Dublin City University,
School of Computing
This is Technical Report no. CA-0301.
This is a first draft of ideas for the WWM system,
as at Feb 2001.
Further revisions will not appear here,
but rather in a number of
In the first part of this paper,
a change in methodology for the future of AI and Adaptive Behavior research
It is proposed that researchers construct their agent minds and their agent worlds
on the Internet.
3rd parties will use these servers as components in
In this scheme, any user on the Internet
will be able to
(a) select multiple
minds from different remote "mind servers",
(b) select a remote "Action Selection server"
to resolve the (inevitable) conflicts between these minds,
and (c) run the resulting constructed "society of mind"
in the world
provided on another "world server".
having to consult with
the server authors.
This constructed society may now also be presented
as just another primitive mind server,
ready for reuse by others as a component in a larger system.
From the current situation of isolated experiments
we will move to a situation where not only can researchers use each other's
but they can also use each other's agent minds
as components in larger systems.
Servers may call other servers, and
it is expected that
3rd parties will continuously write
wrappers and filters for existing mind servers,
overriding and modifying their default
behaviour (to produce new, co-existing mind servers).
None of this necessarily means that the mind being used
ever leaves its server (or that its insides are even made public).
Hence the term, the "World-Wide-Mind" (WWM),
referring to the fact that the mind
may be physically distributed
across the world, with parts of the mind at different remote servers.
Part of the motivation for the WWM
is that if the AI project is to be successful,
be too big for any single laboratory to complete.
So it will be necessary
both to decentralise the work
and to allow a massive and ongoing experiment
with different combinations of components
(so that we are not locked into any particular
layout of decentralisation).
Central to the WWM scheme is the expectation
that researchers will not agree on how to divide up the AI work,
and so components will overlap and be duplicated.
Previous work by this author
introduced models of mind where competition took place
between extremely incompatible components,
and where the mind could survive communications failure with
or even complete loss of a number of such components.
The WWM idea grew out of this work,
and this paper
how these previous models are
the type of models
we need in the WWM.
In the second part of this paper,
we move towards an implementation
of the WWM
by trying to define the
set of queries and responses that the servers
Clients (including other servers)
may then implement any general-purpose algorithm
to drive the servers through repeated use of these queries.
In our initial implementation,
we consider schemes of very low-bandwidth communication.
For instance, schemes
where the competition among multiple minds
is resolved across the network using numeric weights,
rather than by explicit reasoning and negotiation.
It is possible that this low-bandwidth protocol
may be more suitable
to sub-symbolic AI than to other branches of AI,
and that other protocols may be needed for other branches of AI.
It is suggested that it may be premature in some areas of AI
to attempt to formulate a "mind network protocol",
but that in the sub-symbolic domain
it could at least be attempted now.
Whether the protocol presented here is adopted or not,
the first part of this paper (the need for a protocol)
stands on its own.
Finally, we suggest a lowest-common-denominator approach
to actually implementing these queries, so that current AI researchers
have to learn almost nothing
in order to put their servers online.
As the lowest-common-denominator approach
we suggest the transmission across ordinary CGI
and responses written in XML.
Distributed Models of Mind,
Society of Mind,
autonomous agent architectures,
HTTP, CGI, XML, AIML.
Part 1 - Introduction
- 1. Introduction
- AI is too big a problem
- Duplication of Effort
- Unused agents and worlds
- Minds will be too complex to be fully understood
Part 2 - The World-Wide-Mind
- 2. The World-Wide-Mind
- Types of servers
- Types of societies
- Types of users
- Using other people's agent worlds
- Using other people's agent minds
- 3. Further issues on agent minds
- Mind AS server queries the Mind servers (not Client)
- Client talks to the World (not Mind server)
- Low-bandwidth communication
- Numeric communication - Q-values and W-values
- The role of Mind M servers
- What is the definition of state and action?
- 4. Further issues on agent worlds
- Why not separate World and Body servers?
- Changing the Body for the World
- Multiple Bodies in the same World
- The joint World-Body model is no restriction
- What if the Mind cannot make sense of the World?
- Real robots
- The name "The World-Wide-Mind"
- 5. How the WWM will be used in AI
- Dividing up the work in AI
- Making AI Science - 3rd party experimentation
- Artificial Selection
- How 3rd party AI researchers will use the scheme
- Bring every agent online
- 6. Objections to the model
- 7. Miscellaneous issues
- Hidden server insides
- Learning servers
- Learning Temperature
- Q-Temperature and W-Temperature
Part 3 - Implementation
- 8. Implementation
- Short, limited-length, client-server transactions
- Client algorithm
- The server may be involved in many runs
- The client controls time and may implement time-outs
- This is not a stimulus-response model
- Mind AS server algorithm
- The Mind AS server may also implement time-outs
- The servers (and client software) may implement any general-purpose algorithm using the server queries
- 9. List of server queries
- World server
- Mind server
- Additional Mind L queries
- Additional Mind i queries
- Additional Mind Feu queries
- Additional Mind AS queries
- 10. How to implement some existing agent architectures as networks of WWM servers
- Hand-coded program
- Initial test - Eliza Mind talks to Eliza World
- The Subsumption Architecture
- Serial models
- Maes' Spreading Activation Networks
- Reinforcement Learning
- Hierarchical Q-Learning
- Action Selection with a single query or multiple queries
- Static measures of W
- Dynamic measures of W
- Strong and Weak Mind servers
- Matching World state definition with Mind state definition
- "Islands" of compatible worlds
- The "island" of the physical world
- Mind servers with different senses in the same Society
- Global Action Selection decisions
- Other Action Selection methods based on RL
- Other parallel models
- The AS server remembering the winner
- Dynamically changing collections
- Nested Mind servers
- Each server calling a different list of servers
- Servers outside the AS loop
- Feudal Mind servers
- The sub-symbolic Society of Mind
- More complex communication between Mind servers
- Is this a sub-symbolic model?
- 11. HTTP CGI using XML
- HTTP CGI
- XML encoding of server queries
- Persistent CGI
- Asynchronous worlds
Part 4 - Future work and Conclusion
- 12. Future work
- Define the server queries
- Define the client user view
- Client use through existing Web browsers
- Dedicated client software
- Long-term prospects
- 13. Conclusion
- Endnote - Showing the world what a mind looks like
- 14. Acknowledgements
Return to publications or