background image
Weblog tools
Blogger <>: Blogger offers a web-interface that is accessible from
any browser. It consists of an empty form box into which users can type their entries.With
a single mouse click, the weblog is posted to the writer's web site and is archived in the
proper place.The result are journal-style entries.
Metafilter <>: Metafilter is an example of a community weblog
that offers three form boxes: one to enter the URL of the referenced site, one for the title
of the entry, and another one to write ones commentary.
Pitas < >
However, the need for non-technical tools also exists within institutions themselves.
There is a high demand for templates or tools for presenting knowledge in innovative ways
or tools that enhance the capabilities of institutions to produce material for diverse channels
and end user devices.This demand should not be overlooked.
Introducing machine intelligence
Most basically, machine intelligence refers to the ability of computers to perform seem-
ingly intelligent knowledge-based tasks and/or "learn" certain behaviours. Intelligent techno-
logy is able to "learn" the user's preferences and patterns of thought to more effectively help
users find things.As Mark Jones,Victoria and Albert Museum, UK, notes:"We are starting to
see this beginning to work on better sites on the web already.You get the sense that your
pattern of activities has been noted and your are getting better service than you would from,
say, a shop assistant who does not know you." (DigiCULT Interview,August 9-10, 2001)
Collins (1988) distinguishes the following elements of computer intelligence:
Representation: Refers to the ability to map human knowledge structures into models;
those models for knowledge representation or ontologies are the basis for intelligent
querying, whereas a the query gives a selective view on the knowledge model.
Categorisation: The computer performs "intelligent" tasks on the basis of
comparison, through similarity and analogy. Categorisation is used in recommender
systems or filtering systems.
Learning: How does a human being learn and how can this process be simulated by
a machine?
Thinking: There are three elements of machine thinking: reasoning, problem solving
and planning:
reasoning uses facts and rules about facts to intelligently perform tasks; during
so-called inferencing, rules are applied to facts to deduce new facts,
problem solving works differently as it starts with a pre-given objective and
applies reasoning to reach this particular objective,
planning means to provide the system with the possibility to take intermediary
steps and reach a goal through immediate goals; thus, the machine can solve
more complex problems more efficiently, as complex tasks are split up in sub-
tasks with immediate goals that can be achieved individually.The system is able
to combine different immediate goals to reach the overall goal more efficiently.
Perception: Vision and imagery.
Collaborative behaviour: System can exchange information and behaviour with
other systems in an adaptive way. Agent needs a model of itself and a model of the
outside world.To communicate with the outside world the agent needs an own
language, an agent communication language and an agent communication protocol.