|
When we started
MyST Technology Partners, we frequently made the comment that important
information should find us instead of us always searching for it. The first
overt instance of that capability came when we implemented SmartPoints -
additional search results that indicate who is likely to know the most about a
search topic and what channels are likely to hold more information about a
search topic. This is pretty compelling when you first start playing with it
(the free public
site
provides this capability).
"in the last
two decades there has been increasing attention on a new set of problems those
dealing with the interactions between multiple computational entities" --
Steven C.
Laufmann
This is an
important indication of a shift that is driven in part by the increasing volume
of information that we have to deal with as well as the drive to create content
that is machine readable. At no time in computing history have companies agreed
on semantic tagging standards until now. Large software firms that are at each
others throats have concluded that supporting XML open standards are key to
the future of successful computing strategies. This has now paved the way for
significant advancement in terms of agent technologies.
Of greater interest
is the idea that agents can work on your behalf to retrieve information in
advance of you needing it. This sounds AI'ish, but it really isn't that
difficult or far-fetched. With Web services architectures it's now possible to
gather additional information relevant to some activity immediately after
the activity has begun. An example of this can be found at blueskycolorado.com. As content writers create
articles about outdoor activity, queries into Amazon.com are dynamically updated to
include the keywords of the articles. We use the Amazon API to make real-time
requests for visitors - so, as new material is added, the system naturally maps
the interests of the writers to books that are relavant - all in
real-time.
I'm not an expert
on agent technologies, but I know one when I experience one. ;-) I
experienced the benefit of an agent just this morning when I reviewed my RSS
feeds from six channels that are being updated every 12 hours by a "channel
gear". A channel gear is an agent designed to keep a collection of content about
a specific subject up to date based on what Google is aware of. This (among the
many machineries) is largely unknown about the MyST platform - we have an
agent scripting mechanism that doesn't require any expertise about the Google
API or the MyST API. As a business content expert, I'm able to script
instructions that carry out the steps of finding information and updating target
channels. But to really make this approach dance, the agents that harvest
important content for me are supported by another type of agent - an RSS reader. A good example of
this can be found on the RSS WinterFest
Site. The content items in the RSS Radar section are all automated queries
against Google, but they also include RSS feed
addresses.
My hope is that as
we continue to build new information systems that seem instinctively aware of
activities surrounding you, the system will naturally invoke behaviors that
essentially say "And BTW, I retrieved this information because you'll be needing
it."
If you have an
interesting application where this might be useful, give us a
shout. | 
