bigger than your own head.
The Balance provides. The Balance protects.
"Nonono...Is not [i]Great[/i] Machine. Is...[i]Not[/i]-so-Great Machine. It make good snow cone though." - Zathras
I would just like to thank you all for keeping my name near the top of the board for the past month.
------------------
--milo
[url="http://www.starshatter.com"]http://www.starshatter.com[/url]