DPRG List  

[DPRG] DARPA Grand Challenge

Subject: [DPRG] DARPA Grand Challenge
From: steve at txtulip.com steve at txtulip.com
Date: Thu Dec 19 02:17:46 CST 2013

Roddenberry ... Plbbt!!! Face it, he was in it for the money, not the science. I remember a schematic of a communicator with transistors and the idea that it was simple because the real complicated stuff was on the ship. Right... I have issues with that. 
   Back to the real world. I have no illusions that the DARPA robots are autonomous any further than the aerial drones we use fire their own missiles. There are things the drones do autonomously but complicated decisions are done by humans. The same would be true with DARPA robots. There is no Mr. Data in the foreseeable future. Mr. Data technology has yet to be developed because the tech we have now is inadequate for something that complicated. Whatever Mr.  Data is made of... It's not TTL.  
   My points still remains, it makes sense to make a humanoid robot because we build so much based on the human design, so a robot going into a devastated area that can use things designed for humans has a higher chance for success.


Sent from AOL Mobile Mail

-----Original Message-----
From: David P. Anderson <davida at smu.edu>
To: dprglist <dprglist at dprg.org>
Sent: Thu, Dec 19, 2013 12:36 AM
Subject: Re: [DPRG] DARPA Grand Challenge

Hi David,

Great to hear your thoughts, as always.  Your invocation of Mr. 
Roddenberry and the Sci-Fi predictions of 50 year ago are, I think, 
extremely germane to this discussion.  However, I have a slightly 
different take.  I'd like us to look at how SCi-Fi of that day depicted 
computers, versus how they actually have turned out.  I think that's 
applicable to the current robot situation.

Back then nobody really knew anything about computers, and so the SCi-Fi 
writers had to sort of make it up.   One thing they all had in common, 
as they made it up, was the concept of a computer as a sentient being, 
with desires and motivations, ambitions, opinions, and desires.  Really 
just a human in a box.  (For some reason they all spoke in a monotone, 
but otherwise their behaviour was indistinguishable from that of a 
human.  Sometimes a nice human. Sometimes mean.)

Because we really didn't know anything about them, we anthropomorphized 
them.  That is, we made them just like us. Humans have always done that 
with things we don't understand.   So in ancient times, the tornado 
happened because the wind became "angry" and "expressed it's 
displeasure."    That's the same thing that happened with the depiction 
of computers in the SciFi of the 50s and 60s.  WE made them just like 
us.   Of course the reality turned out to be quite different.

Fast forward 50 years to the present, when we all have computers on our 
desks and in our cars and our smart phones.  Now that we actually have 
computers, and don't have to just imagine them, it turns out that people 
don't think of their computers as living beings, as creatures with their 
own wills and desires.  They think of them as ATM terminals and word 
processors and game consoles and music systems and drafting and 3D CAD 
systems, and on and on.  As "tools."  But not as "beings."  The 
futurists of 50 years ago got a lot right, but they got that one dead wrong.

Now we appear to be in the same place today with robots that we were 50 
years ago with computers.  People don't really know anything about 
robots, except what they've seen in the movies, which is all made up.  
And so a humanoid robot, and even a fanciful creation like Star Trek's 
Data, seems reasonable, with the lack of any real data (pun intended) to 
know any better.

If past is any prologue to future, as your post suggests, then whatever 
our current image of the future of robotics may be, Mr Data included, 
it's likely to be dead wrong.  I'm actually quite optimistic about the 
future of autonomous robots.  We live in exciting times, where the 
ability of individuals to make significant contributions is like in the 
early days of flight.

David, I'm not sure I'm as bold in my certainty as you and Rud appear to 
be.   But my intuition is that history will prove the modern-day 
Alchemists were chasing a dead-end.

Next topic, why wheels are superior to legs for robots. Especially for 
robots in difficult and challenging terrain.


On 12/18/2013 07:54 PM, David Wilson wrote:
>> This goes back to my original premise, that a human(oid) form is not useful 
without a sophisticated human brain to operate it.  I've seen or heard nothing 
to alter that conclusion.   Except wishful thinking.  You know, a Positronic 
brain that suddenly become "intelligent" (whatever that means) by itself, the 
scientists don't understand how it works and don't have to ( a handy plot device 
when the author doesn't know either).  You know that famous cartoon: "...and 
then a miracle happens."
>> I pretty sure the we (you and I) had this conversation about 20 years ago 
(remember, back when AI and useful autonomous robots were always just 10 years 
away? :)  That fact that we are having it again now, 2 decades later, speaks 
volumes.  If I'm right about all this then, God willing, we'll be having the 
same conversation 20 years from now.
> We are fast approaching the 50th anniversary of the Star Trek Communicator and 
the speaking computers that seemed so impossibly out of reach back in the 60s.
> I suggest that even Roddenberry demonstrated a lack of imagination and insight 
into what our personal communicators would become. He created significant gadget 
lust for 23rd-century functionality that looks pathetic and unmarketable by 
today's standards. We might even claim that the iPhone arrived 194 years too 
early for his universe and the first smart phone / PDA gadgets are even older.
> Listen to some of us gray-beards and you might think the group would rather 
lug around 'easy to design' discrete devices. A simple camera. Hand cranked film 
movie camera. HP RPN calculator. Analog cell phone. Micro cassette voice 
recorder. Walkman. Mapsco. A real newspaper. I'm happy to live in a time when 
the difficult to achieve solution (smartphone with a real OS) is so widely 
> Mr Data will appear, eventually. His production date seems to be in the early 
2300s. I suspect that our desire to achieve that goal is so strong that 
Roddenberry will again be off by a huge margin, on both the arrival date and on 
the actual functionality that is achieved.
> David Wilson
> _______________________________________________
> DPRGlist mailing list
> DPRGlist at dprg.org
> http://list.dprg.org/mailman/listinfo/dprglist

DPRGlist mailing list
DPRGlist at dprg.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.dprg.org/pipermail/dprglist/attachments/20131219/a2269997/attachment.html 

More information about the DPRG mailing list