D3 and imagemagick-based dynamic image display.
Coverspace came out of conversations between Peter Leonard, Lauren Tilton, and Ben Schmidt surrounding the THATCamp New England session on "Image analytics"/"Screwing around with non-text."
Example browsers (feel free to add one)
Creating an instance
- Clone this repository (into your webserver directory, if you have one)
- Create an image directory called "images" and fill it with jpgs. (Other imagetypes are trivial to support)
- Run "make". (Or just execute the one python line in the makefile)
- Open up index.html in your browser. (You have to do with a running webserver, unfortunately--the flat file won't load
- Open the page in a web browser, and click around. (For the example: go to http://localhost:8018)
Creating an instance without a webserver on a Mac: When in the directory in terminal, just type:
python -m SimpleHTTPServer 8018 & open http://localhost:8018
- python w/ wand library
Currently, you CAN sort on the filename: so if you make your filenames into something useful (eg datestamps) you can arrange by date.
- Allow user-generated metadata to be keyed against the filenames (so we can sort by year, for instance.)
- Zoom in on a particular region (probably by actually expanding the display area of the page, though there are other options.)
- Incorporate filtering based on metadata.
- Add entropy to the color distribution elements.
- Think about how subdivisions of the images might be useful for something or other.
- Allow underscores in filenames.