CACM_JOCCH_one-third_page_vertical:Layout 1 7/30/09 5:50 PM Page 1ACM
and engineering phases of computing,
as well as the eventual use in deployment as a computer. This understanding tells us to use an experimental and
engineering process when developing
new formal models and methods of
computer sciences for our new devices,
paralleling the process of developing
new models and instruments to tackle
new phenomena in rest of the natural
sciences. A computational logic for
a system arises, but we then abstract
away from the specific device to a formal model of it. Programming these
new devices is then a matter of looking for a natural internal process logic
of the system, as opposed to forcing a
one-size-fits-all model of computation
onto some candidate computing system. Rather than looking to impose
top-down the machinery of standard
logic gates, we should look at the natural behaviour of the system and what
‘gates’ or subroutines or problem-solving it is intrinsically good at. By extracting an intrinsic computational logic of
their physical components we can harness the true potential of unconventional computers.
ology—and even, with the interactions
of social machines, for networks of
human beings. We believe this could
be of immediate practical importance
to scientists in those areas, enabling
them to describe high-level functioning of complex systems, and to find
new and unforeseen connections between disparate systems and scenarios. These process languages could be
as revolutionary for the physical sciences as for computer science.
JOCCH publishes papers of
significant and lasting value in
all areas relating to the use of ICT
in support of Cultural Heritage,
seeking to combine the best of
computing science with real
attention to any aspect of the
cultural heritage sector.
Using our physical understanding
of a substrate to inform a computational logic does not mean that such
a logic is the only one possible. Just
as a quantum computer can run as either quantum or classical, other non-standard systems may be capable of
supporting multiple computational
models. This again is found throughout the natural sciences: for example,
in physics a particular system might
be modelled as a continuous fluid, or
as a collection of discrete particles.
With different potential computational representations of a system under investigation, the key is to extract
out the ones that do something useful
and novel and better than other substrates—and then use that computational theory to engineer our next generation of computers.
Computers have come a long way
since the days of valves and punched
cards. Now computer science itself is
branching off in new directions with
the development of unconventional
computing technologies. As the domain of computer science grows, as
one computational model no longer
fits all, its true nature is being revealed.
Just like astronomy, computer science
could describe physical systems in abstract language with predictive power,
and thereby drive forward the dual interplay of technology and theoretical
advancement. New computers could
inform new computational theories,
and those theories could then help us
understand the physical world around
us. Such a computer science would indeed be a natural science.
Copeland, J. et al. Time to reinspect the
foundations? Commun. ACM 59, 11 (Nov.
Horsman, C. et al. When does a physical
system compute? In Proceedings of the
Royal Society of London, 470:20140182,
Horsman, D.C. Abstraction/Representation
Theory for heterotic physical computing.
In Philosophical Transactions of the Royal
Society, 373:20140224, 2015.
Horsman, D.C. Abstraction and
representation in living organisms: When
does a biological system compute? In G.
Dodig-Crnkovic and R. Giovagnoli, Eds.
Representation and Reality in Humans,
Animals, and Machines. Springer, 2017.
We can then go further. With an abstract computational language that describes the native operation of unconventional devices, we would then have
a logical language in which to describe
the physical systems themselves, even
outside a specifically computational
device. Computer science could then
provide high-level logical process languages for physics, chemistry, and bi-
Dominic Horsman ( email@example.com) is
a Postdoctoral Research Associate at the University of
Vivien Kendon ( firstname.lastname@example.org) is a Reader in
the Department of Physics at the University of Durham, U. K.
Susan Stepney ( email@example.com) is Professor
of Computer Science in the Department of Computer
Science, University of York, U.K.
Copyright held by authors.