Linked Open Data Harvesting Node

[jvanulde]

There are 3 purposes for the harvester/node:

1. Publishing features available from the node.
   • As representations?
   • Use case = simplify browser indexing of data
   • Use-case = enable data providers to join the network without implementing URL resolver
2. Publishing cross-node relations available from the node.
   • As representations
   • Use case = nodes joining network implementing resolver
3. Publishing node connections.
   • As links
   • Use case = same as (2)

LOD Harvesting Node Prototype

[jvanulde]

@heryk @denevers can we add features to the LOD Harvesting Node Prototype to demonstrate use case 1?

[denevers]

Add feature as "find features to put in there" or "add features you already have encoded and load them in" ?

[jvanulde]

Probably best to use features we already have.

[denevers]

All we have is in the catalog

[jvanulde]

Are the feature resources accessed via '/data/' path? The features need to be referenced such that we don't rely on a resolver.

[denevers]

most. but since /data/ redirect (303) to another link, it implies that are available directly under that other link. I realise I don't understand completely what you are looking for,

[jvanulde]

So in use case 1, links to node features need to be rendered in the Node landing page. This way, network participants that don't have a resolver can still participate. So perhaps we just provide direct links to WFS GetFeature requests.