Haptics past and future: Gloves and the Octopus Butler
Robot
My interlocutor for this month: The Wii
Phenotropics; web mashups, brains, bacteria, and how to escape the curse of neo-Von Neumannism
My interlocutor for this month: Suffering programmers everywhere
Book reviews in American Scientist:A good
place to start
is
here.
...and here's Salon's story
on it. Ray Kurzweil has written a counter-manifesto
here. Here's
an essay about it in American Prospect. Here's
Upside's article on it, and here's
a lot of ranting about Upside's version of what I said on
Slashdot.
Here are
links related
to specific topics explored in the Half-manifesto"
The
"Singularity"
On
the problem of "Spiritual anxiety" as
egged on by "Cybernetic Totalists"
A good
place to start
is the debates on edge.org, such as here
or here.
Pragmatic
arguments against the
Artificial Intelligence research agenda
A good
place to start is this
book.
For online materials, start with this chapter from a book on Consciousness Studies. There are two opposing chapters on machine intelligence, with the "pro" position supported by Danny Hillis. I take the "anti" position.
My take on the Kasprov/Deep Blue Chess Game, originally published in IBM's Think Magazine
"Memes"
... and then see a rather friendly discussion with Richard Dawkins, the meme man himself, that was originally published in Psychology Today.
"Karma
Vertigo" is an older essay
identifying
some of the special problems of digital cultural persistence, also
first
published in Global
Business Network's magazine. Also see
this this essay.
In effort to make utopianism respectable once again, I gave talk on a 1000 year optimistic scenario. These are some of the powerpoint slides from that talk, though alas some browsers won't read them.
The
Association for
Computing Machinery asked various people to
write
essays on the subject of "hope" in the next fifty years of computing.
These
were published in the fiftieth anniversary special
issue
of the ACM's Communications
magazine.
Here's my contribution, titled "The
Frontier Between
Us".
I stand by everything I said in the above piece, even the bit about the
pernicious longevity of MS DOS influence, which, far from becoming
truly extinct,
now haunts us in stealth.
Here's a vintage interview which captures some of the delirious, infectious way I used to talk about VR in my twenties.
Finally
added some images of
vintage VPL VR equipment from the 1980s in answer to many
requests-
will add more and better photos and documentation soon.
On Sept 9, 2004, I gave a talk for Bay CHI, titled "Why VR has not (yet) become a widespread technology." This list of the top 11 reasons was read by most attendees in advance.
Here's
an interview about Phenotropics, from the main Java website.
And here
is the inevitable slashdot riot. Most of the
cranky
complaints
are
about one particular issue, and in this case I have to say the critics
are absolutely correct. I goofed by not catching an incorrect
number
when I checked the interview for accuracy. The number was
stated
as 10,000,000. It should clearly have been higher.
But that
value isn't important to the idea. If you want to know what
this
is all about, read the interview! Many thanks to all the
people
who
have written extended comments on the ideas in the interview and sent
them
to me directly. I will read and respond- might take me a
while!
Here's
the second part of the interview on the java website. This
portion
covers some of my earlier activities, like VR.
If you missed this seminar on Phenotropics, here are my characteristically sloppy slides.
Title: Should Computer Science Get Rid of Protocols Altogether?
Abstract: Computers are finally beginning to connect to the physical world with a little of the facility displayed by living organisms, which are generally able to interact reliably with an environment even though unplanned elements are often present. Medical instrumentation, robotics, and advanced user interfaces have all been improved recently by using techniques such as pattern recognition and predictive filtering. But operating systems and programming languages are still conceived of in the terms of mid-twentieth century engineering, in which sending signals on wires between slow machines was the central metaphor, and the protocol was the only solution. Perhaps it is time to take some of the advances from recent systems that interface to the real world and apply them to the world that remains strictly inside the computer. Components in such systems would connect together through pattern recognition of each other instead of adherence to protocols. This approach might be called “phenotropic”. While the phenotropic idea is still not fully formed, there are reasons to hope that phenotropic systems would display more useful failure modes, facilitating adaptive systems and avoiding crashes, and might also grow to larger sizes than traditional protocol-adherence-based systems.
My
Blog on HuffingtonPost is (or was- I lost interest after a
while) stimulated by politics and headlines, but
enduring ideas are addressed, and I think you will find that the
entries are still worth reading. The names and places can
always
be changed to keep up with the headlines of the future.
Go back to Jaron's home page.