Warning:
This wiki has been archived and is now read-only.
Use Cases & Requirements/Media Linking and Bookmarking UC
Contents
Media Linking and Bookmarking UC
Participants
| role | name |
|---|---|
| lead | Michael |
| supporter | ??? |
Abstract
Having a global identifier system for addressing media fragments allows to create links to these fragments, and bookmark them (potentially for making a playlist). It also allows embedding of just the relevant segment into another Web page (e.g. blog).
Background
Users - Humans vs. Machines
On the Web we typically find two distinct types of agents: human users on the one hand and machines (software agents, etc.) on the other hand. Both agents come along with potentially different requirements:
| issue | human users | machines |
|---|---|---|
| content | interaction, accessibility, design, usability, etc. very important | encoding, access (HTTP, etc.), etc. of primary interest |
| metadata (schema + instance level) | not directly of interest (beside tags, filter) | essential in order to discovery, access, query, etc. |
| addressing | implicit (search engine result, etc.) | essential |
| common/standardized formats | not relevant (what counts is the 'output') | essential (need e.g. a common data model such as RDF in order to exchange, etc.) |
Web Of Data
While the Web of Documents (largely the Web as we know it today) primary focuses on the human user, the Web of Data supports both humans and machines in their respective tasks. For the human aspect see Tom Heath's recent IEEE IC article How Will We Interact with the Web of Data?; the machine point-of-view will be discussed in the following.
- universal, accessible identifiers (URIs) + data model (RDF) + vocabularies (RDFS, OWL, etc.)
- linked data principles
- global descriptions vs. fine-grained descriptions
Scenarios
In the following, three scenarios within this UC are described in detail. The first two scenarios address human users, the last machines.
Bookmarking
Tim is a Last.fm user. He wants his friend Sue to listen to a cool song, Gypsy Davy. However, not really the entire song is worth it, Tim thinks. He wants Sue to listen to the last 10 seconds.
Playlist
Mash-up of tracks from different sources, etc.
Interlinking Multimedia
Enable the fine-grained, linked-data-compliant addressing of multimedia assets in order to use it in the Web of Data.
Discussion
Bookmarking. TBD
Playlists. Having media fragment URIs further enables media (video) playlists to be created by putting together a list of links to media fragments, as has been done in Metavid Sequence Editor (being discussed at about 3:37min).
This is a very simple form of mashing-up without destroying the original content - a usage that is not possible without fragment URIs. Non-destructive mash-ups are important from a copyright and DRM POV.
The fragment playlist can even put together fragment URIs from different servers. It can be shared in an RSS feed or through a XSPF playlist specification.
If we were to allow addressing of the different media tracks in videos separately in addition to addressing just temporal fragments, we could even have a mash-up where the video and the audio comes from different servers. This may be beyond our mandate.
Interlinking Multimedia. Based on interlinking multimedia ...