Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!ucsd!ucbvax!decwrl!wuarchive!texbell!texsun!newstop!sun!stpeter.Eng.Sun.COM!cmcmanis From: cmcmanis@stpeter.Eng.Sun.COM (Chuck McManis) Newsgroups: comp.robotics Subject: Re: Divert NASA budget to robotics Message-ID: <138873@sun.Eng.Sun.COM> Date: 13 Jul 90 17:53:30 GMT References: <10518@batcomputer.tn.cornell.edu> <#*S$G+&@ads.com> <848@sagpd1.UUCP> <$}T$X6&@ads.com> Sender: news@sun.Eng.Sun.COM Distribution: comp Organization: Sun Microsystems, Mt. View, Ca. Lines: 40 In article <$}T$X6&@ads.com> pkenny@ADS.COM (Patrick Kenny) writes: > Let me ask you a question, > -Do you want to go out in space every day for countless years with a wrench > in hand turning nuts to create a space platform. ? > -Would you like to work in a moon coal mine 3 miles below the surface digging > out minerals to manafacture steel to create some shelter. And since you > would be the first explores on the moon, you would have to live in a small > space ship for the years it would take you to create something big enough > to be productive in.? Wrong question but I'll throw in my .02. This question (like most) has a simple opinion based answer and raises some other difficult to answer question. The opinion based answer (in this case based on my opinion) is that if by doing the manual labor Man will be on the moon in this decade rather than in the next or maybe the decade after that, I'll go today. Other questions this one raises : Consider for the moment that it is already very difficult to teach robots to do something you or I can do today, in conditions we can experience today. Attempting to design a robot that can construct a space station is an order of magnitude more complex because no human has ever constructed a space station in space and so there is no one to ask things like "what is the most efficient way to move from beam to beam?" The current state of robotics is such that we are almost able to teach a robot to do something "we" as humans are already completely familiar with, we are no where close to being able to teach a robot to go into a situation that we can only speculate about and learn how to do something. Further, since the final test will undoubtedly involve sending it up and trying it out, are you willing to bet that you can debug a robot assembler in few enough less expensive manned flights to be more cost effective than one possibly two manned flights? In my opinion it is much more rational to send a human up to build a space station, then have her come back and help train robots to do it, (who then build 10 more). -- --Chuck McManis Sun Microsystems uucp: {anywhere}!sun!cmcmanis BIX: Internet: cmcmanis@Eng.Sun.COM These opinions are my own and no one elses, but you knew that didn't you. "I tell you this parrot is bleeding deceased!"