²

Okay so the proposal was last week: We have established a data collection model with a process, we just need to create a website and decide on an output now.

The model goes something like this
Data Collection - Data Storage - Data Processing - Output

Data Collection
The plan here is to create a website that collects the information anonymously from a group of people without us interfering. Our smart idea here is to create a website that will prove useful and therefore gain popularity (we hope) so that the website reaches as many people as possible. The websites purpose is to ask the user a series of questions in which the answers are selected from pictures, the website then takes those answers and recommends something to the user. At the moment the recommendations are along the lines of movies, websites, artists and the like. The reason for answering questions by selecting pictures is twofold (and is borrowed from youniverse.com). One, it is really much more interesting than selecting from a series of sentences, because as the adage goes: a picture is worth a thousand words. Two: The selected answers will be more subjective for this reason, there is more scope. The method of recommendation at this stage is (A secret). In this way we hope to provide the user with something they will enjoy but would not otherwise have tried or found by themselves.

Data Storage
At this stage storage will probably be in the form of a MySQL database that is acessed from the flash front end. The hope here is that we can create a database that we can pull useful information from towards the end of the project and use in a different way.

Data Processing
The data processing will either be written in Processing or Flash, and will sort through the (anonymous) collected data to see if we can derive some interesting results that we can present separately from the raw information.

Output
At the moment the main output is the one the website provides the user with: It takes the user inputed data from the quiz, and makes an intelligent (well, web 2.0) guess at what the user would enjoy. Some form of 'this was useful' feedback could be used to alter the image tags over time to something approaching a perfect description. A second layer of output is the entirety of the project, that is to say the data collected from thousands of users and what they clicked on/information about them. For instance we could list the answers to various questions by age/sex/country etc.