tag:blogger.com,1999:blog-8587336.post1394470915066762961..comments2024-03-28T18:18:37.133-07:00Comments on CONTRARY BRIN: More cool, weird, worrisome "stuffs"...David Brinhttp://www.blogger.com/profile/14465315130418506525noreply@blogger.comBlogger25125tag:blogger.com,1999:blog-8587336.post-57003726339222432792007-05-30T14:07:00.000-07:002007-05-30T14:07:00.000-07:00To randomsequences...Evolution is not "optimal", i...To randomsequences...<BR/><BR/>Evolution is not "optimal", it is merely good enough for prevailing conditions and competitions. Nor does evolution search the "entire space", it is subject to significant founder effects.<BR/><BR/>Now whether human foresight and molecular intervention can outperform the tic-tac-toe chicken of evolution to generate a more special species is another question altogether.Enterikhttps://www.blogger.com/profile/04758515647778280562noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-1401867181983625732007-05-30T13:56:00.000-07:002007-05-30T13:56:00.000-07:00I haven't followed the arc of the entire neutral/g...I haven't followed the arc of the entire neutral/good genetics improvement conversation, but that won't stop me from jumping in. In the current state of affairs is should be possible to screen all in vitro fertilize embryos for the worst of alleles; phenylketonuria, hereditary nonpolyposis colorectal cancer, Factor V Leiden thrombophilia, et cetera and thereby deplete humanities genetic stock of deleterious genes. A substantial reduction in genetically caused morbidity is well within our reach, no fiction required...Enterikhttps://www.blogger.com/profile/04758515647778280562noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-58771901225379113282007-05-16T03:00:00.000-07:002007-05-16T03:00:00.000-07:00Slightly OT here; on the perennial "Why Johnny can...Slightly OT here; on the perennial "Why Johnny can't code" topic -- whythelickystiff (of why's [poignant] guide fame) has launched <A HREF="http://hacketyhack.net/" REL="nofollow">HacketyHack</A>, to teach Ruby to kids. Never mind just moving pixels around, it's geared to doing things like blogs and IM systems -- the sort of things that they'd actually be using for real.Steve Gilhamhttps://www.blogger.com/profile/03622573187942388226noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-29723419817145780902007-05-04T09:42:00.000-07:002007-05-04T09:42:00.000-07:00Re. the Internet archive item: it's not so much ho...Re. the Internet archive item: it's not so much holding the program liable--the program isn't going to have to pay monetary damages--it's holding the person, corporation, or partnership liable because the program is acting as its agent. The closets analogy would be if the people at the Internet Archive told one of its employees "go copy that web site unless it has a robots.txt file." The Internet Archive would be the one on the hook because the person doing the copying was acting as the agent of the Internet Archive, following its instructions.<BR/><BR/>The real question here is more about whether a contract formed. For instance, there's already a standard method of talking to spiders which the web site's owner chose not to follow. So should you put the burden on the Internet Archive for not building enough AI into its spider (or running the software int he first place), or should you put it on the web site's owner for not knowing all the standard ways to talk to things that might want to crawl the site?<BR/><BR/>This <A HREF="http://blog.ericgoldman.org/archives/2007/03/can_a_spider_en.htm" REL="nofollow">post</A> points out a second aspect: the law that the court makes in this case might apply to click-through agreements (and their close relatives) that apply to humans as well as to spiders. The question there becomes how much do you have to do before it becomes a binding agreement?<BR/><BR/>Re. the Chery QQ: cute car, but check the safety stats before buying. There's a post <A HREF="http://paultan.org/archives/2006/02/18/chery-qq-crash-test/" REL="nofollow">here</A> showing photos of one crash test.False Datahttps://www.blogger.com/profile/07901490109393153935noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-509617410811378472007-05-04T09:09:00.000-07:002007-05-04T09:09:00.000-07:00Rob Perkins,Um, yes it does. In parallel even. Nat...Rob Perkins,<BR/><BR/>Um, yes it does. In parallel even. Natural selection is a genetic search algorithm - that's where engineers got the idea from. Do you have a better analytic word, Rob? (Other than um)<BR/><BR/>Doug S: Hard to know as of yet. We've been in a population boom for the last few centuries, so primarily one side of the evolutionary paradigm has been in action: propagation. But the selection part has not been as active. If we survive the next century or so, populations (at least planetary) are going to have to saturate. If we don't avoid a die-back (knock on wood), then selective pressures will be very high.<BR/><BR/>Don't you remember the old joke about the meeting of the gorilla evolutionary society which designed the next step in gorilla evolution? Bigger jaws and larger gray spots on silverbacks of course!RandomSequencehttps://www.blogger.com/profile/12259854206507818658noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-779681632839091512007-05-04T06:35:00.000-07:002007-05-04T06:35:00.000-07:00If the retrocausality experiment works, public awa...If the retrocausality experiment works, public awareness of it could have <A HREF="http://www.nature.com/news/2005/050704/pf/436150a_pf.html" REL="nofollow">some undesirable consequences.</A>Nick Tarletonhttps://www.blogger.com/profile/16226303578512704501noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-32323098770493411552007-05-04T03:25:00.000-07:002007-05-04T03:25:00.000-07:00Um... evolution might not get us where we want to ...Um... evolution might not get us where we want to go. Remember the story <A HREF="http://en.wikipedia.org/wiki/The_Marching_Morons" REL="nofollow">The Marching Morons</A>? IQ and fertility are negatively correlated - <A HREF="http://www.crispian.demon.co.uk/LynnPINC.htm" REL="nofollow">stupid people really are having more children than smart people!</A><BR/><BR/>- Doug S.Doug S.https://www.blogger.com/profile/11918949543315280580noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-35252897176399985032007-05-03T22:58:00.000-07:002007-05-03T22:58:00.000-07:00"You just have to search the entire space, like ev..."You just have to search the entire space, like evolution does, and evolution has already brought us pretty close to optimal in the first place."<BR/><BR/>Um... evolution doesn't "search".Rob Perkinshttps://www.blogger.com/profile/13115249244056328076noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-69516661608252261532007-05-03T22:47:00.000-07:002007-05-03T22:47:00.000-07:00Thanks, HH. I read it once in the long ago but the...Thanks, HH. I read it once in the long ago but the inner google doesn't recall any details (other than that it was pretty forgettable)<BR/>What you describe sounds like the premise used in 'Gattaca' (the gene bit, not the arms)Tony Fiskhttps://www.blogger.com/profile/14578160528746657971noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-75595377914940788442007-05-03T22:40:00.000-07:002007-05-03T22:40:00.000-07:00Regarding DARPA's BICA project: funding cut comes ...Regarding DARPA's BICA project: funding cut comes as no surprise, without a built in ROI (i.e. business model). A project along similar lines, only with the ROI built-in (not to mention investment-grade management in the person of Jeff Hawkins, inventor of PalmOS) has been built around this concept of HTM (Hierarchial Temporal Memory), which is based on what sounds like good congitive science (to this lay person, anyway) and described in such terms as most of us can understand by JH himself at http://www.spectrum.ieee.org/print/4982 . Follow the links back to source (& download the sources (in Python) even), to build your own HTM. Fun, wow!Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-8587336.post-77112683696475731112007-05-03T22:05:00.000-07:002007-05-03T22:05:00.000-07:00Actually, I've never read Heinlein, but it is almo...Actually, I've never read Heinlein, but it is almost impossible these days to have a truly original thought. :-)<BR/><BR/>I once invented hash tables, too, but turns out it had been done before. Oh well.Xactiphynhttps://www.blogger.com/profile/08254344563346437079noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-5999468245059507402007-05-03T21:29:00.000-07:002007-05-03T21:29:00.000-07:00TonySounds like "Beyond This Horizon", a story wit...Tony<BR/><BR/>Sounds like "Beyond This Horizon", a story with (in the backround) a socialistic paradise, with pre-conception genetic selection to get 'the best possible offspring', and every adult male armed and ready to fight over the slightest insult.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-8587336.post-28535132237215310172007-05-03T20:59:00.000-07:002007-05-03T20:59:00.000-07:00Um, of which Heinlein tale are we speaking?Harking...Um, of which Heinlein tale are we speaking?<BR/><BR/>Harking back to the last post: <A HREF="http://www.planetary.org/blog/article/00000960/" REL="nofollow">COROT bags its first planet</A> (I gather 1500 light years is something of a record)<BR/><BR/>And if you're worried about bees, beware the return of the <A HREF="http://environment.newscientist.com/channel/earth/mg19425983.700-billions-at-risk-from-wheat-superblight.html" REL="nofollow">Death of Grass</A>. <BR/><BR/>I'm combining a couple of David's points to make a cynical prediction:<BR/><I>...ruling that the software was effectively practicing law without a license. (If this kind of guild protection racket steams you, look into Project HALT or Americans for Legal Reform.</I><BR/>and:<BR/><I>...the solar roof and the usable battery car seem to be unstoppable, now. Time to start taking the names of those who delayed this.</I><BR/><BR/>So, when does some bright spark decide to put up legislation requiring you to be licensed as a utility company before you can generate your own power?Tony Fiskhttps://www.blogger.com/profile/14578160528746657971noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-16498038909732372142007-05-03T20:18:00.000-07:002007-05-03T20:18:00.000-07:00Nope, David, because of the very nonlinearity you ...Nope, David, because of the very nonlinearity you point out. A gene might not only increase the probability of intelligence, but also the probability of retardation, for example. You just have to search the entire space, like evolution does, and evolution has already brought us pretty close to optimal in the first place.<BR/><BR/>Lots of people will try - most biologist don't understand non-linear systems, even thought they work with them (but then again, most people go into biology because they failed calculus). But this is like trying to back-engineer a system composed of multiple amplifiers designed by an alien - you're best off not trying. <BR/><BR/>A genetic algorithm is most likely to make improvements - and the beautiful thing is, we <EM>already</EM> have one! So why bother with the tinkering, other than to avoid the most obvious errors contained in the structural units. We're already mostly clones at that level anyhow, with a bit of variation for ancestral geographic adaptations.RandomSequencehttps://www.blogger.com/profile/12259854206507818658noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-70774209705144603282007-05-03T20:03:00.000-07:002007-05-03T20:03:00.000-07:00Actually, Random, I quite agree with you. Jurassi...Actually, Random, I quite agree with you. Jurassic Park can't happen even if you get perfect dupes of TRex genes, because the code is only half of it. The other half is all the "readers"... the ribosomes and the Egg itself, all of which carry quirky programs.<BR/><BR/>See how I deal with all this nonlinearity in EARTH!<BR/><BR/>And yet, that's the beauty of Heinlein's solution. All you need is a growing catalogue of associations. traits that increase or decrease in probability in association with not only the gametes (sperm & egg) but the mother's mitochondria and eggs and uteral propensities. In time, these correlations get better but they never have to be perfect, to let you pick your egg sperm pair from those with a good chance of being top percentile.David Brinhttps://www.blogger.com/profile/14465315130418506525noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-57966972159506757962007-05-03T19:57:00.000-07:002007-05-03T19:57:00.000-07:00Nah, if you need to simulate development to calcul...Nah, if you need to simulate development to calculate the results, why bother to create the embryo in meat-space in the first place? You might as well keep the child virtual all the way through. And if you're going to do that, you might as well remove the 99% of the genome that would be irrelevant to virtual space.<BR/><BR/>Somethings are essentially impractical and pointless.RandomSequencehttps://www.blogger.com/profile/12259854206507818658noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-76167500394276214682007-05-03T19:39:00.000-07:002007-05-03T19:39:00.000-07:00True, if you add "For now".True, if you add "For now".Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-8587336.post-61967058446184454242007-05-03T19:14:00.000-07:002007-05-03T19:14:00.000-07:00David,The only problem with any genetic engineerin...David,<BR/><BR/>The only problem with any genetic engineering plan is that most of our genetics isn't due to discrete genes - those are just minimal structural elements - but the regulatory sequences that are called "junk DNA". Those aren't discrete, and the language they're encoded in hasn't even been recognized yet.<BR/><BR/>On top of that, there's always the possibility of RNA transmission of heredity. It was found for plants last year. I bet in the next ten years, we find out that it's actually a wide-spread phenomenon. The world always has another trick up it's sleeve.<BR/><BR/>I would bet that the most important human qualities don't come out of simple genes, but emerge out of multiple regulatory systems, and that it will be almost impossible to predict what recombinations will produce (short of simulating the entire developmental process). Genetic engineering for anything other than disease prevention and eye-color is simply a utopian dream.RandomSequencehttps://www.blogger.com/profile/12259854206507818658noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-32618406143810687102007-05-03T18:21:00.000-07:002007-05-03T18:21:00.000-07:00Mark is tying into our earlier discussion of Heinl...Mark is tying into our earlier discussion of Heinlein.<BR/><BR/>The single most brilliant SF concept I ever saw was not some garish space structure built by cosmic BEMS, but the elegant social/technological solution that RAH offered, for how we can use the tools of genetic engineering to "improve" humanity without playing God or inflicting on generations the unforeseen tragedies arising from unwise whim.<BR/><BR/>It is a concept of stunning simplicity and unarguable moral validity, since the resulting child is always one that the couple MIGHT have had anyway, by shee random chance of sperm and egg. No feathers or gills. (Sorry Costner.) Simply the best natural outcome.David Brinhttps://www.blogger.com/profile/14465315130418506525noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-28123644003959237002007-05-03T18:06:00.000-07:002007-05-03T18:06:00.000-07:00I had no idea we already had the technology to art...I had no idea we already had the technology to artificially produce DNA. How long before a couple goes into a clinic and has their DNA scanned. The doctor explains they cannot legally produce DNA not already contained in the couple (other than fix a certain known set of birth defects) but the couple can choose whose DNA to use for each and every trait. An intelligent system can can choose the 'best' for them, if them prefer, though most couples like to pick eye color and such...Xactiphynhttps://www.blogger.com/profile/08254344563346437079noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-55975103905016811032007-05-03T15:50:00.000-07:002007-05-03T15:50:00.000-07:00Re: retrocausality. If I had to bet, my gut (long...Re: retrocausality. If I had to bet, my gut (long ago ex-physicist) says that reverse-causality will be demonstrated and the interference pattern will match the experimenter's choices even though the results were recorded before those choices where made.<BR/><BR/>I'm just not certain how the next experiment will unfold, where the experimenter controlling the switch is able to see the display while making the decision. Maybe the display will be neither wave-like nor particle-like, but a some post-quantum state of indeterminate.<BR/><BR/>As an aside, I think as one gets older one looses a belief in the traditional concept of free-will. Not to say I place any credence in determinism, that's a false dichotomy, but there's more going on. As for retro-causality, I do think, that my children somehow had something to do with determining who I married.Unknownhttps://www.blogger.com/profile/09899896624634559665noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-12112689444456220492007-05-03T15:17:00.000-07:002007-05-03T15:17:00.000-07:00RE the colony collapse disorder:Some G-ddamned obj...RE the colony collapse disorder:<BR/><BR/>Some G-ddamned objectivist translated <I>Atlas Shrugged</I> into bee, the workers realized the pernicious lie of altruism, and set out on their own.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-8587336.post-65418352826434634852007-05-03T14:55:00.000-07:002007-05-03T14:55:00.000-07:00The singularity: will we recognize it?Seriously, w...The singularity: will we recognize it?<BR/><BR/>Seriously, we've had these psychological and technical relocations before. The agricultural revolution, the printing press, Newton, etc. No one recognizes the shift in information processing at these stages, because the organism doing the processing is a new organism, invisible to the old.<BR/><BR/>It's like the Fermi Paradox solutions: the aliens are out there, and even here. But they are invisible to us, because they are at another scale of temporal, spatial and intellectual existence. It's like expecting bacteria to know that there exist multicellular organisms: all they see are other single cells.<BR/><BR/>The literary meta-mind is only visible to those steeped in literature. That took centuries to develop after the invention of the printing press, long after the literary meta-mind had already come to life and become the dominant element in human society throughout Eurasia. If the singularity happens, I expect us unicellular organism will go along quite the same as before - maybe after a century or so we might notice that we've become part of a new entity.RandomSequencehttps://www.blogger.com/profile/12259854206507818658noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-12914761976361406852007-05-03T14:42:00.000-07:002007-05-03T14:42:00.000-07:00David,But what about the bees??Honeybee die-off th...David,<BR/><BR/>But what about the bees??<BR/><BR/><A HREF="http://news.yahoo.com/s/ap/20070503/ap_on_sc/honeybee_die_off" REL="nofollow">Honeybee die-off threatens food supply</A><BR/><BR/>Talk about a sci-fi nightmare.RandomSequencehttps://www.blogger.com/profile/12259854206507818658noreply@blogger.comtag:blogger.com,1999:blog-8587336.post-14556408943033781192007-05-03T14:30:00.000-07:002007-05-03T14:30:00.000-07:00Progess in surveillence link doesn't work -- Wired...Progess in surveillence link doesn't work -- Wired can't find the article (whatever it is).Unknownhttps://www.blogger.com/profile/07295870097608491535noreply@blogger.com