OERInterop-ContentAccess

<- Back to OER Interoperability Principles

Provide access to editable versions of OERs
 * If you distribute non-editable versions of OERs, also provide access to the editable source formats.
 * Use the most open license you can
 * Use the most open file format you can
 * Use the most semantic file format you can
 * Provide transcripts and other accessibility options for rich media
 * Enable user generated content

-

Recommendations:

1. A standard API for converting objects from one format to another (transcoder in the sky). Tools would ideally be local but could be external. For instance user finds a wav file. User needs mp3 for their use. Clicks the "convert me!" button and another asset is created and downloaded to the user. Alternatively the converted file format could be stored on the server and available for other users. Best practice describes a continuum from low bandwidth capability (conversion and storage of modified files is not done on the site to everything being done within the site).


 * open office libraries
 * ffmpeg
 * imagemagik
 * gimp
 * audacity
 * send to wiki

2. machine to machine - provide simple api to get xml dump of any oer item (sites sharing with other sites)
 * metadata
 * dublin core
 * oai-pmh
 * simplification of existing spec (highest payback) - go down to the level of pedagogical materials
 * Harmony project to unify DC, LOM, and mp3 metadata
 * www.dls.ucar.edu/people/kginger/nsdl_dc/index.html
 * the content itself

3. machine to person - metadata - issues of grain size. Metadata goes inside and outside the object. To provide capability for semantic linking between versions. Versions need to be addressable (and/or url addressable).


 * Scrumpdiddlyumcious
 * annorate
 * trailfire

4. Fair use (?) - multiple licenses for a document, particular license for a component

structure attribution

-

Business model for the transcoding services