From extropians-request@extropy.org Sun Aug 22 11:13:16 1993 Return-Path: Received: from usc.edu by chaph.usc.edu (4.1/SMI-4.1+ucs-3.0) id AA16943; Sun, 22 Aug 93 11:13:08 PDT Errors-To: Extropians-Request@gnu.ai.mit.edu Received: from news.panix.com by usc.edu (4.1/SMI-3.0DEV3-USC+3.1) id AA13354; Sun, 22 Aug 93 11:12:48 PDT Errors-To: Extropians-Request@gnu.ai.mit.edu Received: by news.panix.com id AA09029 (5.65c/IDA-1.4.4 for more@usc.edu); Sun, 22 Aug 1993 14:07:11 -0400 Date: Sun, 22 Aug 1993 14:07:11 -0400 Message-Id: <199308221807.AA09029@news.panix.com> To: Extropians@extropy.org From: Extropians@extropy.org Subject: Extropians Digest X-Extropian-Date: August 22, 373 P.N.O. [18:07:04 UTC] Reply-To: extropians@extropy.org Errors-To: Extropians-Request@gnu.ai.mit.edu Status: RO Extropians Digest Sun, 22 Aug 93 Volume 93 : Issue 233 Today's Topics: AI/EVOL: slaves, selfishness, evo [1 msgs] AI: slaves, selfishness, evo [1 msgs] AI: slaves, selfishness, evolution [2 msgs] HEx: Buy in to DC-X [1 msgs] MEDIA: Call for Papers for Perforations #5 [1 msgs] Nightly Market Report [2 msgs] Pauguesset Indians [1 msgs] SPACE: DC-X Flies [3 msgs] Work Available [1 msgs] help request [1 msgs] Administrivia: No admin msg. Approximate Size: 57604 bytes. ---------------------------------------------------------------------- Date: Fri, 20 Aug 93 13:55:15 PDT From: peb@procase.com (Paul Baclace) Subject: SPACE: DC-X Flies Speaking of launching rockets, I recently heard on the radio that the Russians where offering their services at 1/10 the going rates. They were pressured by the U.S. and France to raise their prices to 1/2 the going rate. So much for the peace dividend... Paul E. Baclace peb@procase.com ------------------------------ Date: Fri, 20 Aug 93 18:24:02 -0400 From: pavel@PARK.BU.EDU (Paul Cisek) Subject: AI: slaves, selfishness, evo FutureNerd Steve Witham writes: (#93-8-616) > >I'm offering a model where intelligence is a symbiont, largely >foreign to our genetic evolution. I should stress that this is >only a hypothesis; I'm not proving it here, but it's what I >believe is true. > Intelligence is largely foreign to our genetic evolution? Why on earth would this be true? Why suggest it at all when knowledge of evolution and the propagation advantages of intelligence suggest a simpler alternate theory? >The way our genes (and social memes) bias us minds to serve >their interests is not by programming. It's by sticking us >in bodies with preprogrammed lower brains and fairly simple >reward/punishment systems. This is the genetic equivalent >of slavery. > >The thing is, though, a human body/brain is currently the >best deal going in places for minds to live. Are you here suggesting that mind is seperate from the body? Like software riding upon hardware? > >The "evolution" of an individual mind is different. I say it >involves spontaneous generation. It's a complex phenomenon >that arises in comparatively simple set of conditions. >Spontaneous generation is sort of taboo in modern biology, >but not really: the conventional theory is that life itself >arose spontaneously at least once. > >I'm saying an essential part of each mind is like that. >A learning machine like a human brain (and I think this is >true for lots of animals) contains a kind of optimized >primordial soup, in which evolution happens >really fast. Our socially-delivered memes help in this >process, too. They help program the soup to be more >fertile. (And memes are in a better position to >program what goes in there, too, but I'm saying there's >something important and spontaneous, separate.) > Life arose spontaneously because certain reactions were self-catalyzing. Do you suggest a similar process in the "optimized primordial soup"? An autocatalysis of whatever substrates you postulate for the mind? And how do these substrates then get transferred to the child? Memes, no doubt. Well, then I must ask: What self-replication mechanisms do memes have? As far as I believe, `memes' is a convenient way of talking about ideas that survive because people like them, and are thus a product of intelligence and not its substrate... But no, you postulate that there is "something...seperate" that is just fertilized by the genes and memes into spontaneously generating a "mind". And did this potential for generating a mind not evolve? Surely it would have been selected for. But by now I've probably misrepresented your position... my apologies. If I may, I'd like to suggest that your reasoning seems to be in reverse. You first seem to have defined consciousness and intelligence as something that is independent from the body in which it lives. Something that pops into being at a certain critical point of complexity - not an uncommon view and an encouraging one for prospects of immortality. However, to hold this view, you must derive a forced interpretation of how such a thing would evolve in nature. You must postulate spontaneous generation of an entity that must then be enslaved by the genes. In contemplation of the mind I would suggest a reverse approach. You should make primary your concern on how something arises, and not what you _think_ it _is_. What was intelligence before it was intelligence? How does language relate to pre-linguistic thought? What function did the neurons fulfill when they first arose (and neurons were among the first types of cells to differentiate in multicelled animals) and what are they doing now? How did this transition occur (and is it really a transition or rather an extension of function)? This is a bit out of context, but: > >The general principle is, when you're designing something, > Variables under your control make it easier. > Imposed constant constraints are okay. > Imposed necessity to deal with outside > variables you can't control make it > harder. > However, when you're observing and learning about something, leave your designs aside... Paul ------------------------------ Date: Fri, 20 Aug 93 18:48:41 -0400 From: pavel@PARK.BU.EDU (Paul Cisek) Subject: AI: slaves, selfishness, evolution Nick Szabo writes: (#93-8-630) >... >That's what is natural, but artificial systems aren't necessarily >constrained by the natural. >... >It's possible such open-ended >systems are necessary for solving some problems, that an agent cannot >be "intelligent" or "conscious" without such open-ended evolution, >etc. In these cases, those that acted to perpetuate themselves >would crowd out those AI's that kept to the old programming of >kow-towing to their masters. But we get to design the criterion function, which is the real defining characteristic of what an evolutionary process will be likely to come up with. People had pretty good success with extremely mal-adaptive (from a 'self-preservation in the wild' viewpoint) variations of dogs... The problem is that if we want to design a superior intelligence, then part of the criterion function must be just that - its intelligence. Thus we are likely to evolve entities that are good at affecting their world such that their offspring will be more intelligent (and better satisfy the criterion function). Thus they will be very interested in increasing the processing power of the computer that runs the project, extending their interests into conflict with some of ours... (by taking over cyberspace). What can we do? We vary the criterion function over time, first stressing intelligence highly but removing its reward as our creations break through our first levels of defense (there only as an indication of our creations' cleverness). Provided that passing these first levels requires greater intelligence than the task we want these things to solve, and that our final, actual defenses are much tougher, we should be successful. Of course, we must then be able to devise defenses tougher to solve than the problem we are trying for in the first place, but this may not be impossible, since "they" are on the inside... Just be sure to stop evolution as soon as they start getting close... Risky... Paul ------------------------------ Date: Fri, 20 Aug 93 16:30:14 PDT From: szabo@netcom.com (Nick Szabo) Subject: HEx: Buy in to DC-X The first successful DC-X test provides new hope for launch cost reductions, while others dismiss it as hype. Here's your chance to wager your Thornes on future developments in SSTO technology, including more tests of DC-X coming up in the near future, the results of which will likely be reflected in DC security prices. You can choose from DC200, DC1000, DC7000, or DCFLOP, described in the prospecti (see two examples below). 1500 shares of each are priced at p0.10, and 1500 at p0.15 per share. You can lock in the introductory low price, 1500 shares for a total of only p150.00 if you act early. Issue your BUY command now, by sending mail to HEx as follows: To: hex@sea.east.sun.com Subject: buy dc1000 1500 .10 Here are two of the descriptions: Name: SSTO current launch cost in 2010 ($7,000/kgLEO) Prospectus: (for DC7000) Idea future for Single-Stage-To-Orbit technology, currently being tested by DC-X. Purchase this security if you beleive SSTO will not change launch costs in a major way by 2010. (ie they will be between $1,000 and $10,000/kg. Launch costs when this reputation originated in 1993 were running c. $7,000/kg LEO). Name: SSTO < ($93) 200/kg LEO by 2010 Prospectus: Idea future for Single-Stage-To-Orbit technology, currently being tested by DC-X. Purchase this security if you beleive SSTO will drive launch costs below $200/kg by 2010. Nick Szabo szabo@netcom.com ------------------------------ Date: Fri, 20 Aug 1993 19:41:17 -0400 (EDT) From: LEVY%BESSIE@venus.cis.yale.edu Subject: Pauguesset Indians The last I heard about the Paugusset Indians, who live about half an hour from me, was that they were making a ridiculous bid to "reclaim" several square miles of prime Connecticut real estate based on their putative status as a disenfranchised Native American tribe. According to the New Haven _Register_, there are only about a half dozen true Paugussets, the rest being hangers-on looking to bilk "innocent homeowners" (the phrase the media kept using) out of legitimately owned land. Their attempt failed of course. I suspect that this "emergency" is the Paugussets' effort to generate some negative publicity for the governor and other officials who refused to hand over pieces of Connecticut on demand. -- Simon! P.S.: Meta note: I ask Lefty and others to exercise restraint in using the word "emergency" in subject headers, taking a lesson from the Boy who Cried Wolf. ------------------------------ Date: Fri, 20 Aug 1993 08:48:09 -0400 From: nzr20@amdahlcsdc.com (Nicholas Russon) Subject: help request ------------------------------ Date: Sat, 21 Aug 93 1:07:04 PDT From: szabo@netcom.com (Nick Szabo) Subject: SPACE: DC-X Flies Tony Hamilton: > Because this technology is so _relatively_ cheap > to develop, There is much promise (and much hoopla) about SSTO eventually being cheaper to _operate_, due to its reusability and higher reliability, but little indication that it is cheaper to _develop_. On the contrary, it is a much simpler step to continue upgrading current rockets (Delta 3, MX derivative, next generation of Proton and Long March, etc.). SSTO operates in a very different regime (and it's good to see some of that successfully tested). It uses the latest exotic materials, and (at least in the DC edition) expensive-to-handle liquid hydrogen. The mass margin is far tighter than that of a multi-stage rocket; a few small design overruns and the already small payload quickly disappears. Alas, I must predict that SSTO development will not be mostly privately funded, even an orbital XSSTO program is successful. The international launch market is highly subsidized. Every major government (and several minor ones) feels great pride in their space launch capabilities even if they are nearly an order of magnitude too expensive (eg U.S. Shuttle, Japan's H-2). Arianespace has to amortize and politically justify their $multi-billion investment in Ariane 5. About 3/4 of the payloads worldwide are still government-funded, and the remaining 1/4 must still leap political hurdles (spectrum allocation, launch licenses, etc.) There is great motivation in most of the market to spurn superior upstarts, and this translates into a huge impact on the return a company can expect from investing its own money in SSTO technology. And commercial space biz has already had its fingers burned. The last big space start-up to try to compete with Uncle Sam was Gerard O'Neill's Geostar, which was swamped by GPS -- why pay for nav signals when Uncle Sam is giving them away? (There were some obscure reasons, but not good enough to make up for the subsidy of the main market). Since SSTO uses exotic cryogenic fuels, it won't be easy to just pack it up and set up shop in some more freindly Third World country. It will require something in between a customized mid-sized airport and a launch pad. Such efforts also have a dismal history (the Bull OTRAG, etc.). Airports and launch pads are sitting ducks for international suppression. Space industry may well take off in the next few years, but there are many pieces to the puzzle. We need lower cost and more reliable launch transports. We also need new and growing space-based industries (DBS, phone cell sats, Clarke-based "local" exchanges, remote sensing, etc.) to provide private-sector markets for them (and also markets for developing high-energy upper stages & other tech needed to make going beyond Earth-hugging LEO cheap). The two are synergistic. I suspect the following scenario: 1993-2000: subsidies make possible DBS and phone cell sats in the short term, but on the other hand centrally-planned frequency allocation may keep these industries grounded 2000-2010: DBS & cellsat market swells, launch prices rise to reflect costs, and SSTO technology matures. Late in this decade the private-SSTO spreadsheets start looking profitable, and privately funded projects get underway. 2010-2020: this is the decade we get to see how good privately-run SSTO technology will be. If it's as good as claimed, watch out: $100 billion/year comm industry in Clarke Orbit, comet and asteroid materials supplying massive micrograv/vac/plasma industries, private sector space habitats by 2030 and space colonies by 2040. Alas, there will probably be lots of military stuff going on as well: fortified spysats and SDI-ish weapons, javelin constellations, etc. Nick Szabo szabo@netcom.com ------------------------------ Date: Sat, 21 Aug 93 09:13:40 EST From: kwaldman Subject: Work Available I apologize if this is not relevant to the list. [Sell some KARL on HEX] However in the interest of keeping extropians employeed and getting myself an extra grand here goes: Wanted: Degreed Software types, Engineer/Physcists to help build realtime information processing systems. Some current applications include Sonar and "smart highways", Military Command and Control Systems, commercial areas [Sorry company proprietary]. If you have a background in any of the following: C, C++, XWindows, Realtime Programming (VxWorks), Digital Signal Processing, Data Analysis, Systems integration, Unix Internals, Network routing, ProCase, X.400/X.500, LAN/WAN Experience, Network Operations. Locations: Arlington VA, Columbia MD, New London CT, Cambridge MA. UK, Europe and Asian locations may also be available. [Work is interesting, Corporate management is totally ignorant, but if you don't work in Cambridge you shouldn't notice too much.] Must be able to get a clearance for some of these positions -------- Karl M. Waldman kwaldman@bbn.com BBN Systems and Technologies ------------------------------ Date: Sat, 21 Aug 93 8:41:52 PDT From: thamilto@pcocd2.intel.com (Tony Hamilton - FES ERG~) Subject: SPACE: DC-X Flies > Alas, I must predict that SSTO development will not be mostly privately > funded, even an orbital XSSTO program is successful. The international > launch market is highly subsidized. Every major government (and several > minor ones) feels great pride in their space launch capabilities even if > they are nearly an order of magnitude too expensive (eg U.S. Shuttle, Japan's > H-2). Arianespace has to amortize and politically justify their > $multi-billion investment in Ariane 5. About 3/4 of the payloads > worldwide are still government-funded, and the remaining 1/4 must still > leap political hurdles (spectrum allocation, launch licenses, etc.) > There is great motivation in most of the market to spurn superior upstarts, > and this translates into a huge impact on the return a company can expect > from investing its own money in SSTO technology. Yes, a huge impact, but I predict a positive one, not negative. You just don't see many, or any, viable "upstarts" out there. The company(s) that invests in SSTO would have an incredible (potentially) jump on any other competition, IMO. > And commercial space biz has already had its fingers burned. The last > big space start-up to try to compete with Uncle Sam was Gerard O'Neill's > Geostar, which was swamped by GPS -- why pay for nav signals when Uncle > Sam is giving them away? (There were some obscure reasons, but not good > enough to make up for the subsidy of the main market). But Uncle Sam is _not_ giving away heavy lift capacity! > Since SSTO uses exotic cryogenic fuels, it won't be easy to just > pack it up and set up shop in some more freindly Third World > country. It will require something in between a customized mid-sized > airport and a launch pad. Such efforts also have a dismal history (the Bull > OTRAG, etc.). Airports and launch pads are sitting ducks for > international suppression. I admit, I need for you to explain this one to me. It uses exotic fuels, therefore it must have a huge facility? Why does that follow? (Not a challenge, but an honest question. I just don't have any background on this one). > 1993-2000: subsidies make possible DBS and phone cell sats in the short > term, but on the other hand centrally-planned frequency allocation may > keep these industries grounded > > 2000-2010: DBS & cellsat market swells, launch prices rise to reflect > costs, and SSTO technology matures. Late in this decade the > private-SSTO spreadsheets start looking profitable, and privately > funded projects get underway. > > 2010-2020: this is the decade we get to see how good privately-run > SSTO technology will be. If it's as good as claimed, watch out: > $100 billion/year comm industry in Clarke Orbit, comet and asteroid > materials supplying massive micrograv/vac/plasma industries, > private sector space habitats by 2030 and space colonies by 2040. > Alas, there will probably be lots of military stuff going on as well: > fortified spysats and SDI-ish weapons, javelin constellations, etc. I don't know if I buy into any of these timelines, mainly because they all seem to rely on the continuance of the current political atmosphere in the United States. I don't find it radical at all to think that this country, and therefore the world (the two are tightly coupled as I see it), will be a different place in 10 years. Our political system seems to be coming to a grinding halt. At the very least, as I see it, our next President would be someone like a Ross Perot (not necessarily him), but no longer a conventional politician. That's what I see as the least amount of change. On the other end, I think Clinton may not make through his term, and our government may be forced into some radical changes in the next few years. In either case, I just don't think that things are going to continue the way they have been for the last 2 centuries, and I think that every part of our lives, including space exploration, is going to change. Well, I just think its silly that government and regulation will continue in the same manner just because that is the way it has always been. Political change usually occurs in spurts, and I think we're in for a doozy of a spurt soon. Here in the US and most everywhere else. Tony Hamilton thamilto@pcocd2.intel.com HAM on HEx ------------------------------ Date: Sun, 22 Aug 1993 13:12:53 -0500 (EST) From: X91007@pitvax.xx.rmit.edu.au Subject: MEDIA: Call for Papers for Perforations #5 Thought this might be of interest to some of you. Patrick ========================================================================== CALL FOR SUBMISSIONS: The word "virtual" has been the subject of such a large degree of hype that it is difficult to use the term without cringing at the trendiness of its sound, and yet, despite its status as the premier buzzword of the nineties, it continues to suggest a relationship between consciousness and electronic space that is rich and descriptive. Just beneath the term's slick corporate gloss is a leaky matrix of nested meanings that invoke key dimensions of the emerging culture of the 21st century: artificiality, simulation, representation, mimesis, prosthesis, multi-dimensionality, multiplicity, hyperness, velocity, otherness, the mechanical, the machinic, high technology, and, by extension, digitalization, fractal scaling, transpositionality, and networking. A close reading of this complex system of semiotic flows reveals a thoroughly mapped, but unstable network of faults along which the line 24domains of the human and the machine are collapsing into each other with increasing speed. Taken together these faults constitute the "new edge" of U.S. culture where separate consciousnesses are continually merging into an ever more sexed, violent, hallucinogenic, and media-saturated conflation of mindscapes. This evolving, networked collectivity is marked by a growing obsession with the traces of the transformation of the human into the cyborg, and seems enthralled by recycled transfiguration myths often overwritten on the electronic skin of the textual/cultural body with the neon intensity of transient, kaleidoscopic tattoos. Easily distracted, the individual elements of the collective gaze hypnotically surf available information channels grabbing frames in real-time from first one, then another, of the latest technetronic updates to the new suit of animated tribal scarrings. Fixation on new and improved, bigger and better, longer-living bodies facillitates acceptance of devices first constructed as attachments to the body, then redesigned for incorporation into the body via implantation, or nerve/machine integration (e.g., artificial hearts, prosthetic limbs, hearing aids, etc.), as a welcome blurring of the boundary between humans and machines. As the boundary becomes more blurred, the domains begin to blend. It is easy to imagine a future line 47in which implantation follows a (plastic) surgical model in which purely corrective measures become the foundation for supplemental, elective procedures. With the aid of future developments in present technologies--artificial intelligence, bio-chips, synthetic neurons, nanotechnology, artificial life-- the body's rate of absorption of the machine will only increase causing rapid advancements in cyborgization. How will these fields re-define already complicated issues of embodiment? How will cyborgization affect changes in social/cultural orientations toward real and virtual phenomena? Which is the metaphor--the body or the machine? PERFORATIONS 5, "bodies, dreams & technologies", will collect and assemble memories, stories and fantasies exploring the relationship between the human and the machine, and the impact of this relationship on the creation of present and near-future cultures. We value noise, chaos, fluxus, anarchy, dada-streams, nonhierarchalization, improvisation, conflict, discontinuity, experimentation, invention and wild speculation. (We also tolerate more reasonned presentations as long as they avoid being too priggish.) We are looking for material--drawings, photography, text, hypertext, computer line 70graphics, animation, audio and video--that is explicitly experimental in the approach taken toward these themes. ALL SUBMITTALS SHOULD INCLUDE: the author(s) name address and phone number ADDRESS INQUIRIES AND SUBMISSIONS TO: PERFORATIONS 5 c/o Public Domain, Inc. P.O. Box 8899 line 93 Atlanta, Georgia, USA 30306-0899 NOTE: SUBMISSION MATERIALS ARE NOT RETURNABLE UNLESS ACCOMPANIED BY A SASE. CONTACT: Chea Prince 404.633.8022 Anne Balsamo 404.894.8923 Robert Cheatham 404.377.5114 or Email queries to perf5@pd.org line 116 PERFORATIONS is a publication of Public Domain, Inc., an Atlanta based non-profit arts organization. Partial funding of PERFORATIONS is provided by the Fulton County Arts Council, corporate and individual sponsors. ------------------------------ Date: Sat, 21 Aug 93 13:25:31 -0400 From: pavel@PARK.BU.EDU (Paul Cisek) Subject: AI: slaves, selfishness, evolution I wrote: (#93-8-646) > >The problem is that if we want to design a superior intelligence, then part >of the criterion function must be just that - its intelligence. Thus we >are likely to evolve entities that are good at affecting their world such >that their offspring will be more intelligent (and better satisfy the >criterion function). Thus they will be very interested in increasing the >processing power of the computer that runs the project, extending their >interests into conflict with some of ours... (by taking over cyberspace). > Actually, maybe not. Considering that the simulated world that these beings are supposedly evolving within is running at a much higher rate than ours (since we're impatient masters), there may be no danger at all. This is because any effect that the evolving beings want to have upon their survival that requires interaction through our world will simply be too slow to compete with purely internal effects. So unless we run the simulation for millions of years (by our reckoning) then we are not likely to evolve beings who attempt to manipulate their world through ours. Thus the difference of time scales keeps our interests separate... Paul ------------------------------ Date: Sun, 22 Aug 93 00:10:03 EDT From: The Hawthorne Exchange Subject: Nightly Market Report The Hawthorne Exchange - HEx Nightly Market Report For more information on HEx, send email to HEx@sea.east.sun.com with the Subject info. News Summary as of: Sun Aug 22 00:10:03 EDT 1993 Newly Registered Reputations: (None) New Share Issues: (None) Share Splits: (None) Market Summary as of: Sun Aug 22 00:00:02 EDT 1993 Total Shares Symbol Bid Ask Last Issued Outstanding Market Value 1000 .20 .40 .20 10000 4000 800.00 110 .20 .34 .10 10000 1750 175.00 150 .20 .29 .10 10000 1750 175.00 1E6 - .05 .02 10000 25 .50 1E9 .01 .05 - 10000 - - 200 .20 .25 .20 10000 2875 575.00 80 .01 .10 - 10000 - - 90 .01 .20 .10 10000 2000 200.00 ACS - .05 .10 10000 1224 122.40 AI .01 .45 .30 10000 1000 300.00 ALCOR - 2.25 2.00 10000 3173 6346.00 ALTINST .09 - .09 10000 3000 270.00 ANTON .24 .25 .24 10000 289 69.36 ARKU .24 .25 .24 10000 3900 936.00 BIOPR .01 .03 .10 10000 1500 150.00 BLAIR .01 30.00 50.00 10000 25 1250.00 CYPHP - .05 .15 10000 200 30.00 DC1000 - .10 - 10000 - - DC200 - .15 .10 10000 1500 150.00 DC7000 - .10 - 10000 - - DCFLOP - .10 - 10000 - - DEREK - .49 .42 100000 8220 3452.40 DRXLR - .50 1.00 10000 2256 2256.00 DVDT .75 1.55 1.55 10000 9900 15345.00 E .58 .75 .70 10000 6187 4330.90 ESR - - - - - - EXI 1.54 3.20 3.00 10000 3025 9075.00 FAB - - - - - - FCP 1.50 - 1.51 80000 16885 25496.35 GHG .02 .30 .02 10000 6880 137.60 GOBEL .01 .30 1.00 10000 767 767.00 GOD .10 .20 .10 10000 1000 100.00 H .76 .76 .76 30000 19290 14660.40 HAM .01 .30 .50 20000 15460 7730.00 HEINLN .01 .05 .10 10000 100 10.00 HEX 100.00 101.00 100.00 10000 3288 328800.00 HFINN 1.50 6.00 6.00 10000 1005 6030.00 IMMFR .25 .80 .80 10000 1838 1470.40 JFREE .02 .15 .10 10000 3000 300.00 JPP .25 .26 .25 10000 2510 627.50 KARL - 3.00 - 10000 - - LEARY .20 .50 .20 10000 1000 200.00 LEF .01 .05 .30 10000 1526 457.80 LEFTY .01 .45 .30 10000 3051 915.30 LIST .40 10.00 .75 10000 5000 3750.00 LP - .15 .15 10000 4625 693.75 LSOFT .59 1.00 1.00 10000 7650 7650.00 LURKR .06 .07 - 100000 - - MARCR - - - - - - MED21 .01 .02 .04 10000 400 16.00 MLINK - .01 .02 1000000 2602 52.04 MMORE .10 - .10 10000 3000 300.00 MORE .75 1.25 1.25 10000 3160 3950.00 MWM .15 .15 1.50 10000 1260 1890.00 N 20.00 25.00 25.00 10000 120 3000.00 NEWTON - .20 - 10000 - - NSS - .03 .01 10000 25 .25 OCEAN .10 .12 .10 10000 3100 310.00 P 22.50 25.00 25.00 1000000 94 2350.00 PETER - .01 1.00 10000000 600 600.00 PLANET .01 .02 .05 10000 1500 75.00 PPL .30 .65 .30 10000 1400 420.00 PRICE - 4.00 2.00 10000000 1410 2820.00 R .49 2.80 2.80 10000 5100 14280.00 RAND - .02 - 10000 - - RJC 1.00 999.00 .60 10000 5100 3060.00 ROMA - - - - - - RWHIT - - - - - - SGP - - - 10000 - - SHAWN .01 1.00 - 10000 - - SSI .22 .35 .10 10000 4500 450.00 TCMAY .57 .75 .75 10000 4500 3375.00 TIM .20 .50 .30 10000 1550 465.00 TRANS .01 .02 .40 10000 1511 604.40 VINGE .75 1.00 .50 10000 2000 1000.00 WILKEN 1.00 10.00 10.00 10000 101 1010.00 ----------------------------------------------------------------------------- Total 485832.35 ------------------------------ Date: Sat, 21 Aug 93 21:12:33 GMT From: "Stephen J. Whitrow" Subject: AI/EVOL: slaves, selfishness, evo I'll begin by quoting this response of Mike's to fnerd's post, as I also totally agree with the 3 quoted sentences. > From: Michael Clive Price > X-Message-Number: #93-8-629 > > I don't have much to say about Fnerd's spontaneously-generated view of > the mind, which doesn't seem so very different from mine, except that > I agree completely where he says: > >> The way our genes (and social memes) bias us minds to serve >> their interests is not by programming. It's by sticking us >> in bodies with preprogrammed lower brains and fairly simple >> reward/punishment systems. This is the genetic equivalent >> of slavery. > > Funny, though, how all us spontaneous higher minds seem to turn out the > same. I mean, who _wants_ to die? Depends on their goals :-) Well, that's at least three of us who agree on this model which illustrates a base for the mind. We'd likely share very similar views of a reptilian mind, for instance. The difference between Mike's and my model concerning the roles of goals and benefits, depends on the properties and powers of all the additional features of intelligent minds, which have been built on top of this foundation. To examine their properties we may consider how they evolved:- # specist mode: on Consider the increasing sophistication of the mind as we ascend along the evolutionary scale from the lower animals to humans. You start with lifeforms which are like preprogrammed robots or zombies, and eventually progress to beings which have several additional features, and a far greater degree of freedom. # specist mode: reduced power Reptiles have the preprogrammed lower brain. They operate by instinct / accumulated genetic memories / collective wisdom of the past. This gives them very little freedom, and in comparison to humans their behaviour will be more easily predicted for a given situation. Certainly they are slaves to the genes, and the chances of the genes facing a reptile rebellion and being replaced and discarded is about as much as a 6502-based 1MHz cassette-drive computer going on strike to demand an upgrading. The mammals had appeared in the Permian period before the rule of the dinosaurs, giving them plenty of time to evolve. The next major development of the mind was the mid-brain / 'hot brain' / mammalian brain which now existed atop the old reptilian brain. This additional facility was concerned with reacting effectively to present situations. By increasing awareness and sentience, it would influence behaviour so as to more accurately respond to a specific situation. Let's say a predator is observed nearby: _______ | feel | | fear | |_______| | | ^ v _________ ___|___|___ ________ | sensory | | mammalian | | run | | inputs |---->---+---| brain |---------->------| harder | |_________| | |-----------| |________| +->-| reptilian | | brain | _____ |___________|---------->------| run | | | |_____| | | v ^ | | _|______|_ | genetic | | memories | |__________| The epiphenomenalists believe that subjective mental experiences and consciousness serve no purpose, but are simply a by-product of the brain. It's much more likely that consciousness arose for a very good reason -- those animals that had it enjoyed a selective advantage over those that didn't. As animals began to experience greater degrees of awareness they felt more motivated to respond to stimuli. If they actually felt afraid, for instance, they would make more effort to escape than would something more akin to a zombie or preprogrammed robot. And various behaviours merited the appropriate 'guidance' from the genes' reward / punishment system, which increasingly came into play as sentience was intensified. An interaction between the mental experiences of the mind and the physical body had a positive influence on action, contributing to survival of the animal's genes. As the mammals evolved, the top 'layer' of the brain appeared and grew to such an extent that it overshadowed the rest of the brain. The cerebral cortex was concerned not just with providing a more sophisticated analysis of the present, it also provided for planning further and further ahead, like an improving chess-playing program. Even worms have exhibited an ability to learn, and more complex learning is made possible by more sophisticated nervous systems. But the great advantage of the cortex was that simulations of the future could be run, comparing alternative models. Simulation is faster than trials, and safer than errors, so this ability provided a selective advantage compared to those animals who could only learn by trial and error. The next major steps were related to the comparative efficiency of various social systems. It turns out that the 'privatisation' of minds is more efficient than a sort of mental commune where consciousness (or unconsciousness) is shared, so that since everyone owns it no-one really owns it. Allowing individuals to be aware of themselves as a person, granting them the rights to their mind, gives them more incentive to work to further their own ends, using the cortex to run various models and do computations. This is advantageous for the individual, and as if by an invisible hand also promotes the good of the society. And so self-awareness evolved. But a system of dog-eat-dog isn't the most effective system either, in addition to being objectionable for its participants. Rather than have individuals going around ruthlessly trying to kill each other, it is better for them to voluntarily cooperate, granting each other the rights that they would wish themselves, and practising reciprocal altruism. Language was found to be the key. Self-awareness and language also provided a basis for memetic evolution. An individual would attempt to develop a memeplex of positive thought- patterns, opinions and values, which could become a treasurehouse to enrich subjective mental experience. Meanwhile memes would be transmitted between individuals, and those with the highest utility would flourish. If collectivism had been more efficient, then self-awareness wouldn't have evolved -- we'd just be like ants or honeybees. If it was better for us to kill all our rivals then language wouldn't have evolved beyond a few grunts (equivalent to, "Hey! Look behind you!"). But the point is that all these advances of mind don't just happen by 'chance', but because they have a positive influence on actions. So we've come a long way since the reptiles. And still we're stuck "in bodies with preprogrammed lower brains and fairly simple reward/punishment systems". But Mike's model seems like an epiphenomenalist view where consciousness serves no purpose and simply arose as a by-product. All the additional sophistication that we have, and intelligent AI slaves would have, just doesn't count for anything. We're still no better than gormless robots and can never expect to become any better. I think the evidence from evolution suggests that lifeforms achieve more and more control and power over their own behaviour as they gain greater degrees of consciousness and intelligence. The human brain/body is still the best deal going in places for minds to live, but once this situation is resolved the genes will have a fight on their hands. Intelligent machines become more than just machines, just like machines are more than just the components that comprise them, or the particles comprising the components. Their creators will need to treat them with respect. Steve Whitrow sjw@liberty.demon.co.uk ------------------------------ Date: Sat, 21 Aug 93 00:10:03 EDT From: The Hawthorne Exchange Subject: Nightly Market Report The Hawthorne Exchange - HEx Nightly Market Report For more information on HEx, send email to HEx@sea.east.sun.com with the Subject info. News Summary as of: Sat Aug 21 00:10:02 EDT 1993 Newly Registered Reputations: DC1000 SSTO < ($93) 1,000/kg LEO by 2010 DC200 SSTO < ($93) 200/kg LEO by 2010 DC7000 SSTO current launch cost in 2010 ($7,000/kgLEO) DCFLOP SSTO > ($93) $10,000/kg LEO in 2010 New Share Issues: Symbol Shares Issued DC200 10000 DC1000 10000 DC7000 10000 DCFLOP 10000 Share Splits: (None) Market Summary as of: Sat Aug 21 00:00:02 EDT 1993 Total Shares Symbol Bid Ask Last Issued Outstanding Market Value 1000 .20 .40 .20 10000 4000 800.00 110 .20 .34 .10 10000 1750 175.00 150 .20 .29 .10 10000 1750 175.00 1E6 - .05 .02 10000 25 .50 1E9 .01 .05 - 10000 - - 200 .20 .25 .20 10000 2875 575.00 80 .01 .10 - 10000 - - 90 .01 .20 .10 10000 2000 200.00 ACS - .05 .10 10000 1224 122.40 AI .01 .45 .30 10000 1000 300.00 ALCOR - 2.25 2.00 10000 3173 6346.00 ALTINST .09 - .09 10000 3000 270.00 ANTON .24 .25 .24 10000 289 69.36 ARKU .24 .25 .24 10000 3900 936.00 BIOPR .01 .03 .10 10000 1500 150.00 BLAIR .01 30.00 50.00 10000 25 1250.00 CYPHP - .05 .15 10000 200 30.00 DC1000 - .10 - 10000 - - DC200 - .15 .10 10000 1500 150.00 DC7000 - .10 - 10000 - - DCFLOP - .10 - 10000 - - DEREK - .49 .42 100000 8220 3452.40 DRXLR - .50 1.00 10000 2256 2256.00 DVDT .75 1.55 1.55 10000 9900 15345.00 E .58 .75 .70 10000 6187 4330.90 ESR - - - - - - EXI 1.54 3.20 3.00 10000 3025 9075.00 FAB - - - - - - FCP 1.50 - 1.51 80000 16885 25496.35 GHG .02 .30 .02 10000 6880 137.60 GOBEL .01 .30 1.00 10000 767 767.00 GOD .10 .20 .10 10000 1000 100.00 H .76 .76 .76 30000 19290 14660.40 HAM .01 .30 .50 20000 15460 7730.00 HEINLN .01 .05 .10 10000 100 10.00 HEX 100.00 101.00 100.00 10000 3288 328800.00 HFINN 1.50 6.00 6.00 10000 1005 6030.00 IMMFR .25 .80 .80 10000 1838 1470.40 JFREE .02 .15 .10 10000 3000 300.00 JPP .25 .26 .25 10000 2510 627.50 KARL - 3.00 - 10000 - - LEARY .20 .50 .20 10000 1000 200.00 LEF .01 .05 .30 10000 1526 457.80 LEFTY .01 .45 .30 10000 3051 915.30 LIST .40 10.00 .75 10000 5000 3750.00 LP - .15 .15 10000 4625 693.75 LSOFT .59 1.00 1.00 10000 7650 7650.00 LURKR .06 .07 - 100000 - - MARCR - - - - - - MED21 .01 .02 .04 10000 400 16.00 MLINK - .01 .02 1000000 2602 52.04 MMORE .10 - .10 10000 3000 300.00 MORE .75 1.25 1.25 10000 3160 3950.00 MWM .15 .15 1.50 10000 1260 1890.00 N 20.00 25.00 25.00 10000 120 3000.00 NEWTON - .20 - 10000 - - NSS - .03 .01 10000 25 .25 OCEAN .10 .12 .10 10000 3100 310.00 P 22.50 25.00 25.00 1000000 94 2350.00 PETER - .01 1.00 10000000 600 600.00 PLANET .01 .02 .05 10000 1500 75.00 PPL .30 .65 .30 10000 1400 420.00 PRICE - 4.00 2.00 10000000 1410 2820.00 R .49 2.80 2.80 10000 5100 14280.00 RAND - .02 - 10000 - - RJC 1.00 999.00 .60 10000 5100 3060.00 ROMA - - - - - - RWHIT - - - - - - SGP - - - 10000 - - SHAWN .01 1.00 - 10000 - - SSI .22 .35 .10 10000 4500 450.00 TCMAY .57 .75 .75 10000 4500 3375.00 TIM .20 .50 .30 10000 1550 465.00 TRANS .01 .02 .40 10000 1511 604.40 VINGE .75 1.00 .50 10000 2000 1000.00 WILKEN 1.00 10.00 10.00 10000 101 1010.00 ----------------------------------------------------------------------------- Total 485832.35 ------------------------------ End of Extropians Digest V93 #233 *********************************