Imagewiki iPhone Application
Project Summary
Keywords: Locative, Visual Studies, Visual Art, Social Cartography, Ubiquity, Hyperlinks, Device and Experience Design.
Medium: C, Objective C, Ruby on Rails, Postgres, Unix, Coda, and HTML/CSS, iphone
Context: Interactive art, sociable media, design
Concept, Design and UI: Paige Saez
Engineer: Anselm Hook, John Wiseman
Overview
The imagewiki is a visual search tool for mobile devices. It allows for the ability to turn images into physical hyperlinks, conflating visual culture with a community-editable universal namespace for images.
Using the imagewiki a person takes a picture of something with their camera phone and then adds a description for the image. For example, I take a photograph of graffiti at a bus stop; I add the name of the artist and where he/she is from and list where else you can find work by that artist. When other people perform a visual search using the imagewiki tool they discover the description I added to the image. They can edit my description and add content and links to the image. This collaborative editing of the images meaning creates a dynamic, evolutionary process of understanding, interpreting and participating in the direct affect of visual culture in situ with the images themselves.
The study of our visual culture acknowledges the reality of living in a world of cross-mediation--our experience of culturally meaningful visual content appears in multiple forms, and visual content and codes migrate from one form to another. Through the use of the imagewiki tool and the experience of presenting the project as an experiment in collaborative ownership.
With this work we hope to create a place of dialogue around the question the value/ownership of images, intellectual property rights, and the creative commons.
Politics of Representation
Our culture is complicated by layers of visual and image based representations of meaning. Images accrue value through their culturally assigned meaning and this meaning has value. Corporations often create brand associations between images and their products. Photographers and artists often protect an ownership over their images. Appropriation, mis-representation and theft of imagery are common. Many images/icons have meaning for cultures, communities and religion. Images are both taboo and totemic.
We therefore believe that there will be significant social, political and even legal challenges to the idea of associating meta-data with images. This is one of the key areas we want to explore.
Our progress to date
1. Through our experience with the visual artists while working in technology we have discerned a need for a commonwealth approach to the medium. We have gathered a small team of software developers and image recognition experts to create an altruistic open database for image meta-data.
2. To enable a community driven image recognition service we have provided a website and Application Programming Interface (API) backed by the Scale Invariant Feature Algorithm (SIFT). This program also supports email and mobile versions of the service. All three of these interfaces- across web, email and mobile allow for contributors to add images from anywhere; desktop or treetop. The work is open source, open data and open participation.
3. We are now at the stage where we want to improve the speed, and polish the experience - we are happy with having proven the concept but would now like to actually use it to change the world.
Questions remaining
1. Search providers great and small are gearing up rapidly to service this emerging potential for visual search as a new market. But who gets to dictate what information is returned on a visual search of a Coke can? How about a visual search of a child's' drawing of a Coke can?
2. How do we crowd-source trustworthy statements about the things we see? What are the Patent issues of various image recognition approaches such as SIFT? What are the implications for Trademark owners such as stock photographers and Flickr enthusiasts?
3. Will there need to be a Creative Commons license to cover this? If we think of images as a new kind of hyperlink then does visual search then become a valid "resolving opportunity" like DNS? Will the Image DNS be owned by Network Solutions or the people? Copyright has not even begun to touch this space.
Program Goals
The hope is for guidance, time, support and funding for "imagewiki", an open-source project that we believe is unique and very important to the world.
Because it deals with the relationships between the creation and ownership of images, the introduction of physical hyperlinks in the consumer market and invasive computing, it exists at a highly political and volatile intersection of copyright law, creative commons licensing, intellectual property rights and trade-mark licensing. Negotiating this space, interpreting and helping define this space is the goal of the project.
There may be opportunities for monetization of imagewiki without compromising the core values of being committed to open source, open data and open participation - expertise in that landscape would also be welcome.
Project Information Document
Look and Feel. The project mimics a web 2.0 based service where users can use their mobile phones to take a picture of an object in the real world such as a piece of graffiti, a band poster, a painting, a store, a book and get back information on other peoples comments on that subject. The interaction design and usability tests that we have done have led us to a conclusion that the simplest possible interface is best - so the UI is quite minimal - a picture is taken and salient information is returned. Users are also given an opportunity to add their own comments to subject matter. [ Also see addendum below ]
Proposed Budget. The application is complete, it works! But it requires polish, promotion and support for new phones such as Android and the newer versions of the IPhone. We estimate 4 months for two people full time at about $5000.00 a month Canadian plus $10,000 for additional hardware and travel costs for a total of $50,000. We have other contributors but they volunteer their time for trade in kind and for early access to the work.
Prototype Description. We are fundamentally makers who strongly believe in "Praxis makes Perfect" and we don't emphasize on paper design or descriptions. We actually BUILD working examples and prototypes and use that as a basis for dialogue and revisions - using an agile development philosophy of small milestones. The nature of digital media, the ease of construction, and our own skills, make it this the best route. We have a working demonstration of both the http://imagewiki.org website and a downloadable IPhone client that you can install on a cracked IPhone. [ Please see addendum below for more details ].
Distribution Strategy. For free. Creative Commons Licensed. On the Web. Viral. Building a community around the core values. Engage local communities and find ways the tool is applicable to their needs- we have many use cases on this and we have done a lot of research into how small non-corporate communities can benefit from the tool.
Materials List
We have our own infrastructure including 1/2 Rack with a 4u database server and 4 web-servers with fall-over redundancy and a firewall/router with fall-over redundancy at a fixed cost racked in Portland, Oregon - on the USA fiber backbone with low latency and good connectivity to Asia and Canada as well.
We all personally use Macbook Pro's and have all the hardware we need in that area.
We use a wide variety of open source tools and technology including OpenBSD, SIFT image recognition software, PostgreSQL database, Ruby on Rails and a wide variety of other technologies that we grab as needed. Since we are strictly committed to open source - we have no costs here.