Re: Proposed Universal data types

From: J. Andrew Rogers (andrew@ceruleansystems.com)
Date: Thu May 19 2005 - 21:54:33 MDT


Marc Geddes wrote:
> All of the universal knowledge domains are in some
> sense equivalent i.e you can translate a description
> of something in one domain to a description in
> another. But I think they need to be seperated out -
> different representations are useful for computational
> tractibility for one thing.

Again, why? You are making a sweeping assertion, but so far it completely lacks justification
and is essentially baseless from the viewpoint of the audience. Notorious cranks have made
more interesting assertions on this list.

> The truth is I'm
> not worried about anyone stealing the ideas, because
> no one here (on SL4) has the faintest prospect of ever
> implementing a real AGI within the Bayesian framework.
> I'm sorry to inform you AGI wannabes that the world
> is quite safe from your puny efforts.

Ah, the hubris of ignorance. So you are intimately familiar with the work of everyone on the
SL4 list? Hey wait a minute, you are on SL4...

I'm neither agreeing or disagreeing with you here, but you might want to rein in that ego, lest
it get injured by passing traffic.

cheers,

j. andrew rogers



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:51 MDT