Stephenson's Snow Crash revisited

Todays Thinker's discussion caused me to buy a copy of Snow Crash by Neal Stephenson. I've seen it referenced frequently in the forums and thought it was time I read it. Within a few pages I realized I had already read it, probably more than ten years ago when it first came out. Now that I am actually in a virtual reality the book has a different impact than it did originally. Most interesting was the discussion on page 64 of the Bantam paperback edition where Neal Stephenson introduces a key element in the novel: the possibility of reading emotions on avatar's faces. The character Juanita has made the "Black Sun" successful by inventing a process that enables avatars to "show something close to real emotion" and the the Nipponese "more or less ignore what is being said...and pay attention to the facial expresions and body language of the people they are talking to".

If you follow the book closely you will realize that it depends greatly on this device. Despite existing in a virtual reality the characters are able to interact on the basis of shared, observable emotional responses.

What is curious is that no explanation is given for how the emotional resposes of someone at a keyboad could be transmited to the avatar. The teckies will solve that.

What is even more interesting is the unspoken condition that seems to hold in the novel that no one at the keyboard has the option to mask or distort their real world emotional responses while in the virtual world.

So, a question: how would you like it if SL was like the world of Snow Crash: your emotions at the keyboard are transmited to your avatar's face and body without the option to edit or mask?

I think even without being able to work the gesture controls like a master puppeteer, people in SL manage to convey emotions quite handily.

They do it with the smilies: : ) or : (

I find people can speak volumes with just a : (

I think some elaborately mask their feelings or true emotions but they leak through in different ways -- abruptly flying up and down...pacing around with the avatar...bumping or pushing you...sitting down and collapsing on the ground...there are a lot of little things that aren't even in the repetoire of "gestures" per se that convey a feeling.

I wouldn't like that so much as I'd like a way to let people know what I'm doing, since sometimes I just jump away from my keyboard without thinking, or I minimize into another app.

LOL! Well I'd certainly stop people talking to me quick enough, if my avatar is rolling its eyes, pulling disdainful faces and making 'wanker' hand signs as well. Uh... I mean.. uh...

Anyhoo, I tried to read Snowcrash. I admit I didnt finish. I really don't get what all the fuss is about. I think it's poorly written, takes a great idea and manages to tell it in the most trivial way possible so as to make a potentially exciting concept seem positively dull.

Agreed. Mostly. And yeah... I do that too.

Same here, I was really disappointed in it and only finished read it because so many people talked about it here. Phrases like "Deliverator" just made me cringe to be honest. Not a patch on William Gibson or Jeff Noon as far as cyberpunk goes.

/120/7c/35707/1.html

We had a discussion here ... I argue that Stepheson really doesn't fit into cyberpunk, and I believe fans of cyberpunk are often disappointed when they read the book and expect that.

Although I'm not entirely sure how you can safely seperate Snowcrash from being cyberpunk, my main argument is that I just thought the book was poorly written and the story well, not really amazingly engaging, in which respect you can compare it to many other genres if you'd like to run with it There, I've freed you from the need to discuss just in terms of cyberpunk.

Now prepare me a 10,000 word dissertation on it

Uh 10,000 words how are they suppose to sign it for petes sake..??

LMAO

Shadow

I'd like to join the group shit-in, but I loved this book.

You're all dead to me.

The story was nice, the writing was poor. What really got me was the concepts that were new for the time.

One of the more intresting and partialy-amusing thing to remember is that most of the virtual worlds that got created since the book was released, were somehow based on inspired by the metaverse described in the novel.

The author is just better at making intresting concepts and designs then telling a story.

please folks, I didn't introduce this thread to discuss Stephenson's writing skills. Frankly I don't think that would be worth the trouble.

The point is that Stephenson needed to invent a device that conveyed authentic emotional content from real life to the web in order to make a story possible.

We have very crude devices for conveying emotional content in SL. We have no way to confirm the reality of the emotional content. I might be smiling right now, or not, but putting in a smiley face doesn't make it true.

The real question is, "Are we, in SL, part of a subculture that doesn't care about authentic emotional content?"

Perhaps (and Neal Nomad might be looking very apprehensive as he says this, or not) we are part of a subculture that isn't interested in emotional content, or perhaps we are getting just enough to be content,

or perhaps we are afraid of emotional content, .

Yup, the story got bogged down in a lot of places for me, especially when he was on about the connection to languages and ancient civilizations and blah blah blah. The descriptions of the metaverse and the burbclaves were enjoyable, though, as well as the fighting done near the end on the floating city made of old boats and varied detritus. I must admit I never finished about the last 20 pages of the book; I suppose I should do that.

Do I care about "authentic emotional content?" I'm not sure exactly what that means. Perhaps we have "inauthentic emotional content." I suppose we construct our own emotional interpretation of everything in SL on a person-by-person basis. I don't expect my SL experience is quite the same as anyone else's. But I suppose the same thing could be said about real life, as well.

Agreed. These people are not shiny.

I'm all for conveying the emotions, but not hooked into my real ones for all to see without censure! Better gestures, emotes, whatever... fine. But really, no... I absolutely do not want to feed any realtime biometric data into my avatar. The thought is more than a little scary.

I mean, be honest here... if your avatar displayed your real emotions and body language all the time, you're going to offend half the people you meet. lol

I guess it would encourage far more honesty and far less roleplay, though. But then it wouldn't really be a Second Life. It'd be a virtual first one

"Well, yes, to be honest I'm bored with you. The horrified-cum-disgusted look? Oh, well your avatar is the ugliest piece of crap I've ever seen, and your pathetic attempts at flirting are kinda repulsive. Actually, you creep me out. Period."

I kinda liked the book.

As for emotions, it would be interesting if they could be more spontanious and natural than the current system. Perhaps someday they'll be a mood ring (or polygraph?) device one can wear to detect and convey emotions? Or a mask which detects changes in facial muscles and applies the same to the avitar?

Prokofy pointed out that in SL one can convey a wide range of emotions:


One could argue that these could be faked, but emotions are faked in RL too.

You could pull it off with a refinement of those console game camera-based add-ons that "put you in the game". You'd probably have to "train" the interface to interpret your real facial gestures and translate them to the avatar, but it should be possible with existing technology. It doesn't seem like it would even be especially difficult.

Of course, for best results you'll have to play with little reflective dots on your face and one on the tip of your tongue...

I wonder if it's almost possible now to use a web cam and streaming video to put your RL face on your avatar. Assuming that you figure out a way to stream video from your web cam, you could attach a mask prim that acts like a "screen". Imagine walking up to someone on your land and making faces at them!

Well, for starters, the Metaverse in Snow Crash used voice chat for communication, if I remember correctly. When one 'goggled in", I immagine the goggles could be programed with some biometric and/or facial expression readers.

Something that interests me as far as virtual worlds go (but I won't write a diertation on it here) is the issue of avatar body language and how it might be implimented so that something like SL's gestures could come as naturaly as talking or typing. When we're just typing (or for some, using voice chat), I tend to think of it as comming from the person behind the keyboard instead of the avatar: there's that layer of disconnect. One can be reading the text and understanding with very limited knowledge of what the avatar is doing...usualy just standing and facing you, with the occsional animation thrown in like a smilie would be. But someone with enough patience can configure gestures so that it would be possible to carry on a whole conversation without a single typed word, and that's actualy using the avatar as the channel though which communication occurs.

Actually, it was more like a DLP projecting light on a pair of glasses you wore.

Anyhow, more to the point of the thread and less about opinions on Snowcrash itself --

Any physical system would require you to be at your computer at all times. I think it would also ruin the "window" we have between ourselves and our avatar selves which might be uncomfortable for some people.

What could be useful however, is a more streamlined way to allow people in SL to know what you're doing behind your avatar. It doesn't have to be emotional, but it could be more than "afk."

DEAD

To answer the original question, the decks used in Snow Crash painted the virtual world right onto the user's retinas with a laser that was emitted from some kind of device on top of the deck. If the tech can do that, it would seem trivial to also have the deck scan the person's face. The deck is certainly capable of reading body position: Hiro executes the moves necessary to slay the Nipponese opponent in The Black Sun in real life and winds up outside his flat because of this.

What wasn't explained was, what happens when you have your eyes turned away from the deck...?

BTW, read Diamond Age. God that book is awesome.

Yeah he never explained what happens when someone gets in the way or you turn the wrong way really. That kinda bugged me.

The original question...

It depends on how its acheived. I actualy am usualy pretty emotionless when staring at the computer, sometimes I smile, sometimes I laugh, but rarely do I go through the actual emotions when I might do "lol" "" "" "" etc. in my head I'm having that attitude, but my face remains unchanged. This may change if *everyones* emotions were expressed in real time, such as in real life conversations. It would be interisting to see the social side effects of such a thing.

How it is done though depends alot on what happens. If its automaticly reading your face all the time, that would be a bit different than say if you could just make a face, then manualy push it through. Just like how I manualy have to type I may manualy have to smile then in some way inform the computer to display that smile (either through thinking it or through a buttonpress or something).

The point of how our gestures in SL currently achieve this in some ways is a good one. You can read an avatars behavior somewhat, if they're bored, or nervouse, or their personality type from how they move (playfull, serious, personal-spaced).

The way SL currently is built though makes reading faces a bit hard.

There.com actualy did this in a slightly better way with the chat being based on "circles" of people with your camera mostly focused on their upper body, and face. The chat then recognizes phrases which change the avatars facial expressions to somewhat slightly exhadurated expressions of normal faces.

With SL, our cameras are usualy in 3rd person and focused pretty far, not giving you too good a view of someones face unless you go into first person, zoom, or over the shoulder tab. So I fear alot of people may not at first even recognize that such emotions are being displayed, unless SL's default camera views change.

It would also be interisting to see what camera mode people walk around with most often if such emotions could be displayed. I myself would probably use the over the shoulder Tab view to look at the last person who typed, but even this may not help as the person may not be close enough, facing your direction, etc.. So then real life circumstances can come into play where someone may say something but may be facing the wrong direction so you can't "read" them.

If we advanced to a point where we had some devices such as those in Snow Crash, then the whole perspective of SL may change. The 3rd person view may become a view mostly used by builders or during exploration, and first person may become the norm for socializing.

So in short, I would like to see the expressions thing happen, if not for the pure interest in seeing how it effects people.

Lately, I have been in conversations with people who have had their poses actively change while we talked (moreso than the random poses changes that happen normally). I don't know if this is an object that does it for them, or they are doing it manually. Either way, it made me feel like they were more engaged, as none of them involved yawning or looking away!

Having an object listen to what you type and apply gestures/animations based on your words could be easily made. The server overhead would suck, but a creative person could have a ball with this. It could even be expanded to listen to the words of the person talking to you and react automatically, even if the words you eventually typed ran counter and your face changed right away.

It would be cool if an objectionable person got within range of you and began talking - your av would involuntarily grimace before smiling a big hello!

EDIT:

I started typing this before I had read Oz's post. He had kind of covered this already - sorry, Oz!

How dare you! Hehe, its ok, I didn't really go into that and I don't own the topic.

That is a good idea, wouldn't be too impossible to make either, hmm.

I've already setup gestures to respond to when I say stuff like "no" (shake head) "dunno" (shrug), etc. But it looks kinda hackney when you have a sentance like... "No, I really like it." and your avatars head shakes.

So probably doing something with a script rather than pure gestures is the way to go. Especialy with behavioral changes based on whose around you. Like maybe your movements become more goofy around one person, or more lovey with another, or frigid with another.

Mmm.