Mike,

On 9/20/08, Mike Tintner <[EMAIL PROTECTED]> wrote:
>
>  Steve: If I were selling a technique like Buzan then I would agree.
> However, someone selling a tool to merge ALL techniques is in a different
> situation, with a knowledge engine to sell.
>
> The difference AFAICT is that Buzan had an *idea* -   "don't organize your
> thoughts about a subject in random order, or list, or tables or other old
> structures  etc. organize them like a map/tree on a page so that you can
> oversee them". Not a big idea, but an idea, out of wh. he's made money, &
> clearly appeals to many..
>

Addresses a different audience than I was looking at, but yea, I think I see
what you are getting at.


> If you have a distinctive idea, wh. you may well have, I've missed it &
> you're not repeating it. "A tool to merge all techniques" is a goal, not an
> idea. You have to show me that you have an idea - some new insight into
> general system principles applying to ,say, repair.
>

There is a large body of experience with various knowledge engines of
decades past. My "ideas" are tiny bits of glue that were missed in long past
projects that were hastily designed, programmed, presented, and abandoned.
In some of these cases, whole approaches were abandoned because of tiny
problems in their design or coding. I am just taking the considerable time
(now ~6 years) to methodically work though the myriad issues and identify
viable approaches to the challenges that buried long past projects. As I
have said here before, if not for Weizenbaum's book, Dr. Eliza or a very
similar program would have been developed by 1980 and the Internet
Singularity would have arrived on the heels of the first Internet
deployment. Weizenbaum precipitated a computer disaster on a scale fully
comparable to the Perceptron disaster, yet still, no one sees it.

  And if you are to do focus groups, you will also have to have a new idea
> to show them & test on them.
>

Hmmm, I hadn't even thought about focus groups. I consider this area to be
way too subtle for any but computational linguists and similar sorts of
experts to participate. So far, the folks working on the Russian Translator
have been the most helpful. There is no broad masterstroke of genius behind
Dr. Eliza, but instead countless seemingly insignificant details make it
work where prior efforts failed. "Little" details like making users answer
questions by editing their problem statements rather than answering the
questions directly. Made separately, these decisions would push Dr. Eliza
into the same holes that past systems fell into. Instead, it must be
designed as a working system. Do you think that I could be wrong in this
presumption?

Steve Richfield



-------------------------------------------
agi
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=8660244&id_secret=114414975-3c8e69
Powered by Listbox: http://www.listbox.com

Reply via email to