Ramblings and Thoughts from May 15, 1997

Ramblings from May 15, 1997

This is an e-mail I sent to a couple friends Spider and Brad after I saw a talk by Jaron Lanier. That's pretty much all the background you need. It rambles a lot - but it was put together based on my recollections a day afterwards along with the notes I scrawled on a pocket calendar I happened to have with me at the time that was my only writing surface available. The only modifications I made below was to add a couple hyperlinks where appropriate.

The e-mail in its original form from 5/15/97

So yesterday I went to this talk by Jaron Lanier which is a name I'd heard of but didn't really know anything about until yesterday. Mainly went because his announced topic was going to be about the architecture of computers and the intentional (and unintentional effects) they will have on our lives.

Anyway, turns out Lanier was the guy who coined the term Virtual Reality, and the founder of VPL(?) which was the original company doing vr stuff (force gloves, goggles that sort of stuff from what I could tell). Apparently he got forced out by the bean counters and is now a writer/musician/lecturer/philosopher.

So apparently he had just flown in from Warsaw Poland and was totally wiped out. He couldn't remember where he was supposed to go, so he called up his name crossed with Stanford on the web, and voila, found the lecture announcement with the directions. He also noted at that time that it was for a class in the philosophy department focussed on epistemology. So, on the fly, while totally sleep deprived, he changed his talk completely around to match that.

He ended up talking for about 2 hours (possibly a bit more - he had just finished "lecturing" when I had to run to choir rehearsal) about various and sundry topics, swooping off on tangents, and getting back to his point. It was fascinating. He's the kind of person that I would love to spend a night flaming with. Of course, he'd dominate the conversation since he's actually done a ton of stuff, and I've done nothing at this point. but anyway....

Some of the highlights of the talk in no particular order: (this is mostly for my benefit to remember his stuff but figured you might be interested).

His attack on intelligent agents which I followed pretty closely cuz I know Brad's work and stuff. Basically he was posing the question of whether such things are getting smarter or whether we are dumbing ourselves down to make them look smart. In particular, he noted the credit rating algorithm which we all adhere to despite the fact that it doesn't predict bad credit risks etc - yet we completely modify our behavior (using credit cards, keeping them paid up etc) so that we don't run afoul of it. His claim was that similar things would happen with agents - that by not doing things ourselves, and accepting the agents' recommendations without reservation, we would be dumbing ourselves down not freeing ourselves up. I'm not sure I got everything, but if you're interested, he has a couple papers on it on his web site. Just looked back at my notes, and saw the quote "change self to fit the image of self on the computer" - it'd be easier for us to change to fit the computer than to get the computer to change to fit us - we'd be dumbing ourselves down rather than the computer getting smarter - same result either way - our viewpoints would match but not the desired goal.

His idea of sedimentation and locking in. Basically the point was that once a standard reaches a certain point, it becomes "sedimentized" and unchange-able without an extraordinary amount of effort due to backward compatibility etc. This mainly applied to the computer industry obviously but I think it could easily be extended to others. Anyway, examples he cited are MIDI (as he described it, a friend of his was just trying to get 2 synthesizers to talk to each other one weekend in Silicon Valley, hacked this protocol together, which got picked up and eventually slammed into place despite the fact that it sucks according to him and despite the efforts of many people to replace it with Hyper-MIDI, MIDI2 etc etc), MS-DOS (need anything be said?) or the concept of files (he noted that these days CS classes tend to treat the concept of a file as a fact of nature rather than the convention it is).

I don't remember him applying it to other areas, but it's pretty easy to see how one could. When people's beliefs in any area get locked in place they hold to them in the face of almost all evidence. There was one quote in Speaker for the Dead (Orson Scott Card), "People question all of their beliefs except the ones they _really_ believe, and those they never think to question" (paraphrased). Things like religion of any sort, capitalism and the power of the free market, the supremacy of democracy etc.

Pretty obvious stuff actually, but what's interesting is the conclusion he drew from it. Since we are currently in the age which is possibly _defining_ all the standards which will become sedimentized and form the basis of future society (since computing will be ubiquitous) we should think more carefully about what we are doing. His term for it was "Karma Vertigo" - in terms of looking around and seeing how much karma we control as to the direction of society, you can get dizzy because there's so much of it at stake. (I really like the phrase, regardless of the meaning anyway :) )

Obviously, he's pushing for us not to do something stupid like MS-DOS or COBOL or equivalent. His example for a great standard was TCP/IP and how it was a monument to the great ideals of our age, openness, sharing of information, etc and that with such a standard, the WWW or something like it was inevitable where people could publish anything they wanted on line and have many other people see it. (he also cited it as the best example of a workable anarchy - nobody told people to put stuff on the web, nobody paid them to do it - they just did it because they wanted to - course that's changing now with businesses trying to get on, but still an interesting side note).

I'll get back to Karma Vertigo in a bit, but let me touch on some of the other stuff he talked about. Actually he started off talking about cuttlefish (squids and equiv) - there's a "giant cuttlefish from Palao" which was recently discovered which has essentially perfect chameleon control of its skin - in fact, it's got a great "video card" as well and can update any bit of the skin in real time. He said that it would slide in front of a reef and the image of the reef would slide across its skin at the same rate it moved, making it wholly camouflaged. He also said that it had control of a layer of muscle under its skin so it could change its outward shape as well. Pretty cool stuff - he used it to leap into ideas about communication and stuff like that - i was really groggy at this point (only got 3 hours of sleep the previous night and this was in a warm sunny room).

But he later moved into a concept which he called "post-symbolic communication". Right now we use symbols to communicate ideas - his (somewhat corny) example was a sentence like "as cuttlefish we stand on porcelain platters on the rings of Saturn" - we can all visualize that, but to actually do it is essentially impossible right now. but through the magic of symbols, he can communicate the idea to us without having the tech to actually do it.

then he posited ubiquity of VR where children grow up using it to create virtual worlds from a very young age, the way kids these days grow up completely net-savvy. Then maybe it _would_ be possible for them to _create_ the world/ideas they're trying to convey almost as easily as it is for us to create the symbols since they would have such facility with the world building tools of VR. What would be the grammar of such a communication - how would it be used? interesting questions.

It reminded me of an idea in a book I read (Beggars in Spain, Nancy Kress - this idea was the only thing worth taking from the entire book :) ), where a generation of "super-brain" children developed these word or concept nets where you could demonstrate an argument to another person by stringing concepts together into a 3-dimensional structure with all the supporting details tied in, the history, the side points/footnotes. I really like the idea as you can probably tell. Anyway, the tie-in to his concept of post-symbolic communication is obvious I think.

Touches on another point he made - mankind _needs_ long term problems and goals - on the scale of centuries. now that we've mastered nature which was our former goal, we need the sort of problems like mastering post-symbolic communication, exploring virtual reality etc. These are "healthy" technologies which don't involve fucking up the environment or etc but yet give us new frontiers to explore.

Looking through notes, just tossing stuff in here for my later reference - the guy who did the cuttlefish research was Roger Hanlan(sp?) of Wood's Hole. Recommended the book "The Denial of Death" by Ernest Bicker (sp?) "You can't argue with a zombie" weird paper on his web page. "Don't do what comes naturally, push the limits, break the walls" - pretty obvious - again, the encouragement to explore frontiers rather than stick with the status quo.

Back to the Karma Vertigo thing. He noted that right now we're essentially going through a "Constitutional Convention" of cyberspace which will end up being more important than the U.S. one in terms of impact on number of people. Right now, various visions of the future are being tossed around, explored, thrown away, and we can be a part of it all. Those that are involved are deciding the way the masses will view the world via computers for years, decades, possibly centuries - and since he seemed to take it as a given that computers will be ubiquitous, it's the same as deciding what reality is for the same time period.

Similar to a Brad quote. "In particle physics, you'll be hacking the realities of at most 50 people in the world. In building cyberspace, you'll be hacking _everybody's_ reality".

So once again, we get back to the question, WHY THE FUCK AM I IN PHYSICS?!!?

Good question. I don't see a particularly good way to attack the issue that most interests me about all of this stuff, which i've mentioned to both of you, that being virtual communities and how we interact with them. How do people interact with other people via computers? What effect does the form of the computer interface have on the actual communication? I've kinda been exploring this inadvertently through my own experiences via stuff like muds, zephyr, e-mail, discuss meetings, net-news, web pages, mailing lists, etc. never IRC though :)

I need to get to the point where I'm solid enough in my job at work to start to explore this sort of stuff on my own time and get to a point where I can actually accomplish something on it, rather than just idly talking about it all the time. I know, I know, I should just punt the degree here and figure out a way to get paid to do this, but dammit, i've put 2 years in here, it will be a relatively cushy job in a couple months, and i've got the resources of the university to draw on here where i can do stuff like go to talks by Jaron Lanier on a whim, not to mention the computing resources (well okay stanford's sucks compared to MIT :) ) and libraries and stuff. Plus being able to do stuff like play 7 hours of grass volleyball on the weekends just rocks :)

I dunno. I think I've identified where I think I want to go with my life and something I might actually be good at just because i'm one of the few people in this world who _does_ most of his socializing over the net, via the computer - that kind of insight doesn't hurt when i can analyze my own reactions as well as the actions of others via such a medium. Now I just need to figure out how the hell I can support myself doing it...as well as motivating myself to _do_ something instead of talking about it.

Anyway, I'm just going to sit here, listening to the tape of the chorale's last concert and staring off into space, rather than working on the problem set I should be doing. Special relativity just doesn't hold a candle to this stuff for me.

i guess my first step should be to take the time, sit down and read through Lanier's web pages and writings and then write him with some of my ideas, see what he thinks and maybe he can suggest possible avenues to explore. Then again, based on his comment above, it seems like he'd encourage me to find my own way which is probably the right thing anyway.

Oh I should mention - I think I did above, but anyway - he was a great speaker I thought - just full of cool ideas, and even while sleep deprived was able to come up with transitions between them that made sense. I should have punted volleyball today to go see his second talk on post-symbolic communications, but the day was just too nice to think about staying inside when there was vball to be played :)

Whoo. This is long. but hopefully interesting. and even if it isn't, well, I needed to type it up for my files anyway :)

Toss me comments, ideas, etc if you feel like it and have time,
Perlick


Eric Nehrlich's WWW home page / nehrlich@alum.mit.edu