Today it’s usually easier to use a search engine for information than find it locally, which is not optimal from a digital sovereignty point of view. Part of the problem is that we lack good open source tools to provide context and graphical search of local documents. These tools present plain-text lists for search results, which means people with good graphical memory find information slower. We think it’s a huge opportunity to show the context of search hits in a graphical form to find information faster. Technically, this will mean taking an existing file synchronization and sharing (FSS) solution, hosting your documents on-site. Then improving LibreOffice to index content in documents with their context. We will build a secure REST API on top of this in Collabora Online which provides good performance. Finally we will integrate with a search engine, e.g. Apache Solr to create a proof-of-concept search page that allows searching in all documents hosted in a FSS solution. This will serve as an example how to integrate our solution to other projects like Nextcloud.
