Open joemudge opened 10 years ago
jczaplew
commented
10 years ago I love this idea and would be excited to help build a prototype. The general concept of gamifying voluntary data collection can be applied to myriad disciplines, and it would be valuable to create a framework/process for developing such an application.
I also lack experience in game development, but could help on the programming side.
joemudge
commented
10 years ago Great! Thanks for your interest. I realize this is a bit of a large and complex project. I think we can make progress on this idea without the assistance of a game developer by making a 'skeleton' of an app that can:
jczaplew
commented
10 years ago Some references that describe similar projects - http://mobile.wisc.edu/news/field-research-project-webird/ and http://mobile.wisc.edu/teaching-and-learning/field-research-plant-identification-at-the-biocore-prarie/
joemudge
commented
10 years ago Nice find. This Mobile Learning Incubator group at the University of Wisconsin seems like they're doing some pretty interesting stuff. Between the birding app and the plant id app, they seem to have a lot of the pieces of what I was originally envisioning already worked out. I think the key difference between what I had in mind and what they're working on is the component of mass collaboration of image identification. The WeBird app seems to require users to identify their own birds, which limits the usability of the app to avid birders. If users could take pictures of unknown birds, then have those birds (eventually) identified through a mass collaboration process where users see a few pictures of birds from a reference collection and are asked which reference species image the unknown bird image looks more similar to. This would allow kids to learn about birds while they're participating in the mass collaboration effort, and also when they eventually have their own unknown bird image identified by the crowd.
joemudge
commented
10 years ago The code for plant ID iOS app from the Mobile Learning Incubator at UW is available on GitHub here: https://github.com/UWMLI/FieldResearchTool I this code could actually be a good starting point for the mass collaboration image ID side of things. If we could generate descriptions of species characteristics that even a kid would understand, we could get each user to arrive at a possible species ID for each image they evaluate. These individual IDs could be validated by getting multiple users to come up with IDs for each image and using the most frequent ID as the confirmed ID.
wclenhardt
commented
10 years ago In case you haven't seen this, Project Noah is somewhat similar although not quite a game. Might have some insights there. It started as a standalone app, then was taken on by NatGeo.
joemudge
commented
10 years ago Thanks for sharing projectnoah.org . Those folks seem to have a similar concept and motivation as what I had in mind, however by adding a game aspect, I'm hoping to attract players that don't necessarily already have an interest in nature, which hopefully would provide a broader user base. They might be worth contacting at some point to see if they'd be interested in sharing a common database on the ecological data collection side of things. It seems like it would be silly to collect two separate sets of similar geotagged, identified, species image data.
joemudge
commented
10 years ago I contacted the makers of the WeBird project at the University of Wisconsin and they pointed me toward www.arisgames.org , which is a free, open-source platform for creating/authoring/testing/editing/and playing location-based games for iOS devices, with a capability of incorporating image data collection as a component of the game. I'm working my way through the training documents for this platform right now and I'm hoping that it will facilitate the creation of the data collection side of the game.
joemudge
commented
10 years ago I've also been working on the actual game aspect of the project, which I'm thinking would be good to also ground in ecological theory to give it another dimension of educational content. One idea I had was to allow players to use their identified species image collections to compete against each other (or against an AI) in a strategic, habitat colonization simulation game. There's a theory in plant evolutionary ecology called C-S-R triangle theory. It basically states that each plant species has evolved a survival strategy based on a unique tradeoff between (c)ompetitive ability against other species, (s)tressful environment tolerance, and (r)eproductive ability. There are lists of what species characteristics are associated with each of the C-S-R axes, and lists of C-S-R classifications for many plant species. There are also JAVA-based cellular automaton simulation models that simulate interactions of species with different C-S-R strategies on a grid over a number of timesteps. What I was thinking would be a fun game would be to provide the players with a simulated habitat grid, and allow players to take turns deploying individuals of different species (with different C-S-R strategies) from their collection of identified images to the grid. Once each player has deployed a set # of initial individuals from their collection to the grid, a simulation is run, showing the colonization of each player's species to empty parts of the grid and competition between species at occupied parts of the grid. The player who wins is the player whose placed individuals and their offspring control a greater proportion of the grid after a set # of timesteps. Players would want to place their individuals to limit competition between species from their own collection and to outcolonize, outcompete, and/or outlast the individuals of different species from the other player's collection. I just came up with this game idea over the last few days, and I'm very open to changes or other ideas. Also, although the C-S-R theory was initially developed for plants, the general idea still holds for other kinds of organisms and the theory has been expanded upon, into something called UAST (Universal Adaptive Stategy Theory), so I'm sure it'd be possible to to make a game like this work for other groups of species too, or even multiple groups together, with preditor-prey interactions as well as competition. I imagine we'd probably want to start simple though. I've got a .pdf of an ecological modelling paper that contains JAVA code of a CSR cellular automaton simulation model that might provide a good start for the gaming part of things if we decide to go with this idea. ( dx.doi.org/10.1016/j.ecolmodel.2006.12.039 )
joemudge
commented
10 years ago I'm not a programmer, but I'm hoping that between: 1) the free, open-source platform for creating data collection oriented location-based mobile games (www.arisgames.org), 2) the Github code for creating a smartphone based species identification tool ( https://github.com/UWMLI/FieldResearchTool ), and 3) the JAVA code for a cellular automaton simulation model of species competition, ( dx.doi.org/10.1016/j.ecolmodel.2006.12.039 ) that we have the initial components of most of the big pieces required to make this ecological data collection game a reality.
PIRatE-Lab
commented
10 years ago Guys,
This sounds really fun. I have been thinking of a game that explores ecological fragmentation, focusing on road kill. We have about 9 years worth of data on the road kill rates here in southern California (Centered in Ventura County, but extending into Los Angeles and Santa Barbara County). We survey about 50 roads and have documented more than 5,000 kills of everything from badgers to bears to weasels. Separate from our research (supported with an iPhone app called Splatter Spotter in need of updating), I have been hoping to get more into the public outreach by creating a game that gives the player the options to managed road differently (install wildlife crossings, slow speeds on roads, etc.). My thought would be for folks to pick the time of the year and or use the most recent data so that the kill rates, and other underlying data change and are not static. It would be great to blend the two (citizen science plus ecological, interactive game).
You can see a quick five minute summary of the road kill project here:
Or a longer, but more recent one here:
Organizational Page: GameEco
Goal:
To develop a video game that also features of outdoor activity, environmental education, and scientific data collection.
General idea:
(For those familiar with Nintendo's Pokémon franchise, the idea could be considered loosely analogous to a real world version of the Pokémon video game.)
The premise of the game would involve players ‘capturing’ geotagged and timestamped digital images of real species in their actual habitats on their smartphones. ‘Captured’ individual pictures of species would be added to their personal collection, and also uploaded to a central database generating geotagged and time-stamped data on species occurrences, abundances and ranges.
Incentive for ‘capturing’ more individuals of different species could be achieved through some mechanism to allow for players to compete with their personal collection of 'captured' individuals of different species against the collections of other players.
The game would require collective identification of unknown species images through comparison to reference images of expert-identified species, possibly through some in-game reward mechanism whereby points for successful identification of random unknown images could be used to acquire improvements to a user’s personal collection.
Proof of concept:
Each separate component of this game has already been successfully implemented elsewhere:
Benefits:
The success of a game of this type could generate unprecedented data on species abundances and distributions, and enable; (1) better ecological monitoring, (2) tests of novel and/or previously untestable ecological hypotheses and (3) better predictions about future changes in biological communities.
It would also increase the ecological literacy of game users, which is important because ecological illiteracy has been identified as a main impediment to societal prevention of environmental degradation.
In addition, it would merge video gaming with outdoor activity, which many American children do not get enough of.
My background:
My research background is in community ecology. I lack the experience in game development that would be required to make the game entertaining and I also lack the programming skills required to build the game.