Exploring PossibilitiesThose two words form the far from simple answer to some fundamental questions. Observation and experiment at various levels has led me to realise that Life, the Universe, and Everything is ultimately just exploring possibilities. Asked what I am doing the answer is the same, as it is when asked what you and I should be doing. The consequent question is How? not Why? Any answer to the latter can be no better than "because that is what we/they do." To get good answers to How? you really need to pay attention, that every day more scarce currency. Firstly you need to take very seriously the take home message from the models reported in Stu Kauffman's The Origins of Order. These show that when you are in unexplored territory the winners are those who jump far and land on a productive spot, but when you are already on your way up your mountain the best strategy is to move incrementally, following the local gradient at each step. The implications of this can be developed at way too much length for here, but clearly include the notion that if you want to keep exploring possibilities, it helps if you can usually find your way back to somewhere close to where you have been before, some home base conditions even more than location. Secondly, and one of the purposes of this site, it might be useful to take into account some of the initially unfamiliar implications of the results I have been getting from experiments with toy systems, a habit which goes back long before I started playing with Ed Fredkin's replicating cellular automata in 1983 and which is clearly getting worse with age, in part provoked by Stephen Wolfram and more recently facilitated by the wonderful work of the Golly gang. Foremost is the observation in ever more intriguing guises of the creative synergy between (even deterministic) chaos and emergent order. Chaos and order are not opposed. You need both. |
On this page |
On other channels |
Emergence/Systems/ComplexityAround 1986 I had reason and opportunity to discard all previous assumptions and start a fresh look at the question of how the world really works. That search quickly led me to what at that point of time it was fashionable to call complex systems, a field I have come to learn has had as many names as it has directions of entry over at least my life time (post WW2). More recently I have been happiest with emergence as the umbrella label for N N Taleb's black swans and Rumsfeld's unknown unknowns, in part because emergence allows proper celebration of Darwin's endless forms most beautiful across all the great domains that feed our curiosity. The essence of emergence is that it is not strictly predictive but rather a way of dealing with the many surprises that are both consistent with natural laws and the outcome of some previously untested combination/history. You study emergence through analysis which may lead to a degree of statistical predictability over time, but which, like plate tectonics, biological evolution and much of astronomy is primarily a tool for making sense of observed patterns after the fact. Such analysis is neither strictly science nor anti-science. To it science is a toolset, in much the same way that math is a toolset to science. Since the early 1990s when Liddy Nevile showed me a 1986 paper by Marcia Salner, I've been particularly motivated by the question of how we might get a critical mass of younger people to take onboard the generality of emergence short of waiting for them to accumulate a lifetime's experience or suffer the two crises of confidence that Salner identified as necessary to progress from the naive belief that right and wrong are knowable, via relativism, to a general systemic understanding; neither of which route comes with guarantees. Participation in ISCE's 3rd International Workshop on Complexity and Philosophy in Stellenbosch in early 2007 and my subsequent analysis of the workshop papers for presentation at Complex '07 on the Gold Coast highlighted the great double focus of the field, a separation within a coming together that is equally pervasive within Melbourne Emergence Meetup Group. That separation can be most easily described as between the natural and social worlds, with very few of us truly seeing the consistent whole. Following motions adopted at the monthly meeting of the Melbourne Emergence Meetup Group the Kororoit Institute Proponents and Supporters Association was incorporated on 15 April 2011 to provide a more formal platform to pursue opportunities that have developed through the Meetup Group, Bill Hall's TOMOK and, particularly, our Putting Community Knowledge in Place program discussed below. The Association is moving rapidly towards operating as a Pre Institute. |
Putting Community Knowledge in PlaceThis was the theme of a Special Session some colleagues and I put together for the Melbourne 2010 Knowledge Cities World Summit in November. It remains a working description of a program we are fleshing out with origins in the Melbourne Emergence Meetup Group and Bill Hall's TOMOK. (There were and are several other people actively involved.) For me this grows out of the struggle of the Moonee Ponds Creek Co-ordination Committee to have community knowledge efficiently brought into a prolonged strategic planning process. It also draws on involvements in the problems of the City of Brimbank inherited from my late mother's St Albans History Society and an extensive network on local interest organisations those connections lead to. Right now, I'd like nothing more than to be able to make practical use of my unique combination of knowledge of Mount Defiance from hiking the trails across the summit to diving the bommies below the recent rockfalls and lookout to help facilitate the local member and transport minister get an outcome for the next 75 years of this issue which was never in The Plan, despite being a long obvious risk to our greatest tourist asset. More broadly, the aim is to glue together human and technical systems to encourage the collection, curating, utilisation and recognition of information currently only residing "in the heads and bottom drawers of non-traditional owners" so we can get better supported planning and implementation outcomes. The complementary benefit of our participation was to sharpen awareness of the conditions for successful establishment of knowledge precincts, the core mission of the Summit, an objective which allowed its judges to choose Melbourne as the world's leading knowledge city, albeit without competition from North America. The insight of Summit co-chair Klaus Kunzmann, who I first heard at GAMUT preconference, broadens the lessons from our continuing misadventures with Docklands Science Park. |
Experiments with Discrete SystemsPrecursors: Developing templates for sporting fixtures under complex constraints were the most practical application of a younger lifetime dabbling in combinatorics and adjacent fields, including a fascination with computationally intensive random sequence generators. Pattern Breeder: A generalisation of Ed Fredkin's pattern replicating XOR rule cellular automata (CA) released January 1986 at Macworld Expo San Francisco and described in Scientific American that September. A 25th anniversary demo version runs as a Golly Perl script. Life in a Tube: While Conway's ubiquitous Game of Life CA (GoL) settles down to static and cyclic patterns on a 3D plane, when constrained to a narrow cylindrical grid a couple of simple engines can drive endlessly complex forms. Discovered by chance in early Mac days and returned to with various new GoL implementations. agars: While GoL patterns settle on an otherwise empty grid, much more diverse behaviour can be elicited by filling the grid with repeating stable or cyclic patterns before minimally perturbing it. Such experiments provided grounding in period doubling and similar epiphenomena of more active CAs. Trapper: A small region blockaded at each end arises frequently in Life in a Tube with any tube circumference up to ten cells and is equivalent to several 1D 3 colour symmetric neighbourhood CA rules. Analysed traps with widths up to 130, identifying some repeating with periods over 4 billion. Tick Tock: Moving beyond spatial grids, a minimalist rule acting on the nodes and edges of a simple graph to produce a completely new graph each iteration. These quickly show the emergence of persistent structure which was not visible in their seed patterns. LivingOnTheEdge (LOTE): Generations CA rules extend the better known 2 colour rules such as GoL's live and empty states by adding what is effectively a dying state that persists for a number of iterations, ignoring and being ignored by its neighbours. Generations 345/3/6 was identified and named LOTE in 2000 but neglected until Golly 2.0 added Generations support. I found LOTE's development from simple seeds qualitatively more interesting than other CAs and over two years and three computers have run 500 such seeds to 100,000 iterations and many beyond, developing software to control and snapshot such long runs, to render and archive compressed and enhanced images of whole patterns and to pan and zoom animations of fine details. WMPVN: Late in 2010, an opportunity to make a presentation about Golly to the Australian Open Source Developers Conference (OSDC) coinciding with the release of version 2.2 had me testing some of the newer features and discovering how easy it was to build rules for my new weighted Moore plus Von Neumann neighbourhood which promises numerous comparators to LOTE. For updates see Cellular Automata Project news. |
Book recommendationsEssentialsA Different Universe: Reinventing Physics from the Bottom Down by Robert Laughlin. One of the best popular books on physics ever. Why complexity is often irreducible. Why of the 24 different solid forms of water, none were predicted in advance, but all were readily "explained" after the event. Why the "fundamental" laws of physics are no such thing. Why there is no such thing as empty space. Will change the way you think about the world. (Thanks, Tim.) At Home in the Universe: The Search for the Laws of Self-Organization and Complexity by Stuart Kauffman Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution, and Epistemology by Gregory Bateson Out of Control: The New Biology of Machines, Social Systems and the Economic World by Kevin Kelly The Nurture Assumption: Why Children Turn Out the Way They Do by Judith Rich Harris A World Without Time: The Forgotten Legacy of Godel and Einstein by Palle Yourgrau Snowball Earth: The Story of a Maverick Scientist and His Theory of the Global Catastrophe That Spawned Life As We Know It by Gabrielle Walker Animals in Translation: Using the Mysteries of Autism to Decode Animal Behavior by Temple Grandin and Catherine Johnson Recent*David Copperfield and Oliver Twist by Charles Dickens. Continuing the Guttenberg-inspired theme of seeking a deeper feel for how worldviews have changed since 1850. The Malay Archipelago, the Land of the Orang-Utan and the Bird of Paradise; A Narrative of Travel, With Studies of Man and Nature by Alfred Russel Wallace. The context for Darwin and for Indonesia. The Adventures of Tom Sawyer and Huckleberry Finn by Mark Twain. Explains Sarah Palin. I obtained these for free from the iBooks store and read them comfortably enough on my ageing Touch with my ageing eyes or, more recently, on my newer phone on trains, but I haven't found a way to link to those editions so I've substituted Amazon. My
|
Information technologiesProperly trained in Systems Analysis and Design at the beginning of my career, together with enough assemblers and other old languages that I have retained a solid understanding of how things work underneath and consequent appreciation of how many miracles underpin our too easily accepted and seductive modern information technologies. Nowadays I'm only fluent with Perl and MySQL, though when needed I can muddle my way through other cornerstones of the interweb. After a career that led me into contract programming early, in late 1981 I was contracted to design the server-side system for Australia's first online information service and quickly became captive to the cyberspace dream, marrying that to the Macintosh user interface and from mid-1984 getting serious about trying to develop a graphical Public Information Communications and Access (PICA) system. Though we were still eight years too early because of the immaturity of essential infrastructure, we soon learnt that we were some way behind the first to start in that direction. After the detour which had me convening the first meeting of independent PostScript developers in early 1987 in San Francisco and contributing the final chapter to Roth ed's Real World PostScript, in the early 1990s I enrolled for an MSc science-technology-society program offered by the University of Melbourne's department of History and Philosophy of Science, in part to ensure I gained the earliest possible access to the internet. And so I became the IT voice on an ever changing consultancy team providing education technology policy advice. Having twice sworn off ever learning another programming language and having received no encouragement to persist any longer with my long pursued goal of processing cricket results online, in 1997 I acecpted an invitation to move to Sydney which gave me the space to pursue connections with those who already were online, get roped into forum maintenance and eventually develop my own TransForum software in Perl which turned out to be the best possible language for an ageing programmer who appreicated the inconsistencies inherent in real world data collection. Returning to Melbourne, from 2000 I have applied Perl, MySQL and other cornerstone technologies to backend systems for inhouse and client-facing websites, and Perl in particular to my own research into discrete systems. |
PublishingWhile the publishing urge had already hit me by 6th grade, my first lasting project was a seasonal weekly newsletter with Grant Edmondson for Strathmore Football Club, More News, launched in 1970 and still going strong. This taught me the basics of type, paper, ink and deadlines. Realising I did not then have the contact network needed to get into business in IT, in 1982 I joined Computerworld Australia to write industry news and technical analysis, progressing to their microcomputer magazine and then Australian Macworld. Of the first two LaserWriters Apple brought to Australia in 1985, one came to my desk for review, quickly having me teaching myself PostScript and soon making PICA Pty Ltd an early leader in the introduction of desktop publishing to Australia. This led to providing technical support for various publishing projects including several books due to my mother's role with St Albans History Society. During the mid 1990s, I contributed to and had final content assembly responsibility for a series of reports on education technology policy for Australian government agencies. ![]() At the start of 2000, Joe Selvaggi and I were recruited by a company that was to merge with and become Niche Media, the by then and since publishers of Australian Macworld, to bring their print properties into the dot com bubble which was pricked as we were getting started. So we finished up taking away what we could save of our efforts to form the core of our boutique hosting business. This also provided an introduction to a publishing services business for whom I initially documented their IT requirements then stayed on to help implement them across much of the decade. In 2008, my long preserved entity for personal business dealings, Meme Media, published my mother's and my friend Heather Vicenti's stolen generation autobiography. |
Steal my slogansOur systemic servants do not good masters make. Money can't be saved. Neither your faith nor your job absolves your responsibility. Those who live off the margins should be required to live at the margins. My respect for your belief is proportional to the effort you have made to substantiate it. Only the young die good. Giving thanks to the space, time, energy, matter and other lives that have allowed me to tell my lies on this old and damp ball of rock. |
Community involvementsHaving done more than anybody's share of formal committees and associated record keeping, nowadays I have a strong preference for agenda-free gatherings, but I keep getting dragged back, not always unwillingly when it is something close to the heart. Having not long returned to what had become Friends of Moonee Ponds Creek, I was soon drawn into issues with the future of the Moonee Ponds Creek Co-ordination Committee (MPCCC) which had for a decade served the mutual interests of the Friends and the municipalities the Creek runs between, with most of the councils having grown environment departments in the interim. The wash up is that I'm in my second year as President of MPCCC, we have managed to move the MPCCC's records out of storage and into shared space at Melbourne's Living Museum of the West, our working party has just signed off on a much overdue Strategic Plan, and we now need to get serious about considering MPCCC's future role. When I inherited and decided to stay on in my mother's house, I also inherited most of the records from her involvement in St Albans History Society and have been able to stall starting to get them sorted until the History Society too gained access to that space at the Living Museum to accommodate a digital archiving setup acquired through a couple of grants. In the interim, History Society members led yet another defence of Errington Reserve's retention for public recreation. This has also led me to take a wider interest in local affairs within and around the politically troubled City of Brimbank, including as a member of the City's Economic Development and Transport Committee. On June 1st 1982, on my first assignment for Computerworld Australia I made contact with several people who have had major roles in my life since then, none more so than Bill McPherson who initially invited a group of us to join him on Tuesdays for dinner at his restaurant on the corner of Errol and Victoria. After this group had nurtured several projects, it took on the name 17 Group during an attempt to extend its reach to those for whom Tuesdays might not be convenient by meeting on the 17th of the month no matter which day it fell on, at one of which I presented my Conversation Piece which was to provide the theoretical grounding for TransForum. A core group continues to meet on the second Tuesday of the month at a cafe a block from Bill's old restaurant. I was an established regular at Melbourne Perl Mongers when we decided to run what became the first Open Source Developers Conference in Melbourne in 1994 and have only missed one Australian OSDC since, presenting at several and being a fallback member of what became the supervising OSDClub committee when there was a shortage of candidates. After I returned to Melbourne and accepted I could not even think about returning to my previous level of commitment to local sporting organisations, I made some conscious decisions about the best ways of preserving sufficient connection to my past, one part of which was to regularly attend the Sydney Swans' Melbourne games. The other was to go back to spending the core summer holidays at Cumberland River which had been a central part of my life before the floods of 1983 washed our camp out to sea. It provides an important connection with old friends, bush walking and diving as conditions permit, photography, videoing and time for reading. I get back to Strathmore Sports Club a couple of times each footy and cricket season, plus finals and funerals, but still hardly feel like I've been away. Unfortunately, there is no comparable way to visit the North West Cricket Association nor the former Cricket Union of Victoria and too much of their history is almost certain to die with me, especially the idiosynchratic techniques I honed for grading competitions and drawing up highly constrained fixtures. |
Cellular Automata Project news11 June 2011: Having at least sampled all 64 of my target WMPVN rule families decide I should also look at the balance of the 16 Generations rule families that I long ago declared comparators and promptly discover a 612,750 iteration Methuselah in 3458/38/6 which just might be one for the record books. 3 May 2011: The surveys run, it was back to searching more of the target rule space. Topping a run of novel mechansisms, WMPVN-45678c_459ac_16 proved to be a not entirely unfamiliar p14s39t space filler, but with a difference. Unlike other space fillers, the rule is just on the active side of the settling versus chaotic core threshold, with a polarised p168s700t reprocessor boosting the otherwise even slower growth of core within more rapidly growing 56x56 settled grids. This was the kind of rule I'd been hoping to find since early days of studying eroding agars in Life. 7 Mar 2011: Checking neighbouring rules reveals that the shoulder growth mechanism also operates in WMPVN-45678c_459_17, producing a much sparser internal pattern of gutters and braids. Population grows around 1.75 times faster than the first found rule and a seed production run leaves us with 30 as against the 14 already run to 100,000 for the first one, so the next couple of months promise to be dominated by completing surveys of this new rule and the replicator rule, the latter still having more than half its 50 to finish. 19 Feb 2011: Finally put work on Generations 345/3/6 on hold after a last round of renderings are uploaded and a few links added to the placeholder site where I still intend to add extensive commentary, if not new data. Within hours discover a widening puffer engine in WMPVN-45678_459_17 spectacular enough to trace its unremarkable origins and run from there in isolation to add the 20,000th iteration to my wallpaper collection. 7 Feb 2011: Completed first pass identification and mapping of mechanisms which most influence large sub-sections of the 45678xx/45xxxx/n rule space, colouring blocks of a progress spreadsheet accordingly and making the still numerous gaps easier to identify. 28 Jan 2011: Discover Trident and Burner which dominate two of the 16 rule families where flutterby otherwise reigns, in Trident's case for rules with 13 or more states. 22 Jan 2011: Running the last of the first 50 viable seeds found for 45678c/459abc/12 building some first rough statistics on the development of an expanding skin of replicators surrounding chaotic core and the less common patterns which penetrate such skins. The original delta engine which anchors the spectacle of the one viable seed initially found for 45678bc/459ab/24 makes a persistent reappearance forming a comb off the third scrum line reactivation. 19 Jan 2011: Render the last significant Generations 345/3/6 pattern after it has run to 200,000. 16 Jan 2011: Up in the early hours for what was expected to be a quick rollover and hit a run of surprises. The import of finding p163t Fredkin-style replicators in 45678c/459abc/12 had still not sunk in while I was posting an explanation of my 2007 Stellenbosch Cosmogenesis schematic to EDU-Talk later in the day. 14 Jan 2011: A single long lived but ultimately settled seed in the first batch tested with 45678bc/459ab/24 has me trying head on collision seeds and turning up one that is viable and about to reveal a spectacularly rich collection of mechanisms. 13 Dec 2010: Present summary of plans for the Cellular Automata Project to Emergence Meetup, especially the WMPVN and Generations territory to be focused on to leverage experience with 345/3/6 and one late discovery under 345/3/6. 6 Dec 2010: Left-right reversing "flutterby" traveling 13 cells diagonally in 58 cycles seen frequently in what turns out to be a quarter of my target WMPVN rule space, or at least the part thereof with 16 or more states, producing a common background of central core and eight radiating compass points, the diagonals somewhat shorter than the four primaries familiar from other rules. 5 Dec 2010: Having noticed a simple pattern between the generated rule trees between adjacent state numbers, complete and start using script to generate sets of rule trees for WMPVN rules. Start what turns out to be the last, at least for now, Generations 345/3/6 seed run to 100,000. 26 Nov 2010: Present paper at OSDC Golly: how an open source project took over the study of cellular automata which has had me exploring a bit more deeply, in particular the use of rule trees to implement a "weighted Moore plus Von Neumann" neighbourhood in which the directly adjacent cells count 2 and the diagonals count 1 giving a survive/born range of 0-12 bits c.f. Generations' 0-8. |
DisclaimersBeyond the brief summary far above, the central importance of emergence is being addressed separately and is just assumed here. The importance of cellular automata is similarly assumed, having been dealt with by Wolfram in 2002, though in a way that may not have won many converts and underplaying both the excellent fit between human visual perception and 2D CAs and the potential to define more navigable rule spaces. This page first uploaded 30 January 2011 with underdone layout and links. First update and external linking 14 February. Several links on this page are to placeholder pages or worse and are included as reminders to self that it is past time to do something about them too. |