MapleSim 16

After an initial lockstep phase in which MapleSim upgrades were closely tied to those for Maple, release 6 was held back and appeared shortly after Maple 16. It may be that this marks the point when the simulation package asserts its independent existence as an application in its own right, rather than a younger sibling in the shadow of its older stable mate.

Not that the links with Maple have loosened in any functional sense; in fact, their potential for integration continues to grow. Maple can now transparently open MapleSim (and Modelica, of which more in a moment) models within its worksheets and make thread safe MapleSim function calls from its TPM (the high level, multithreaded, Task Programming Model). Information from the MapleSim model, including system equations, can be accessed, as can MapleSim's full solver repertoire. And Maple can set or change MapleSim parameters and solver settings, so is able to operate as a symbolic front end. The two products remain firmly complementary, but not co-dependent; MapleSim is developing a wider vision.

[more]

Not trawling but drowning

[Having finished writing this, and just as I am about to post it, I look at the accretion at its end and realise that it is going to drive Julie Heyward, who views footnotes with disfavour, to distraction. Apologies in advance, Julie.]

My fascination with the idea of drowned geographies started (back in the prehistoric days of 1977) with a Richard Cowper¹: short story, Piper at the gates of dawn². It then took a quantum leap forward when I later heard a folk song about "the trawler fleets of Trowbridge"³.

Trowbridge is a smallish town in the British county of Wiltshire. It is a fair way inland, and not near any significant body of water. The lyrics were intended as a nonsensically humorous send up of traditional sea related folk song, but they latched into my imagination and stayed there vividly. I was writing a fair amount of short fiction at that time, and tried many times to do something Cowperesque with this image of trawler fleets sailing out of a once landlocked town. It never happened, but the idea never went away, either. I have, ever since, played with maps and tried out the idea of flooding different landscapes to different contour lines – first on paper and then, in later years, digitally. To be honest, in retrospect, the maps were probably of greater interest to me⁴ than the nominal objective for drawing them...

All of this floods (excuse the pun) back now because I have just read Ray Girvan's JSBlog post on Floodland, which I immediately went out and bought because (a) it plays to this drowned geography weakness of mine and (b) Ray's recommendations are usually good⁵; it's joined the end of the "to be read" queue.

More immediately, however, I was (given this fascination of mine) obviously unable to ignore Ray's mention of a global flooding visualiser at flood.firetree.net. Hardly had I finished reading his post than I was over there and (of course!) flooding Wiltshire to see the effect on (where else?) Trowbridge.

There is, alas (or perhaps, if you are an inhabitant of Wiltshire, fortunately) no way to flood the countryside around Trowbridge to such a depth that trawlers could realistically operate out of the town. At thirty metres rise in sea level, Trowbridge would remain landlocked. At thirty five metres (the site won't do this; I had to revert to manual inspection of Ordnance Survey maps) it would acquire a coastline on a shallow lake in a wetlands region.

Despite its inland location, Trowbridge has a "Bythesea Road". At a little less than forty metres rise in sea level, this would live up to its name by becoming the town's seaside promenade.

At forty metres it would become a town on one minor arm of an inland sea (as shown in my illustration here – click it for a larger view).

At fifty, most of the town itself would be submerged, leaving two small parts of it at north east and south west on a pair of islands. At sixty metres it would disappear almost entirely, apart from a few scattered islets at the mercy of the tides just off what would now be the coast of Steeple Ashton ... in a substantial sea which would certainly support a fishing industry but not trawler fleets.

No matter ... I enjoyed the adventure of vicarious post apocalyptic disaster. Tomorrow I shall almost certainly flood somewhere else...

---

1. I made no linkage then, and don't now, with J G Ballard's earlier Drowned world which, though it powerfully affected me in other ways, failed to evoke a concrete transformation of the world I know. Ballard's world was an obliterated world, not a transformed one.
2. To answer the obvious questions: there is a clear connection with Wind in the willows, but it isn't relevant here; possibly also with the Pink Floyd album released about eight years before, though I'm not aware of it and haven't checked. There can be no direct link to Van Morrison song written about twenty years later. Pink Floyd and Van Morrison also both drew their inspiration from Wind in the willows)
3. I can't, unfortunately, give any details of the song, as I'm unable to find any record of it ... I suspect that it was penned by the group who performed it, or by someone known to them personally, and never travelled widely enough to leave a permanent fossil record.
4. In Tesseract (no connection with the later Alex Garland novel of very similar title), a longer fiction which I never finished but which served as the spawning ground for several shorter ones, I also tried flooding various landscapes in South Wales not with water but with time ... the further the protagonists moved from modern infrastructures like motorways or centres of population, the further back they sank into the past ... it was fun to work on. I'm a dreamer at heart.
5. It was Ray who put me onto another superb novel in which drowned geography plays a major part: Ronald Wright's A scientific romance.

---

• J G Ballard, The Drowned World. 1962, London: Victor Gollancz.
• Richard Cowper The custodians, and other stories. 1976, London: Gollancz. 0575020962
• Alex Garland, The tesseract. 1999, London: Penguin. 0140258426 (pbk.)

Nasties under the bed

Just found, in Games and culture...

“This paper explores the characterizations of enemies in military-themed video games, with special attention given to the games Conflict: Desert Storm and America's Army. I demonstrate how the public enemy of America's Army is one not confined to any nationality, ethnicity, or political agenda. This marks a significant departure from games such as Conflict: Desert Storm. I argue that the production of this abstract enemy--what I call the unreal enemy''--is significantly shaped by a biopolitical system that intertwines the military and electronic entertainment industries. This arrangement delocalizes power, distributing it through a network of institutions and subjects. Throughout, I use ethnographic examples that explore how this abstract enemy has been constructed and juxtaposed against more concrete and personal figures, such as the America's Army Real Heroes, individuals upheld as the embodiment of personal achievement in the U.S. Army. I conclude by asserting that the unreal enemy of America's Army is, ultimately, an enemy that is not exclusive to a video game, but one that exists as an anonymous specter, ever present in the militarized American cultural imaginary.”

---

Late addition, 2011-07-19. The following is a comment to this post, by Dr.C, but seems worth promotion to full visibility here.

“I am not sure I completely agree with this. At the present time the generic "enemy" seems to be more of mid-Eastern extract. There are examples where people will refuse to fly in an airplane next to anyone who is swarthy. The irony is, of course, that we invaded "them" and not the opposite.”

---

• Robertson Allen, "The Unreal Enemy of America's Army". in Games and Culture, 2011. 6(1): p. 38-60. DOI 10.1177/1555412010377321
  

Only connect

Once upon a time I spent several weeks, completely alone (long story; don’t ask), on top of a mesa with a precarious microecology, surrounded by desert. My only company was the local wildlife: mostly lizards, small rodents, and an unlikely colony of feral cats at the top of the food chain.

Those cats, as the only visible social species, became the focus of my attention for much of the time – particularly the complex network of behavioural relations that bound their tribe together. There were mentor/mentee relations. There were sibling and parent/offspring relations. There were hunting partners. There were bonded sexual pairings and temporary liaisons. There was a power hierarchy. And there was a rich web of what I could only describe as ‘best friend’ relations. Then there was also a small number of maverick individuals who appeared to be entirely outside all of this, part of the tribe, but operating alone, connected to the main only by apparently random acts of altruism or brigandism.

Studying systems of inter-relational linkage was, at that time, an important part of my day job. But computing resources were scarce in those days, and sheer volume of content and association limited what could realistically be done; things are very different today. While analysis of social networks stretches back at least to the 1940s (arguably to the late 19th century), and social network analysis (SNA) as a formal field has been well established in sociology (to which I shall return, shortly) for decades, ‘it has only recently been discovered by behavioural biologists as a useful tool in the study of animal behaviour’, to quote Amelia Coleing^* in Bioscience Horizons just over a year ago. Coleing goes on to observe that ‘...methods devised to measure social complexity in studies of animal behaviour... often reflect the social relationships between individuals indirectly... social network analysis provides formal descriptors... and by providing quantitative measures... allows testing of statistical models about relationships and structure’. [More]

---

• Amelia Coleing, "The application of social network theory to animal behaviour" in Bioscience Horizons, 2009. 2(1): p. 32.

Eggsperimenting...

I did, in the end, do the eggmobile experiment – with a group of urban café teenagers. By the time we'd thought it through, it cost even more in call time than my first sight expectation; but in the scale of things it was still a cheap enough piece of research. And it enables me to hold up my head when urging others to check facts rather than believe everything they are told.

Cellphones vary in the amount of radiation they produce at the earpiece; generally speaking, it reduces with time as design improves in newer 'phones. A thorough study would look at many different combinations, but we contented ourselves with two: a pair of the latest 3G handsets, and a pair of older (1996) ones. We also set up a control: an egg sandwiched between two inactive phones which were switched off. Given the information available on Snopes, nobody expected the egg to hard boil although the possibility had to be allowed for. More plausible was a discovery that the temperature within the egg would rise detectably. Through holes drilled into each egg, we inserted temperature sensors (one in the dead centre of the egg, one against the inside of the shell at the closest point to each 'phone). The probes were attached to a data logger, and the reliable resolution of the system was 0·1°C.

To cut a long story short ... the temperatures in the egg flanked by new phones did not change at all in the hour, even at the shell closest to the 'phone. Nor did that in the control. In the egg between the older 'phones it rose by 0·3 degrees at the shell, but not by any measurable extent at the centre.

Running the older 'phone set up again with the handsets switched on but SIM cards removed, transmitters and aerials destroyed with a hot soldering iron, and the 'phone bodies wrapped in metal foil, produced exactly the same temperature results. With a 3mm layer of thin polystyrene foam added between 'phone and egg, the temperature rise disappeared entirely. It would seem that the heating cause is simply the warming up of the handset by power flow from the battery, not radiation resulting from communication. Attaching sensors to the 'phones themselves supported this: the old handsets warm up significantly in use, the new ones very little.

Friday monster blogging

For anyone who liked Dr C's crab (last week), or my frog (earlier today), you mustn't miss the latest instalment in the saga (from Ray G) HERE.

Pay no attention to his protestations of innocence...

Morphit

At the risk of repetitiousness, it has to be said: an office spreadsheet product is not, for a wide range of reasons, the best place to do analytic work. On the other hand, office spreadsheets in general and Microsoft's Excel in particular are nevertheless the place where a large and ever increasing amount of analysis is, in the real world, done. For that reason, anything which offers spreadsheet users an independently-implemented set of analytic tools without frightening them away has to be welcomed.

Morphit, from software consultancy The Edge, is a package designed to provide scientists with such an option. [more...]

Bring back Harry O

Another brief post prompted by Nicholas Malleson, whose latest results suggest that giving burglars access to transport reduces the geographical extent of their burglary activities.

Both intriguing and logical.

Be sure to click on the graphic to enlarge it – its full detail is engrossing and (to me, at least) its patterns have a seductive music.

Malleson: Prototype burglary simulation

Just a quick mention for the paper "Prototype burglary simulations for crime reduction and forecasting", written by Nicholas Malleson (see my "Patterns in crime" post of 12 June) and available online at the link provided in the reference below.

While you are there, if you are interested in either crime or data analysis, look at the other articles as well; it's a fascinating journal.

---

• Nicholas Malleson and Pat Brantingham, "Prototype burglary simulations for crime reduction and forecasting" in Crime patterns and analysis, 2009. 2(1): p. 47.

Patterns in crime

Arguments over whether social sciences can truly be described as ‘science’ are perennial; they have been around much longer than I have, and no doubt will run and run long after I’m gone. What is not in doubt is that they are now at least as dependent on scientific computing methods and resources as their physical science counterparts – one project ... author comments: ‘I really can’t stress enough that the project might have ended if we hadn’t been given access to... the NGS.’^[1]

We will never approach the vision of Isaac Asimov^[2], whose ‘psychohistorian’, Hari Seldon made detailed predictions over millennia with an accuracy and precision approaching those of orbital mechanics. The data involved is too chaotic for even a remote approximation to that. Hypotheses can, however, be generated that are testable within bounds useful and valuable in budgetary and policy planning terms – and even an indicative pattern of association, with no hypothesised explanation, can improve the effectiveness of resource targeting. [more]

---

Image: Visualising individuals’ movement through time and space (from Malleson, 2008^[13]; GeoTime® software used courtesy of Oculus Info Inc. All GeoTime rights reserved.)

MapleSim

Working down through a series of guinea pigs, the youngest and least experienced who could still produce a functioning model with useful visualised output was an eight year old who in ten minutes demonstrated an impressive five jointed creation which she calls ‘a wiggly waggly dancer thing’. [more...]

All the world's a stage...

Ecology has come a long way from its first emergence as a distinct field of study in the 1970s. Like me, over the same period, it has moved from cuddly hippiedom to a strongly data-centred systems view of the world.

OK, so that’s an oversimplification. From another view it was a datacentric approach, and increasing access to computerised methods, that enabled and drove the newly separated discipline in the first place. Nevertheless, the engine that drove undergraduates of my generation in ecological directions was the emotional impact of Carson’s Silent Spring¹ or its successors, with the likes of Kershaw’s Quantitative and dynamic plant ecology² simply providing the means. Only later did the study itself become the point of the game.

I am finalising this during a break between sessions of a planning meeting in which Russian hosts discuss details of a major migration study with opposite numbers from every continent except Antarctica. Thirty years ago, even if such a meeting had been imaginable, the attitudes of those participants would have been different. Even now, those participants’ approaches here are subtly different from those which they display on home turf. As with many areas of scientific computing, this is down in large measure to the internet and other distributed information, analysis and communication developments: access to global data encourages not only objective analysis, but methodological comparison.

That globalisation of data also underpins the nature of the field itself. Ecology is, by its nature, interconnective; arguably, it really exists only in those interconnections and not in the nodes that belong in other disciplines. While manageability requires that ecotyping of bacterial systematics³ and global migration of viruses⁴, crop pathogens⁵, birds⁶ or humans⁷ are distinct data collection projects, all are subsets of a larger and ever growing data pool on the biosphere as a total system. Human minds are reasonably good at comprehending the two limits of that range, but the infinitely complex network of levels between is more difficult and without scientific computing would likely never be addressed at all. Since localised human impacts now constitute a major input to global habitat output, filling in those black box blanks is a priority for any ideas of a managed future.

It would be ridiculously over the top to suggest that any real approximation to a detailed whole planet ecology model (or ‘gaia model’ as some describe it) yet exists, but work towards joining up some of the dots is being done throughout the extent of that connective tissue. Particular foci of attention include critical examination of local interventions, construction of supersets from local data, and interlocal vectors. [more...]

---

1. Carson, R., Silent spring. Repr. with a new afterword. ed. 1962, London: Penguin Books in association with Hamish Hamilton, 1999. 0140273719

2. Kershaw, K.A., Quantitative and dynamic ecology. 1964, London: Arnold.

3. Koeppel, A., et al., Identifying the fundamental units of bacterial diversity: A paradigm shift to incorporate ecology into bacterial systematics. PNAS, 2008. 105(7): p. 2504-2509.

4. Nelson, M.I., et al., Phylogenetic Analysis Reveals the Global Migration of Seasonal Influenza A Viruses. PLoS Pathogens, 2007. 3(9).

5. Munkacsi, A.B., S. Stoxen, and G. May, Ustilago maydis populations tracked maize through domestication and cultivation in the Americas. Proceedings of the Royal Society B: Biological Sciences, 2008. 275(1638): p. 1037-1046.

6. PRBO: Pacific Shorebird Migration Project. 2007.Hugo, G., Population geography. Progress in Human Geography, 2007. 31(1): p. 77-88.

Matrix unloaded

Interesting lecture^* on Thursday evening, hosted by BRILLE at UWE and delivered by Stephen J Ball of the IoE. My thanks to Matt Davis, without whose suggestion I would never have been there, and to Adèle Evans whose perceptive comments sent me squirreling off along the lines of possibility afterwards.

The actual content reflected Ball's recent and current work (particularly a Political studies article^*), but there was a lot of lateral detail added by the context and more drawn in by questions afterwards. The "geography" of mechanisms which replace bureaucracy with influence networks was well demonstrated. At a time when the nation state is clearly on its last legs and in a process of unvolution, these network maps hint at the underlying skeleton of what will replace it. Lots of important collateral issues too, such as the end user and involuntary donor impacts of philanthropic action.

What mostly hooked me, though, as a dabbler in network structured analyses myself, were technical aspects of the methodology. There were a number of questions I was itching to ask, in that direction, but wouldn't have been appropriate ... and, besides, I'll get more out of them by trying to resolve them myself. Some of them Professor Ball raised; others were implicit.

One of the issues which he mentioned was "the problem of fitting it on the page" – the scale and complexity of the networks being modelled. All of his diagrams, he repeatedly made clear, were partial and fragmentary, the reality of the whole existing only in their interpenetration.

He mentioned block modelling, but suggested that his models are not at a level of complexity or scale to justify its application; I'm not sure I see that, though I concede that I lack the knowledge of his study to contest it. In any case, it was very clear that both macro and micro structures are equally important: any method of integration must not sacrifice detail.

The lecture was illustrated using descriptive diagrams produced in SmartDraw. Professor Ball's networks are descriptive models – visualised forms of conceptual linkage structures to aid exploration – rather than active simulations of behaviour. The "nature of discourse through the arcs, and the nature of nodal influence thus enabled" (Professor Ball's words, but my emphasis), is the focus of study, not its measure. Linkages are qualitative, or even nominative, not quantitative, concerning directionality of influences and the public/private formal/informal weak/strong competitive/collaborative ways in which they shape one another. With my different perspective, I was itching to see them transferred into Simile (or some other active modelling tool) – though I'm not sure how it would be done. Considering that "how" is one of the things now occupying my attention in spare moments.

A question which Professor Ball highlighted early on, and emphasised, was how to define a measure that proves a policy result. In a very real sense, these networks can only be said to exist if there is such a result. Validation depends on establishing such measures which, to someone with my cast of mind, means quantising the model – though other methods exist, such as seeking a prediction from action at input to change of state at output.

Unlike most of the networks with which I play, the predominant active potential lies not in the interconnective arcs but in the nodes themselves. Some way would have to be found of translating that into arc actions – presumably, at first sight, in the form of equation systems associated with those nodes and hidden away in submodels. Then each arc would gain much of its quantitative value from interaction of those nodal systems – but there would also be value assignments inherent in some arcs, as well. A friendship between industrialist "X" and influential academic "Y" (both of whom are nodes) would find its quantitative expression(s) through interaction of their submodels with each other and also (some system of weighting becomes necessary here, related to closeness of association) with mutual acquaintances and to a lesser extent with other nodes in the network or beyond. ("Beyond" because some nodes have no interconnection, existing in separate networks, but two networks themselves impact on the environmental state within which both operate.)

Then there is the question of effect persistence. One arc in a particular diagram represented a grant from one node (a foundation) to a single element in another (a researcher in a think tank); that would be quite a short term linkage, perhaps as short as a year or even less. Another was a degree of common membership between two nodes – two or three partners in the same merchant bank who are also founding trustees of the same charity, for example – which will tend to greater longevity. Links between influential individuals who attended the same school and university will be longer lasting still (though even more difficult to quantify). A network diagram containing these linkages is a snapshot at one moment in time, its complete validity limited by the shortest linkage duration within it.

Any quantisation of such a snapshot would have to restrict itself to seeking dynamics and rest states within the network at that moment. Study of effects over time would require longitudinal linkage between successive snapshots.

All my instincts are towards modelling a whole network as a pyramid of substructures. At the base would be the system of equations for each node – or, preferably, just one system of equations, with a different table of variable values for each. Node inputs and outputs would link to impact matrices (fairly complex ones) which, in turn, would be expressed as arc effects. That would allow both tuning and examination to be focussed at any desired level (each level being a network in its own right), temporarily ignoring the others. That would correspond pretty well to real life experience, too. Longitudinal linkages would then reflect temporal process between time sliced variants of the same network, much simpler lateral ones (the effect of one discrete network on others in the same policy environment – and on the environment itself.

It's a wonderful playground to imagine ... but an immense one. Anybody got spare high performance computing capacity, and a few years funding, that they don't know what to do with?

---

BRILLE first keynote lecture in an educational policy research series: Policy Networks, Policy Analysis and Post-Neo-­liberal Governance. Led by Professor Stephen Ball, Thursday 30th October 2008, 18.00 - 19.30, Frenchay campus, UWE.

Stephen J Ball, "New Philanthropy, New Networks and New Governance in Education" in Political Studies, 2008. click here for 2collab bookmark

Birds, bees and growing things

[Bee on Rape image ©2007 by Judith Acland]

No area of work is more closely associated with data analysis than agriculture. Statistics as a concept grew out of the need to quantify agricultural production for taxation purposes. Modern statistics as a coherent, mathematically-based field was shaped by a need to quantify agricultural investigation. Most generic methods can be traced back to an agricultural point of origin. The great landmark adventure of recent years, the sequencing of the human genome, concludes an intellectual story line that started with a statistical view of inheritance in peas. Astronomy and mathematics owe most of their history to the need to better predict seasons and the flooding of farmlands.

And it is inevitably so. No other strategic concern of the human race is so much at the mercy of a statistical universe. Within the controlled environments of manufacturing, ever smaller numbers of components escape the ever-tightening loop of quality control. In the fields, we remain concerned with the percentage of crop that can be brought to usable yield.

Agriculture has, to be sure, done serious quality control loop tightening of its own; but fewer of the affecting factors are amenable to human control than is the case in manufacturing – and the degree of control is less complete by many orders of magnitude. If production quality for left-handed silicon widgets is temperature dependent then I can easily control the temperature of the manufacturing environment within a few degrees. If I am a wheat farmer, however, I am limited to modest and partial amelioration of what the climate throws at me. The individual ear of wheat cannot be protected; my aim has to be maximising the number of ears per hectare that survive to be harvested and eaten.

Some day, perhaps, all food will be synthetic: biomass grown in orbital tanks, on nutrient broth mined from the asteroids and Saturn’s rings, then knitted into palatable forms, all under quality control comparable to current silicon chip production. Perhaps. That day is a long way off, though. More modest industrialisations, such as greenhousing, polytunnel cloching and hydroponic production are with us already, but are generally focused on market value position or marginal capacity. In the foreseeable future, a rising population will for the most part depend upon nutrition grown in dirt, under sky, battered by weather. In improving the efficiency with which we discern and exploit the patterns within that environment, computerised methods are as important as in any other area of science.

In this battle to maximise yield, data analysis is applied across a range of scales, but they can be split crudely into tactical and strategic approaches. Strategic application seeks to refine global understanding of the factors that affect success; tactical use is aimed at specific application of that understanding to a particular area, location or context. Linking the two, embracing both, is a mesh of research and modelling cycles.

All of this makes for a huge subject, and trying to address agricultural data analysis in a few pages has more than a hint of hubris. All that anyone can hope to achieve is a representative lucky dip into diverse topics. [read more...]

On drowning fishes ... (1)

Any blogger with a scientific bent, looking for a place of entry, can find an embarrassment of riches in any post by Dr C. His latest, Data, data, everywhere, enough to drown a fish..., is no exception; any reasonable person could spend a lifetime following the openings which it presents. Alas, I have only a few minutes and so will have to be selective.

One issue which Dr C raises, en passant, is the question of relative benefits between computerised and hands on data collection. This is something about which I feel strongly; I'll come back to it (later today, with luck) as it requires more time than I have at this moment and access to material stored on a server not immediately available to me. For now, I'll content myself with a small note on a sideways issue: Dr C comments that "I didn't see it [STELLA] in the list of programs discussed by the Growlery probably because it is a different field."

It's true that STELLA (a modelling software environment from ISEE Systems) is an excellent tool for getting to grips with complex data analysis. It's also true that it wasn't included in the article which Dr C mentions (Getting to grips with power tools) for several reasons. Modelling software is, as Dr C suggests, slightly different from (though not at all unrelated to) analysis.

In the case described, a model has been constructed within which the user can perform comparative analysis within an automated model. This does require either a preëxisting comfort level with the idea of software models in general or a preconstructed model with which to work. And, of course, you also have to understand the idea of comparing reaction curves to start with. (As an aside, one of the loveliest descriptions I have encountered for the calculus approach to modelling is given by Lakshmi, a fourteen year old school student, on the SCW Education site: "Calculus is when you imagine very small bits of a problem so you can get your head round it, then imagine that small bit happening over and over again, forever, to make it back into the big problem again but now you understand it.")

There are several environments of a similar type to STELLA, all of them good in different ways, and it pays (if you have the opportunity) to look around for the one which suits you. These days, UML (Universal Modelling Language) has brought them all into closer coöperation with each other. Anyone interested in my take on this area may like to read Why is Lego the best toy in the world?, written last year.

Beanbag math

Over at SCW Education, Chandra describes how to have fun with math, beanbags and a bunch of 8/9 year olds. It's well worth reading, if only to marvel at the patience required to cover a school wall with taped together sheets of flip chart paper!

Modelling the ideal

(Declaration of debt: This one came to be written partly as a direct result of my conversations in here with Dr. C.)

Plato posited a cosmology in which the variability of what see we around us results from imperfect manifestation of ideal forms. So, for example, the large black horse and the small piebald horse you pass in a field are variant images of an ideal pattern of perfect horsehood. It’s not difficult to square this, at least intuitively, with modern ideas of micromutation and genetic blueprints. It is also a view of the universe mirrored in the idea of a theoretical model – the model being an idealised image of the varying reality (whatever reality may be, once you start looking at it closely). If all of this seems whimsical, consider the work of Dor Abrahamson^[1] and others on how students learn statistics and probability concepts.

Statisticians have a very close, if ambivalent, relation to this business of models and ideals. Like sculptors working a block of marble in search of the form within, they seek to reveal the model which sits as a shining ideal behind the grubby uncertainties of real data – in fact, statistics could be defined as the quantification of deviance from the ideal. This Platonic vision is clearest in classical frequentist statistics with the attempt to place key population descriptors on one of a predefined list of approved mathematical distributions but it’s just as real, if less obvious, for Bayesian approaches. Even where models are not themselves statistical, do not themselves use statistical methods, and eschew any statistical connection, they are usually derived from origins in probabilistic investigation and their relation to reality is therefore a statistical concern.

IBM, for instance, is interested in building cerebral cortex biomimetics using confabulation theory. UCSD’s Robert Hecht-Nielsen, in a talk to IBM’s Almaden Institute on Cognitive Computing last year, emphasised that confabulation architecture contains ‘no algorithms, rules, Bayesian networks, etc’^[2]. Confabulation, however, depends on maximisation of cogency, and cogency is defined by a probability statement of relation between assumed facts, so evaluation of outcomes as mimesis is a statistical exercise in biocomparison (Bayesian and otherwise). I hope to return to this in a future issue.

Less ambitious than modelling the functions of the cerebral cortex itself, though not necessarily less interesting, is statistical study of the braincase which contains it. This is the aim of a research study, still in the unfunded preapproval pilot phase, by a young African academic. Since the work is contentious for sociopolitical and religious reasons, and nascent careers fragile, I won’t identify the researcher or university more closely than that; my primary interest here is, in any case, not the study itself but the informational framework within which it is to be conducted. It asks whether there is an evolutionary ‘trajectory’ for evolutionary development of the cerebral cortex, indirectly inductable from analogous trajectories of three hundred descriptors for the physical forms of the cortex components and the container within which they sit. [...More]

---

[1] Abrahamson, D., Bottom-up stats: Toward an agent-based “unified” probability and statistics, in Small steps for agents… giant steps for students?: Learning with agent-based models. 2006 San Francisco, American Educational Research Association.

[2] Hecht-Nielsen, R., The Mechanism of Thought. 2006 San Diego, California, USA, IBM Almaden Institute on Cognitive Computing.

[3] Mithen, S.J., The singing Neanderthals: the origins of music, language, mind and body. 2005, London, Weidenfeld & Nicolson. 0297643177 (hard).

[4] Rao, C.R., Journal of the Royal Statistical Society, "The Utilization of Multiple Measurements in Problems of Biological Classification". 1948. 10(2): 45pp.

[5] van Vark, G.N., Statistica Neerlandica, "Some applications of multivariate statistics to physical anthropology". 2005. 59(3): 10pp.

[6] Thomson, A. and Randall-MacIver, R. Ancient Races of the Thebaid. 1905, Oxford, Oxford University Press

Living with big brother

[published in Imaging & Machine Vision Europe]

I’m a scientist, using machine vision as one tool in what I do. That tool is changing almost in my hands as I use it, and throwing up clouds of important sociopolitical issues in the process.

Recent issues of IMVE have included Nick Morris’s look at the rôle of MV beyond the factory floor, his particular examination with Tom Wilkie of its place in security applications (April/May issue this year), and Tim Gillet’s piece on facial recognition (August/September 2005) which closes with a commendably balanced reference to Big Brother. Machine vision is indeed spreading rapidly beyond industry, in uncountably many ways; and we have already moved into a more effective and sophisticated surveillance culture than George Orwell’s 1984 could imagine. But I’m not here to cry doom. Yes, the social changes concern me; but science and technology cannot be put back in the bottle, and neither accusing nor defending machine vision on apocalyptically dystopic charges really gets us anywhere. Better to consider what Big Brother might look like, down the line a way, so that we can think about influencing the world into which we are heading.

When considering the unknown, we need models if only as departure points, and fictions are all we have for the future. I have invoked the ready made token of Big Brother (from George Orwell’s 1984) but you can pick from a plethora of alternatives. Our present (in the developed world, at least) owes more to Huxley’s Brave New World, and even that isn’t a very good fit. The DNA scanner above, omnipresence of ANPR (automatic number plate recognition), smart tagged clothing, and pharmaceutical solutions everywhere, point to Ira Levin’s This Perfect Day; but, again, only up to a point. The symbiotic human/machine society of Iain M Banks’ “Culture” novels offers another view. There are combinatorially many avenues which future gazing can validly take, but I’ve settled on two which seem likely to become reality within my working life and materially impact what I do.

Visual approaches are only the start of it. “Imaging” is an inherent part of the hardware/software package which is the human window onto the world, but “machine vision”, while convenient, is a misleadingly restrictive tie to biological metaphor. I’ve just been to see the prototype of a machine which may one day sample your DNA for identification purposes as you pass through a controlled entrance (ticket gate, primary school, supermarket). DNA comparison may not seem to be “vision” – but think of films and TV dramas where the moment of revelation is invariably conveyed by visual overlay of two transparencies.

“Vision” has two parts. The physical eye or camera specifically receives light (or any elecromagnetic radiation), but the software behind either is omnivorous, not necessarily concerning itself with modality. Software processes input into the most useful form. Spend any length of time forced to function meaningfully in darkness, and you will find your brain beginning to form images from the sound, touch and smell stimuli which it is receiving. “Machine vision” is partly an artefact of modern emphasis on eyes as highest grade sensory input, partly of recognition that visual information is more readily susceptible to productive algorithmic handling than, for example, tactile or olfactory input. As methods advance, the distinction between inputs will become purely one of availability and quality in context, the camera being only one tool in a larger rack, and this magazine will naturally evolve into something like “Machine Perceptual Imaging, Europe”.

As machine vision spreads throughout our technological infrastructures, it does so on an ad hoc basis. Today your client needs a site security system; yesterday it was a CCTV system at a football stadium; tomorrow it may be high speed capture cameras for a production line monitoring solution. I, meantime, am applying similar methods to study of malnutrition effects, the behaviour of ants, roadside impacts. But since the rise of the internet, data doesn’t stay ad hoc for long.

Biological sensory inputs do not remain separate. Your visual, auditory, olfactory, and tactile inputs on entering a room, for example, are immediately synthesised into an integrated impression of that room. At the same time, both inputs and synthesis are compared with a lifetime’s catalogue of other situations for signs of threat or advantage. The new input data are simultaneously being added to that catalogue. So it is with machine inputs: we use captured industrial, social, or scientific data in the same way, combining primary sources with secondary to derive value added from our investment in collection. This used to be a deliberate manual process, but increasing connectivity has made automation more and more feasible – not only automated comparison and integration, but automatic selection of what is to be compared and integrated. In a research lab recently I was introduced to a software agent which tracks down any accessible video footage of multiple objects moving in complex patterns (termite colony, shopping mall, pebbles in a tidal race, billiard table, dance floor), analysing and cataloguing patterns before adding the results to a database without human intervention. Commerce similarly mines disparate databases for relationships between customers or behaviours of which those customers themselves are unaware. In industrial context, with many variables being automatically monitored, the logic of automated and ongoing analysis is inescapable.

So, gradually, both local networks and the internet will increasingly resemble autonomous entities – with machine vision hardware as their eyes. Your client’s site security, CCTV, or production line monitoring will not be able to ignore the synergistic advantages of tapping into larger entities. Just as disparate CCTV systems have been integrated into large scale “big brother” ANPR networks in return for security payoffs, so your client will reap the benefits of at least partly submerging local systems in these extended structures. Your machine vision imaging will not, in many cases, be a separate operation but one of many eyes for a larger one with considerable autonomy.

The scope of the integration be will not be limited to industrial input: the same type of mutual back scratching will link you into global input, analysis and catalogue sharing of all kinds. And vision not be a particularly separate sense of that “big brother” organism of which you will be part: it will be part of a holistic undifferentiated sensory array stretching from thermometers to DNA scanners.

While I don’t suggest that this large machine entity will be “aware” any time soon, AI work and algorithmic theories of mind are progressing at a rate which makes any fixed view insecure. Keeping an eye on the likes of Sourceforge’s AI mind project, the American Association for Artificial Intelligence, and similar sites.

But as electronic information handling becomes more efficient and more compact, it will inevitably be concentrated into more localised centres where that offers advantages. Some data, commercially or militarily sensitive, will never be shared, and some analyses are needed very rapidly as realtime feedback into control of volatile processes, which will increasingly suggest the use of integrated perceptive machines instead of human beings.

Honda has a number of humanoid ASIMO robots with sophisticated machine vision. While currently childlike in size and restricted to cute demonstration tasks, they demonstrate generic capabilities which could well develop into data capture and intervention platforms. Obvious first points of application would be hazardous industrial contexts requiring flexibly mobile sensory perception linked to manipulation; extension to other industrial, scientific, military or social rôles takes little imagination. Many of these small entities will, despite their need for local autonomy, be mobile avatars of networks and thence of the global structure, though some will be largely (or completely) self determining.

Knowledge has long been too large for any human individual to contain or contribute more than a small fraction of it; much of it will exist in machine entities without connection to any human individual at all. This is Big Brother being born, for better as well as for worse, and machine perception (vision and otherwise) is at its heart. We need to think ahead: how will we make use of it, rather than just fall into it?

---

How does it work?

Recognising any object involves a variety of mental tricks, but boils down to comparing features of the image formed with those of images previously stored. A face is one of the trickier recognitions, because it moves so much (particularly with expression), is seen from multiple viewpoints (as the head turns and tilts, or with camera angle) and changes with time. The human brain often picks on transient features such as facial hair, but designers of machine software aim for invariants such as distance between fixed physical points: eye centres, top of cheek bones, and so on.

Numerous software approaches deal with this variability, but Hidden Markov Models (HMMs) are important. A Markov chain (from mathematician Andrei Andreyevich Markov, 1856-1922) is a type of stochastic sequence modelling change in system states; an HMM attempts to do this when vital information is unknown and must be deduced. An early application, before machine vision, was to speech recognition and HMMs played an important part in the development of DNA sequencing.

Much current work concentrates on reducing the variability of surfaces to be recognised (the face, for instance). Michael Bronstein’s work with multidimensional scaling and photographic deconvolution at the Isra’el Technion(http://www.cs.technion.ac.il/~mbron/research_facerec.html) is an example. R&D work by the likes of MERL (Mitsubishi Electric Research Laboratories) is focussed on principal component analysis of multiple laser scanned images from the same face. Where the person being recognised coöperates (inn door security systems, for example) the subject can be asked to assume a neutral expression and look at the camera, eliminating many problems orientation and viewpoint. One approach, used for example by A4vision, is to project a grid and analyse the ways in which it is deformed by the contours of the face, thus avoiding much uncertainty inherent in images of the face itself.

Why is Lego the best toy in the world?

With the above question, 14-year-old Sophie Amundsen (protagonist in the teenage best seller Sophie's World) is invited to consider the atomic nature of matter as the first step in a journey beyond the limits of her imagination. Lego (or Meccano, or any similar construction toy) can be an equally powerful metaphor for other ideas, scientific and otherwise, leading the way to other journeys. Modular reusability has always been a virtue and a refuge from insanity, at least as far back as we can peer into human history and probably long before that. And if some of the bits can be left assembled into higher level substructures, reusable later, so much the better.

Way back in the days when a large organisation paid me good money to sit in a windowless room modelling sociomilitary scenarios...

Read more ...

Exploring energy: easing the way to fuels paradise

Depending on one's viewpoint, nuclear power can be the salvation of the biosphere from heat-death, or a fast route to the dystopic futures of Judge Dredd or Riddley Walker. Fuel cells may be the answer to our petroleum fears or the successor to them. Wind power used to be the golden path for environmentalists everywhere, but it has acquired a new face as the despoiler of beautiful landscape and murderer of endangered bird species. Similarly dramatic dichotomies afflict solar, geothermal and wave-power sources.

The underlying common currency of the debate is energy economy... [Read more]