...
Bl: variations on a theme that keeps arising that do have architectural implications. Goal is to triplify as quickly as possible, dump them into a second behind the scenes VIVO that acts as a temporary hopper (?)....transform raw data into format that is usable by VIVO, including disambiguation some projects are rule based, some have custom queries, lots of processes involved in transforming, migrating, publishing. In some cases this can be straightforward, other situations are more complex. Big graph of triples up front and then work with it to add to VIVO. Incremental updates are trickier rather than redoing everything. So other approaches that are incremental are worth exploring.
...
De: entities go into their categories: profs into their named graphs etc. running into problems with spaql sparql translator where it just hangs. When the translator tries to match new trifplified triplified data with existing VIVO data it just hangs. Do not want to have to get deep into Jena in order to do this, would like to find a different way, remove dependencies on Jena. would like to avoid reinventing wheels.
Aw: hopefully the task force can find a way to do this once together, determine a general work flow workflow that works for most.
Bl: need common tools for freely available sources to help people get their VIVOs up and running. Other question is about the architecture of application itself, how dependent do we want to be on central data store and force people to grapple with this model up front. VIVO exposes all graphs so all triples need to be nicely lined up, this can be difficult to accomplish. If this can be simplified or improved to allow layers to be built up over time, so as to not force things to be need to be set up perfectly at the outset.
...
RoF: eventually yes, at this stage still simply trying to quickly get a feel for what is out there. Did not want a big formatlized list formalized list in case I left something out.
...