We started integrating a workflow engine into OSA-application last week. The workflow engine is developed as bachelor’s thesis by Heikki Kurhinen. User interfaces for the batch ingest has been done already for the beta version, but the functionality has been missing.
We are going to develop several micro services for OSA-specific workflows. These workflows will generate metadata automatically, before we ingest documents into our archiving system. Automatically generated metadata during workflows execution is stored temporarily into Mongo database. The aim is to bring automation to the ingest process. Users can ingest several files at the same time and the content will be handled
at the same way during the ingest process. Regardless of ingest style (manually or batch ingest using the workflow engine) the last step of our ingest process is still XML schema validation, that is already available in the current version. We have determined schemas for all data types to guarantee the minimum requirements of the Capture-datastream. Schemas are stored in the content models in Fedora Commons Repository.
At the same time of developing new features into OSA-application, we get the users feedback of the current OSA-application version running in https://osa.mamk.fi.
Of course we are testing OSA-application ourselves, but we can not reach the real-life use cases by testing newly created functionalities mainly. Feedback from project partners is very essential for keeping development roadmap and users real requirements in balance.