Moving this to a NOISE search tool
Currently working on these ideas with Josh Russell, who worked on AstroJam earlier in the year.
Based on the Color Pickr concept users can search work by colour. In addition this can be filtered down to categories eg graphic designers, age, postcode, rating, keyword. Basically any meta data we have on the database.
An intilal random selection of work is shown to start with, you can filter this down by selecting a colour or search term. This should give you an abstract map (ala Dopplr) of work.
This is still in it's vaporware stage but Josh is working on the code to breakdown the colours. The tricky thing from my point of view is incorporating this into a standard search format on the NOISE site. In an ideal world this would be the only way to search and find work but at the back of my mind is the nagging knowledge that I'm going to have a regular search option somewhere. Actually I think whats holding me back with this is the scale of it. In the showcase section it's to small, there need to be two variations. So a showcase search section and a larger scale section for the whole site, as shown below.
Edit:13/7/2008
Edit:15/6/2008
Visual for NOISE catalogue/search tool which incorporates a number of similar elements
Edit:15/6/2008
Edit 18/5/08: Following on from work in Manchester at NOISE, I've been looking at a lot of sites that use meta data to create visual websites. For example oSkope
It's the use of meta data that helps generate the serach results. So this comes back to the upload and site registration tool. In an ideal world what information do I need to collect store in order to buld something like this? You need information aboout work, type format, genre, style, date when work was uploaded and any tags. You also need information about the person who has uplaoded the work. What department are they from, what year are they in, what groups do the belong to.
OK so with the CAGD site at the moment you have:
Contemporary Art Practices, Fine Art, Graphic Arts and Design
BA Level 1, Level 2, Level 3, MA Level 1, Level 2 and staff
Groups
Comments. On your work and others work
Sets
Tags but these are only added after you have submitted. This would have to be done at the upload stage. See how to tag proposal as well.
The unviversity doesn't divide into seperate departments, so the upload and registration process become key in order to to capture and curate information. After that has been done then you can select how to view it.
So what else is missing?
One of my favourite parts of oSkope is it's graph function, so can you alter how work is viewed. You can also view works as grids, stacks and piles. Could I start to use the information above to do somthing similar?
Edit 18/5/08
Older:
Looking at ways of people searching and finding work on the MA space. I want a sort of anti "Amazon reccomends", so you find unexpected work but has some relevance to work you've done or looked at in the past. Not just a random selection.
For example on Amazon reccomends if you skipped the first few links of what they reccomend to you and are given the a list of things five links down the chain. Mapping what users are looking for.
Amazon.com Statistically Improbable Phrases Can this be aplied to people searching for images?
Amazon.com's Statistically Improbable Phrases, or "SIPs", are the most distinctive phrases in the text of books in the Search Inside!™ program. To identify SIPs, our computers scan the text of all books in the Search Inside! program. If they find a phrase that occurs a large number of times in a particular book relative to all Search Inside! books, that phrase is a SIP in that book.
How do you start to visualize this, Musicovery has a great interface, that would be really suited to visualizing and mapping how people are searching and viewing work.
Maybe a graph
Print this post
No comments:
Post a Comment