These are notes pulled from Anselm's original at http://www.hook.org/anselm/essays/20050820.txt, posted here to encourage feedback and discussion.

-- community modelling

• what would it take to build a collaborative modelling tool for communities and watersheds?
• something that lets you and your friends pour in data
• and pour in your best understanding about relationships between elements in your community
• and then actually simulate forward in time to explore possible near term outcomes

• my hope here
• i'd like to get suggestions and input about how to build community modelling tools...
• this talk i hope is actually more just of a roundtable; just a way for us to brainstorm
• i really don't know anything about this domain; i just want to try and help promote a new community focused around this proposition.

• some things that should scare the crap out you
• http://www.rollingstone.com/news/story/_/id/7203633?rnd=1124583200019&has-player=true&version=6.0.12.1059 (the long emergency)
• http://www.amazon.com/exec/obidos/tg/detail/-/0670033375/102-7787066-0659334?v=glance (jared diamonds collapse) (a well reasoned look at several ancient cultures and how they adapted or failed to adapt)

• carbon is a bad one by itself:
• http://www.globalchange.umich.edu/globalchange1/current/lectures/kling/carbon_cycle/temperature_2050.jpg
• http://www.globalchange.umich.edu/globalchange1/current/lectures/kling/carbon_cycle/carbon_cycle_new.html
• http://rst.gsfc.nasa.gov/Sect16/Climate2.JPG
• (shows the deep correlation between co2 and temperature)
• (temperature follows co2 but with just a 10 year delay or so. Right now we are exponentially climbing co2 ladder)
• (thus the implication is that there will be a radical climb in temperature fairly soon).
• http://www.scs.carleton.ca/~schriste/data/Carbon-Atmosphere-Mass_files/State%20of%20the%20World_28025_image001.gif
• (more of the same)
• http://earthobservatory.nasa.gov/Library/GlobalWarming/Images/temperature_vs_co2_rt.gif
• (more of the same)

• and it is just one of several radically shifting trends:
• forest reduction
• fishstock collapse
• diversity elimination
• population density and reliance on mechanized (oil based transport) to keep our cities alive

• effectively we are nearing the end of the kinds of civilizations we cherish and love
• we face a computational limit; we cannot think fast enough to think our way out of this problem
• can software tools help us reason better as a collective?
• can software tools do this fast enough to help us avert the collapse of civilization?

• we _need_ to face the fact that the way we live is about to end:
• our standard of living is already starting to face a long slow decline; working harder, achieving less, having less security
• many of the same horrific things that we heard about in ancient cultures; tyranny, war, starvation will happen to us and our children
• instead of wanking about novelties, or how to create personal wealth or retreating to island getaways...:
• we need to start solving this problem

• here is an earlier post re these ideas with some citations and comments by other people:
• http://lists.burri.to/pipermail/geowanking/2005-August/001835.html

• emerging solutions would probably build on the finally maturing social geo / locative movements - exemplified by say:
• housingmaps.com
• platial.com
• plazes.com
• tagzania.com
• placewhere.com
• etc etc etc...

• of course there are well known historically similar domains and toy examples
• http://ode.org <- 3d physics modelling (i provide the web hosting for this project)
• modern weather modelling
• simcity
• some recent looks at economy management in MMORPG : http://ad-rag.com/123417.php

• and lots of formal domain studies - although these are not "open" and thus not reproducible (but can serve as insight):
• classically one isolates say hydrography or economics or ecology or sociology and model that domain only but there are some cross domain efforts typically grounded on top of gis systems
• http://atlss.org ; modelling the real cost of draining south florida everglades; inteference in food web pyramids and subsequent deferred and hidden financial impact on human communities.
• http://physicsweb.org/articles/world/13/2/3/1 - modelling fish stock reservoirs to better understand the real world behavior. As we know fish stocks have collapsed around the world, destroying what could have been sustainable resource, destroying livelihoods and creating huge economic costs and unused labour.
• some military bases do modelling on typical sandhill crane and turtle activity to predict locations and to minimize base activities from interfering with these endangered species.
• Artificial Anasazi
• http://www.casa.ucl.ac.uk
• http://biota.org ; synthetic ecosystems and organisms modelling; a great grass roots community
• Santa Fe Institute and the whole ALIFE movement ( Tierra being the classic example ). The ALIFE proceedings are highly recommended reading (especially the earlier journals).
• Upcoming book on cities and complexity: http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=10627
• "Critical Mass" by Philip Ball
• "Geosimulation : Automata-based modeling of urban phenomena"
• http://jasss.soc.surrey.ac.uk/JASSS.html
• http://comdig.org/

• why try do as an open process?
• note that web does seem to be shifting in a time sense towards more timely information with broad coverage
• altavista (large coverage but weeks old)
• google (large coverage and days old)
• feedster (reasonable coverage and hours old)
• sms/irc (narrow coverage and instant)

• also the web has shifted in direction of increased 'fulfillment':
• web 1.0 -> largely static content
• web 2.0 -> intelligent filtering (even using mechanisms such as tags) and more social
• web 3.0 -> fulfillment; just in time services, finding hiking buddies when at trailhead or looking at map of

• so web may be natural way for people to collaborate over social models; in this light:
• the web is a kind of nervous system
• we are both endpoints of observation about our local systems
• and local points of computation; that can individually submit our own intuitions, models and understanding
• a level playing field is created where insight from an individual or corporation can be evaluated on its own merits equally.

-- approaches and tools

• user interface
• some kind of spinny globe
• you can zoom into an area and submit new observations about say
• wildlife observed
• local temperatures
• a noted hypothesis or relationship between two things ( water level and fish stocks or pollution level and fishstocks)

• perhaps you can drag and drop say a factory into a watershed and test its impact
• it is an object with inputs and outputs
• water consumed
• electricity consumed
• pollution output
• money output
• products output
• observe effect on animal habitat (correlated to pollution and say noise disruption and land disruption)
• observe effect on economy (due to increased dollar resources)

• let people add their own novel kinds of factory objects or processes or ideas such as say:
• an invention: a rooftop garden kit
• a policy: swapping out personal vehicles for shared vehicles
• a policy: structuring communities around sharing child care resources etcetera

• and then you can explore outcomes:
• what happens if you scale up a policy over an entire city? does it help? hidden side-effects that cause problems?
• and maybe even measure bottom lines; like say global warming etcetera...

• ultimately one wants to be able to enter observations and relationships

• computational aspects

• one also wants to train any model against historical data to get it to be as accurate as possible (similar to financial modelling)
• since this is a largely geographic problem there probably needs to be some kind of cellular automata based model
• it may be too difficult to use a calculus approach such as integration linear equations... cellular automata can do similar predictive work

Sistema de revisão feito pela  Weblocal hospedagem de sites!