Created a simple prototype that connects google and amazon services. The prototype calls the google webservice with an inital search term, chooses randomly one of the returned results and feeds the title of the chosen result into amazon services. Here again a random result element is taken and used as the base for the next google search. We have been able to create chains of 50 elements length.
Comments (0)
Abstracting the wsdl generated google and amazon services by using shallow bean reflection. A bean is represented via a hashmap that contains all its members, as literals, nested hashmaps or nested lists. That should simplify and generalize data access. The members can accessed via a path like mechanism, example:
MapUtils.getComplex( mapInstance , "contactInfo/physicalAddress/city" );
This gets the nested map with the key “contactinfo”, within that map the map with key “physicalAddress” and within that map the literal “city”.
Comments (0)
Extended the prototype to output an xml file and created some basic html representations. The connector from google to amazon is the “pagetitle” and the connector from amazon to google is “productName”.
- Google-Amazon Chain (10 tuples) with the initial term: “George W. Bush”:
- Google-Amazon Chain (10 tuples) with the initial term: “Daniel Yergin”:
- Google-Amazon Chain (15 tuples) with the initial term: “Chinatown Soccer”:
Comments (0)
Extended the prototype to support deeper reflection, now every particle of the webservice response is available in the xml result for processing and to act as connector to the next element of the chain. The updated prototype engine uses either amazon product name, an actor/director name, manufacturer or author to connect to a google request. The connector is highlighted. The next version will improve the amazon queries to create more fittingresponses and use new services providers to connect to. Added semacodes to visualize URLs.
- Google-Amazon Chain (10 tuples) with the initial term: “Heinz Foerster”:
- Google-Amazon Chain (10 tuples) with the initial term: “Persepolis”:
- Google-Amazon Chain (10 tuples) with the initial term: “Trial And Error”:
- Google-Amazon Chain (15 tuples) with the initial term: “Buckminster Fuller”:
Comments (0)
This prototype is using website registrant data, weather reports, satelite pictures, stockquotes and local time in a given area as well as google, amazon and alexa.
- Seed : “Arthur Schnitzler”:
- Seed : “Villa Straylight”:
Comments (0)
This prototype is using website registrant data, weather reports, satelite pictures, stockquotes and local time in a given area as well as google, amazon, alexa and flickr(NEW).
Comments (0)
Radioactive by Aaron Zinman and Judith Donath
“RadioActive is a forum for large-scale mobile-based asynchronous audio discussions. It allows participants to engage in persistent discussions on a number of different topics, similar to usenet. Users may moderate posts, allowing the dynamics of the chat space to change over time. Navigation through the threads is made possible by both audio and visual interfaces that are optimized for the device. In addition to arbitrary topics, users may also engage in discussion surrounding time-shifted media clips. This is a reaction to the phenomena of TiVO and podcasting, all which allow users to access their media in a personalized fashion but completely passively. RadioActive gives audiences a chance to talk back, contrasting the typical unidirectional model of broadcasting. “
Comments (0)
SearchScape is an amazing shockwave based 3d information topography of Manhattan. Some similar approach could be taken to visualize a chain of konnexion datasources. Processing or Shockwave are potential technology platforms to make that happen.
From the site http://www.searchscapes.net/:
Each person constructs his/her image of the city. This image is made out of facts, memories, experiences, stories, news - mostly invisible data, and not only of architecture, buildings and streets. “SEARCHSCAPES: MANHATTAN” is an attempt to create a tridimensional map of Manhattan, using existing data from the web.The objective is to compare the city’s “physical spaces” and “information spaces” (search results). This is an attempt to materialize information: to give it dimension and physicality.
Comments (0)
Created the first image layering prototpye with CSS alpha filters. All images used in the konnexion output are stacked on top of each other with different opacity. Every image used is resized to 800×600. Amount of images in this sample is 68 and seed term “Commodore Vanderbilt”. I got the CSS idea from the Yahoo! API Image Collage. Click to enlarge the screenshot below.
The CSS instruction to create filter elements with a certain opacity in Firefox, Internet Exlporer and Safari is:
display: block; position: absolute; left: 20; top:20; width:800; height:600; -moz-opacity: .40; filter: alpha(opacity=40); opacity: 0.4;
Comments (0)
