As the tooling around STAM is reaching more maturity, it would be a relatively small step now to develop a higher-level WebAPI that exposes some of the functionality via a webservice. Communication would largely proceed by GET/POST of a STAMQL query.
The webservice (let's call it stamd), would load one or more annotation stores into memory and serve requests. It would be somewhat comparable in function to both TextRepo & AnnoRepo. Do to the nature of the current library, however, the implementation would be strictly memory-bound, which might make it less scalable to huge data collections.
The service would be kept pretty lightweight without features such as user authentication (that can be taken care of in a wrapper).
As the tooling around STAM is reaching more maturity, it would be a relatively small step now to develop a higher-level WebAPI that exposes some of the functionality via a webservice. Communication would largely proceed by GET/POST of a STAMQL query.
The webservice (let's call it stamd), would load one or more annotation stores into memory and serve requests. It would be somewhat comparable in function to both TextRepo & AnnoRepo. Do to the nature of the current library, however, the implementation would be strictly memory-bound, which might make it less scalable to huge data collections.
The service would be kept pretty lightweight without features such as user authentication (that can be taken care of in a wrapper).