Log in

No account? Create an account

Previous Entry | Next Entry

cool links

There's a bunch of stuff I've been meaning to link to lately, but I was waiting till I got caught up reading friends' entries. It looks like my habits have settled down to spending Saturday catching up, posting on Sunday, and then responding to posts during the week, which is all I really have time to do. Most of these are re-posted from other people's journals, but I like posting them here because it collects together all the stuff I find interesting in one place, and it's unlikely that anyone has happened to already have seen all of them:

The Patrick Stewart Alphabet song... a really fun video of Captian Picard singing and performing broadway style onboard the enterprise:

a transformer halloween costume that transforms just like in the cartoon!:

Thursday was the 70th anniversary of Alan Turing's publication of On Computable Numbers, which laid the theoretical foundation for what we now call "computers". MIT celebrated by hosting a debate between Ray Kurzweil and Yale professor David Gelernter on the issue of whether intelligent machines which pass the Turing test should only be regarded as "zombies" or if they should be regarded as conscious like any other living, thinking, sentient being:
(As expected, Gelernter comes across looking rather foolish, while Kurzweil is right on target.)

Funny video parody about sexual consent forms:

a really impressive stop-motion animation:

Sitting up straight is bad for your back? When I first read the title, I thought "yippie! my sitting habits aren't as bad for me as I thought!" Unfortunately, it looks like the 135-degree angle they suggest for slouching is the opposite direction from the way in which I usually slouch:
Study Indicates Sitting up Straight is Bad for Your Back

Researchers unveil self-aware robot:

Juggling in a cone... stunningly beautiful:

There's also a bunch of personal stuff that has happened lately, which I plan to make a friends-only post on soon... hopefully tonight.


( 11 comments — Leave a comment )
Dec. 4th, 2006 09:06 am (UTC)
Wow, I'll have to pass on some of those videos!

I haven't looked at the Turing anniversary thing yet, but if it's about consciousness, then it seems odd that they didn't invite at least one philosopher to take part. (I suppose the MIT faculty doesn't have many philosophers of mind, though I don't know what Alex Byrne's focus is.)

And of course, calling that robot "self-aware" is a bit extreme. At first I was thinking, there's a reason that we've got a large amount of our understanding of our own shape hard-wired. But I guess being able to learn about it is useful for when things change. It's definitely an interesting step, but looks like it might be oversold (as science stories tend to be).
Dec. 5th, 2006 12:02 am (UTC)
Odd indeed... Searle may have been a better choice. Although I think he's debated Kurzweil before. One thing that frustrated me somewhat about this debate was that they kept going back and forth between talking about whether one could build an AI which would pass the Turing test out of software, and the issue of whether to call it conscious once it is.

But after sorting through all of that, I think most of it boils down to whether one believes there is such a thing as "private" mental states. Kurzweil (and I) do not believe there is a such thing... all mental states are objectively accessible, even if we can use subjective language to talk about the experience of the person whose mind is in those states. Gelernter's idea of consciousness relies on the existence of such states, but the problem is that definition is useless if such states don't exist. And he didn't provide any convincing argument as to why they would.

And of course, calling that robot "self-aware" is a bit extreme.
looks like it might be oversold (as science stories tend to be).

I agree that science stories tend to be oversold, especially the titles. And this is no exception. However, I don't think it's oversold in comparison to the extreme way in which some people tend to oversell the idea of "self awareness" in the first place. I've always found it irritating when people refer to self-awareness as if it's something difficult or more special than any other kind of awareness. I don't see it as having a special place, the self is just an concept you build up like any other concept. If you can be aware of "stuff" then you can be aware of "self" without much modification to the software. I also think consciousness is oversold in the same way, although it would take a lot more than self-awareness for me to call something conscious.
Dec. 5th, 2006 12:08 am (UTC)
I'd have to agree about awareness not being terribly special. I think. I guess it's hard to say - if I'm willing to say that a thermostat is aware of the temperature of a room, then it's certainly not very special. I might have to say awareness is more than just that, but that's a long way from it being anything terribly exciting.

As for private mental states, I don't really know what to say. I guess that all depends on how private they are supposed to be. I certainly don't want to say that we have infallible introspective access to them. Whether we can get third-person access to them is less clear. I'm also less clear what the significance of that would be.
Dec. 5th, 2006 02:28 am (UTC)

Whether we can get third-person access to them is less clear. I'm also less clear what the significance of that would be.

You might have to hear the debate to see the signifigance. It's significant because Gelernter's entire argument was based on it. Perhaps there are better arguments on his side, but this is the route he went.

He defined consciousness as having private mental states which have no effect on the external world, and cannot be detected by any physical device. He gave examples such as having a happy or sad thought, that nobody else knows about. He then proceded to argue that, because we always have objective access to all the states of a software program, software can never be conscious. Kurzweil rightly pointed out that when people have happy or sad thoughts, they do have effects that can be detected neurologically.
Dec. 4th, 2006 09:46 am (UTC)
I saw the juggling in a cone guy at the philly juggling festival awhile back. He also did a routine sitting cross legged and rolling metal bowls on the ground around him sort of like boomarangs. They'd roll out in an arc and come back to him, he had several of them going at once. He's always coming up with the most innovative stuff.

I'm not sure if I have it on video or not.
Dec. 4th, 2006 11:48 pm (UTC)
cool... I love people who push the envelope, that's the best kind of juggling
Dec. 5th, 2006 02:48 am (UTC)
This is my favorite of his routines:


Here is the video I shot of the metal bowl routine. For some reason I only caught the end of it on video.

Dec. 5th, 2006 06:46 am (UTC)
That was neat... thanks
Dec. 4th, 2006 10:55 am (UTC)
A self-aware robot? Wow, that's brilliant - I always get excited to hear about new developments in artificial intelligence. I wonder how long it will be until we've developed robots that are truly intelligent in the Turing-test-passing, conversational, intelligent sense, rather than simply being good at parsing facts in a disjointed and mechanical manner. To me, I think that the sign of digital intelligence would be the capability to synthesise and generate, rather than simply to analyse and parse.
Dec. 5th, 2006 10:33 pm (UTC)
Hey Jeff, it's Rob from NY. Just saying hi and adding you to friend's list.
Dec. 6th, 2006 06:40 am (UTC)
hey rob, hows it going? Hope new york is treating you well... wish I had been around more while you were here visiting. enjoyed hanging out.
( 11 comments — Leave a comment )


domino plural

Latest Month

May 2017


Powered by LiveJournal.com
Designed by Lizzy Enger