intelligent discovery agents

[x <cyber1@io.org>]

On Fri, 6 Jan 1995 Nesta Stubbs <nesta@nesta.pr.mcs.net> wrote:

>> There was some recent talk about network agent technology on this list.
>>
> does anynoe have any information about intelligent agents?  I mean I know 
> about filter and stuff, but they arent intelligent agents.  I assume one 
> would be something like the WWW worm and other searching scripts that 
> have a database of information to cross-reference their finds and decide 
> what to send back to you.  

You might want to check out Brian LaMacchia's Ph.D. proposal to create
a new class of knowbot, to be termed "Internet Fish".  It is posted on 
http://www.swiss.ai.mit.edu/~bal/bal-home.html

LaMacchia's proposal is interesting in that his 'fish' seem to have 
limited autonomy, thus moving us closer to a content-addressable model
of net info retrieval.

Also of interest is RFC 1728: Resource Transponders, by C. Weider.  The
idea is that there should be meta-information (information about 
information) available for use by info retrieval programs like 'archie'.  
From ftp://nic.ddn.mil/rfc/rfc1728.txt

The following is extracted from LaMacchia's abstract:

> We will design, implement and deploy a system for constructing
> ``Internet Fish,'' a new type of resource discovery tool.  Internet
> Fish attempt to discover new sources of information related to a
> particular topic; characteristics that describe the topic of interest
> may be specified by the user or deduced by the Fish over the course of
> time.  As part of the information-gathering process Fish conduct
> long-term conversations with users; these conversations permit Fish to
> ask for human assistance when necessary and allow humans to
> dynamically reallocate Fish resources.  In addition, Fish facilitate
> *serendipitous* resource discovery; that is, the act of finding
> interesting information in an unexpected place or manner, information
> that we were ``lucky'' to have discovered.