Depending on the application client need, a same Stanbol instance could provide different processing chain.
At least, two types of processing chain could be useful :
- The first one - fast - with just one or two "essentials" engines, provide entities to the users in (near) "real time"
- The second one - slower -with a lot of engines, provided a detailed analysis, and computations that help for classification, summary, detailed enhancements, etc... and are displayed to the user later or when reopen document.
Another types of chains can be useful depending on the type or the source of document (e.g. engine A is not useful for mail and picture, but relevant for scientific paper)
Depending on the chain configuration, the endpoint could be sync or Async (see STANBOL-263), in a push or pull mode.