This document discusses using behavioral implicit communication (BIC) to enable anticipatory coordination in socio-technical knowledge-intensive environments. It proposes extending the Molecules of Knowledge (MoK) coordination model to treat computational environments as shared environments where agents' actions and traces are observable. Enzymes and traces in MoK are redefined to represent epistemic actions and their side effects, enabling the environment to implicitly understand intentions and goals. This lays the groundwork for collective intelligence and anticipatory coordination through the tacit messages conveyed by agents' observable behaviors. An experiment simulates a citizen journalism scenario where search actions in a MoK-based system spread information in a self-organizing, adaptive manner.

1. Anticipatory Coordination
in Socio-technical Knowledge-intensive Environments:
Behavioural Implicit Communication in MoK
Stefano Mariani, Andrea Omicini
Alma Mater Studiorum—Universit`a di Bologna, Italy
AIxIA 2015
Ferrara, Italy
24th September 2015
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 1 / 31

2. Outline
1 Premises
2 Ingredients
3 Towards Anticipatory Coordination
4 Experiment
5 Conclusion
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 2 / 31

3. Premises
Outline
1 Premises
2 Ingredients
3 Towards Anticipatory Coordination
4 Experiment
5 Conclusion
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 3 / 31

4. Premises
Challenges of Socio-technical Systems
Socio-technical systems (STS) arise when cognitive and social
interaction is mediated by information technology, rather than by the
natural world (alone) [Whi06]
STS are heavily interaction-centred, so they need coordination at the
infrastructure level [MC94]
Coordination issues in STS are:
unpredictability “humans-in-the-loop”, whereas software behaviour is
programmable and predictable, human’s is not
scale physically-distributed, open systems, large-scale,
ever-increasing number of people, devices, data
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 4 / 31

5. Premises
Challenges of Knowledge-intensive Environments
Knowledge-intensive Environments (KIE) are workplaces in which
sustainability of the organisation long-term goals is influenced by the
evolution of knowledge [Bha01]
Usually, KIE are computationally supported by STS, so they need
coordination too
Coordination issues in KIE are:
size massive amount of raw data, aggregated information,
reification of procedures and best-practices, and the like
pace data is produced and consumed at a fast pace, huge
frequency of interactions
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 5 / 31

6. Premises
The Idea
Coordination models and technologies draw inspiration from
distributed collective intelligence phenomena in natural systems,
looking for self-organising and adaptive coordination mechanisms
[VPB12, MZ09, VC09, ZCF+11, MO13]
A novel perspective
Similarly, we focus the “social layer” of STS, looking for novel
coordination approaches inspired by the latest cognitive and social sciences
theories of action.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 6 / 31

7. Premises
The Approach
1 We take as a reference the M olecules of K nowledge (MoK )
coordination model for knowledge self-organisation in KIE [MO13]
2 Extend it toward the notion of anticipatory coordination — as a form
of collective intelligence arising by emergence
3 According to the theory of Behavioural Implicit Communication (BIC)
[CPT10]
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 7 / 31

8. Ingredients
Outline
1 Premises
2 Ingredients
3 Towards Anticipatory Coordination
4 Experiment
5 Conclusion
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 8 / 31

9. Ingredients
Behavioural Implicit Communication
Implicit interaction
Behavioural implicit communication (BIC) is a form of implicit interaction
with no specialised signal conveying the message, since the message is the
practical behaviour itself — and possibly, its traces [CPT10].
BIC presupposes observation: agents should be able to observe
others’ actions (and traces), as well as to mind-read the intentions
behind them
BIC applies to human beings, to both cognitive and non-cognitive
agents, and to computational environments as well [WOO07]
Smart environments are pro-active, intelligent working environments
able to autonomously adapt their behaviour according to users’
interactions [CPT10] 1
1
Which is, not by chance, the very notion of anticipatory coordination.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 9 / 31

10. Ingredients
Smart (Computational) Environments
In [TCR+05], an abstract model for smart environments defines two
types of environment:
c-env common environment, where agents can observe only
the state of the environment (including actions’ traces),
not the actions of their peers
s-env shared environment, enabling different forms of
observability of actions, and awareness of this
observability
Three requirements can thus be devised:
1 observability of agents’ actions and traces enabled by default
2 the environment should be able to understand actions and their traces,
possibly inferring intentions and goals motivating them
3 agents should be able to understand the effects of their activity on
both the environment and other agents, so as to opportunistically
obtain a reaction
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 10 / 31

11. Ingredients
M olecules of K nowledge I
MoK
M olecules of K nowledge (MoK ) is a coordination model promoting
self-organisation of knowledge [MO13].
Inspired to biochemical tuple spaces [VC09] and stigmergic
coordination [Par06]
Main goals:
1 self-aggregation of information into more complex heaps, possibly
reifying useful knowledge previously hidden
2 autonomous diffusion of information toward the interested agents, that
is, those needing it to achieve their goals
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 11 / 31

12. Ingredients
M olecules of K nowledge II
A MoK -coordinated system is thus:
a network of MoK compartments (tuple-space like information
repositories)
in which MoK seeds (sources of information) autonomously inject
MoK atoms (information pieces)
atoms undergo autonomous and decentralised reactions
aggregate into molecules (composite information chunks)
diffuse to neighbourhoods
gets reinforced and perturbed by users
decay as time flows
reactions are influenced by enzymes (reification of users’ epistemic
actions)
and scheduled according to Gillespie’s chemical dynamics simulation
algorithm [Gil77]
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 12 / 31

13. Towards Anticipatory Coordination
Outline
1 Premises
2 Ingredients
3 Towards Anticipatory Coordination
4 Experiment
5 Conclusion
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 13 / 31

14. Towards Anticipatory Coordination
MoK Compartments as c-env
The only sort of smart environment enabled in MoK is c-env,
mapped upon a compartment, because [MO13]:
1 no sharing of compartments is assumed
2 enzymes are visible only to MoK reactions
3 enzymes cannot diffuse, thus neighbourhoods cannot perceive them
So, MoK does not support s-env since there is no observable action
reification in shared environments
Also, support to c-env is limited to compartments – not
neighbourhoods – since enzymes cannot diffuse
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 14 / 31

15. Towards Anticipatory Coordination
MoK Compartments as s-env
Compartment can now be shared among users
Enzymes are
diversified to resemble the epistemic nature of the action they reify
made observable to users sharing the compartment they live in
no longer consumed by reinforcement reaction, but now subject to
decay
now generating traces through a deposit reaction
Traces are introduced as the MoK abstraction resembling (side)
effects of actions
different in kind, according to their “father” enzyme
observable solely by MoK reactions
subject to diffusion, decay and to a novel enzyme-dependant
perturbation reaction
MoK compartments extension leads to neighbourhoods representing
c-env, where action traces may diffuse, becoming part of the
environment as they participate MoK reactions
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 15 / 31

16. Towards Anticipatory Coordination
Enzymes and Traces as BIC Enablers I
Model
enzyme(species, s, mol)c
enzyme(species, s, mol ) + molc
rreinf
−−−→ enzyme(species, s, mol ) + molc+s
trace(enzyme, p, mol)c
trace(enzyme, p, mol ) + molc
rpert
−−→ .exec (p, trace, mol)
enzyme
rdep
−−→ enzyme + trace(enzyme, p[species], mol)
species defines the epistemic nature of the action
s strength of reinforcement
p the perturbation the trace wants to perform
.exec starts execution of perturbation p 2
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 16 / 31

17. Towards Anticipatory Coordination
Enzymes and Traces as BIC Enablers II
Reinforcement influences relevance of information according to the
(epistemic) nature and frequency of their actions, so as to better
support them in pursuing their goals
Enzymes situate actions, e.g., at a precise time as well as in a precise
space
Mind-reading and signification by assuming that users manipulating a
given corpus of information are interested in that information more
than other
Perturbation influences location, content, any domain-specific trait of
information, according to users’ inferred goals, with the goal of easing
and optimising their workflows
Traces enable the environment to exploit users’ actions (possibly,
inferred) side-effects for the profit of the coordination process —
promoting the distributed collective intelligence leading to
anticipatory coordination
2
Notice, p is implicitly defined by species, as highlighted by notation p[species].
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 17 / 31

18. Towards Anticipatory Coordination
Tacit Messages Steering Anticipatory Coordination I
Tacit messages [CPT10] describe the kind of messages a practical
action (and its traces) may implicitly send to its observers:
presence “Agent X is here”. Since an action (trace) is observable in
shared compartments (neighbourhoods), any agent therein
becomes aware of X existence and location — likewise for the
environment.
intention “X plans to do action b”. If the agents’ workflow determines
that action b follows action a, peers (as well as the environment)
observing X doing a may assume X next intention to be “do b”.
Accordingly, the environment may decide to undertake anticipatory
coordination actions easing/hindering action b — e.g. because
action b is computationally expensive.
ability “X is able to do ai=1,...,n”
. . . . . .
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 18 / 31

19. Towards Anticipatory Coordination
Tacit Messages Steering Anticipatory Coordination II
. . . . . .
opportunity “ [e1, . . . , en] is the set of pre-conditions for doing a”
accomplishment “X achieved S”
goal “X has goal g”
result “Result r is available”
Enabling Collective Intelligence
Since agents can undertake the above described inferences, MoK
compartments actually act as BIC-based enablers of distributed collective
intelligence phenomena — e.g., anticipatory coordination emerging due to
agent interaction.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 19 / 31

20. Towards Anticipatory Coordination
Tacit Messages Steering Anticipatory Coordination III
In many STS, a set of fairly-common actions may be devised, in spite
of the diversity in scope of the software platforms — e.g. Facebook
vs. Mendeley:
quote/share re-publishing or mentioning someone else’s information
can convey, e.g., tacit messages presence, ability,
accomplishment. If X shares Y ’s information through action a,
every other agent observing a becomes aware of existence and
location of both X and Y (1). The fact that X is sharing
information I from source S lets X’s peers infer X can manipulate
S (3). If X shared I with Z, Z may infer, e.g., that X expects Z
to somehow use it (5).
like/favourite marking as relevant a piece of information can convey,
e.g., tacit messages presence, opportunity.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 20 / 31

21. Towards Anticipatory Coordination
Tacit Messages Steering Anticipatory Coordination IV
follow subscribing for updates regarding some piece of
information or some user can convey tacit messages
intention, opportunity.
search performing a search query to retrieve information can
convey, e.g., tacit messages presence, intention,
opportunity.
Perturbation Actions
Accordingly, each tacit message may cause a perturbation action
contributing to anticipatory coordination — e.g. send discovery messages,
establish privileged communication channels, undertake coordination
actions enabling/forbidding some interaction protocol, notify users about
availability of novel information, etc.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 21 / 31

22. Experiment
Outline
1 Premises
2 Ingredients
3 Towards Anticipatory Coordination
4 Experiment
5 Conclusion
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 22 / 31

23. Experiment
Simulated Scenario
Citizen journalism scenario 3
users share a MoK -coordinated IT platform for retrieving and
publishing news stories
users have personal devices (smartphones, tablets, pcs, workstations),
running the MoK middleware, which they use to search the IT
platform for relevant information
search actions can spread up to a neighbourhood of compartments —
e.g., to limit bandwidth consumption, boost security, optimise
information location, etc.
search actions leave traces the MoK middleware exploits to attract
similar information, actually enacting anticipatory coordination
3
Technical details of the simulations – e.g. tools used and simulation parameters – in
the full paper.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 23 / 31

24. Experiment
Self-organising Adaptive Anticipatory Coordination
Figure: Whereas data is initially randomly scattered across workspaces, as soon as users
interact clusters appear by emergence thanks to BIC-driven self-organisation. Whenever new
actions are performed by catalysts, the MoK infrastructure adaptively re-organises the spatial
configuration of molecules so as to better tackle the new coordination needs.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 24 / 31

25. Conclusion
Outline
1 Premises
2 Ingredients
3 Towards Anticipatory Coordination
4 Experiment
5 Conclusion
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 25 / 31

26. Conclusion
Conclusion & Outlook
Although our experiment focusses on one specific pattern of
anticipatory coordination, results are more than encouraging,
deserving further investigation
Efforts are currently devoted to fully implement and run a MoK
coordinated system on a large-scale scenario—a prototype
implementation of MoK exists, but large-scale deployment has not
been achieved, yet
Special care will be payed to the semantic similarity measure
ontology-based semantic matching is rather unfeasible in pervasive,
mobile scenarios
purely syntactical matching has too low expressiveness
viable tradeoffs may be usage of wildcards, e.g. as in Java regular
expressions4
, or of synonymy relationships, e.g. as done in [PVM+
10]
using WordNet5
4
http://docs.oracle.com/javase/tutorial/essential/regex/
5
http://wordnet.princeton.edu
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 26 / 31

27. References
References I
Ganesh D. Bhatt.
Knowledge management in organizations: Examining the interaction between technologies,
techniques, and people.
Journal of Knowledge Management, 5(1):68–75, 2001.
Cristiano Castelfranchi, Giovanni Pezzullo, and Luca Tummolini.
Behavioral implicit communication (BIC): Communicating with smart environments via our
practical behavior and its traces.
International Journal of Ambient Computing and Intelligence, 2(1):1–12, January–March
2010.
Daniel T. Gillespie.
Exact stochastic simulation of coupled chemical reactions.
The Journal of Physical Chemistry, 81(25):2340–2361, December 1977.
Thomas W. Malone and Kevin Crowston.
The interdisciplinary study of coordination.
ACM Computing Surveys, 26(1):87–119, 1994.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 27 / 31

28. References
References II
Stefano Mariani and Andrea Omicini.
Molecules of Knowledge: Self-organisation in knowledge-intensive environments.
In Giancarlo Fortino, Costin B˘adic˘a, Michele Malgeri, and Rainer Unland, editors,
Intelligent Distributed Computing VI, volume 446 of Studies in Computational Intelligence,
pages 17–22. Springer, 2013.
Marco Mamei and Franco Zambonelli.
Programming pervasive and mobile computing applications: The TOTA approach.
ACM Transactions on Software Engineering and Methodology (TOSEM), 18(4), July 2009.
H. Van Dyke Parunak.
A survey of environments and mechanisms for human-human stigmergy.
In Danny Weyns, H. Van Dyke Parunak, and Fabien Michel, editors, Environments for
Multi-Agent Systems II, volume 3830 of Lecture Notes in Computer Science, pages
163–186. Springer, 2006.
Danilo Pianini, Sascia Virruso, Ronaldo Menezes, Andrea Omicini, and Mirko Viroli.
Self organization in coordination systems using a WordNet-based ontology.
In Indranil Gupta, Salima Hassas, and Rolia Jerome, editors, 4th IEEE International
Conference on Self-Adaptive and Self-Organizing Systems (SASO 2010), pages 114–123,
Budapest, Hungary, 27 September–1 October 2010. IEEE CS.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 28 / 31

29. References
References III
Luca Tummolini, Cristiano Castelfranchi, Alessandro Ricci, Mirko Viroli, and Andrea
Omicini.
“Exhibitionists” and “voyeurs” do it better: A shared environment approach for flexible
coordination with tacit messages.
In Danny Weyns, H. Van Dyke Parunak, and Fabien Michel, editors, Environments for
Multi-Agent Systems, volume 3374 of Lecture Notes in Artificial Intelligence, pages
215–231. Springer, February 2005.
Mirko Viroli and Matteo Casadei.
Biochemical tuple spaces for self-organising coordination.
In John Field and Vasco T. Vasconcelos, editors, Coordination Languages and Models,
volume 5521 of Lecture Notes in Computer Science, pages 143–162. Springer, Lisbon,
Portugal, June 2009.
Mirko Viroli, Danilo Pianini, and Jacob Beal.
Linda in space-time: An adaptive coordination model for mobile ad-hoc environments.
In Marjan Sirjani, editor, Coordination Models and Languages, number 7274 in Lecture
Notes in Computer Science, pages 212–229. Springer, 2012.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 29 / 31

30. References
References IV
Brian Whitworth.
Socio-technical systems.
Encyclopedia of human computer interaction, pages 533–541, 2006.
Danny Weyns, Andrea Omicini, and James J. Odell.
Environment as a first-class abstraction in multi-agent systems.
Autonomous Agents and Multi-Agent Systems, 14(1):5–30, February 2007.
Franco Zambonelli, Gabriella Castelli, Laura Ferrari, Marco Mamei, Alberto Rosi, Giovanna
Di Marzo, Matteo Risoldi, Akla-Esso Tchao, Simon Dobson, Graeme Stevenson, Yuan Ye,
Elena Nardini, Andrea Omicini, Sara Montagna, Mirko Viroli, Alois Ferscha, Sascha
Maschek, and Bernhard Wally.
Self-aware pervasive service ecosystems.
Procedia Computer Science, 7:197–199, December 2011.
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 30 / 31

31. Anticipatory Coordination
in Socio-technical Knowledge-intensive Environments:
Behavioural Implicit Communication in MoK
Stefano Mariani, Andrea Omicini
Alma Mater Studiorum—Universit`a di Bologna, Italy
AIxIA 2015
Ferrara, Italy
24th September 2015
Mariani, Omicini (Universit`a di Bologna) Anticipatory Coordination: BIC in MoK AIxIA 2015 – 24/9/2015 31 / 31