Feedback/suggestions from impresso review at Dagstuhl

NEWSPAPERS

[ ] Newspapers page: Unclear: which parts are missing within the span of a newspaper
[ ] No legend: how was this calculated, what are we missing? For @mduering to add to FAQ entries
[ ] Suggestion: export newspaper metadata for additional analysis outside of the platform. Feasible @danieleguido ?
[ ] Missing: Which titles have article segmentation? Feasible @danieleguido ?
[ ] Unclear: how are newspapers sorted? Is this random?
[ ] Suggestion: Width of newspaper lines could reflect the amount of content/year
[ ] Request: compile a list of newspapers based on their metadata: for example short duration
[ ] Suggestion: add date filter and location filter to the newspaper page (to select newspapers from a given time and space)

[ ] Order of newspaper titles unclear- why not alphabetically or the same order as in /newspapers?
[ ] Entity suggestions: why is there no “more” button?
[ ] Unclear: how do I do an “OR” search?
[ ] Suggestion: Project back distribution of search results on the temporal view in /newspapers: article frequency, languages etc.
[ ] Wildcards, regex are requested
[ ] Suggestion Search: give it more screen real estate, make it a separate process: I compile all parameters, then press “Search” rather than displaying results right away

FAQ

[ ] OCR QA is missing. to do @mduering
[ ] “Relevance” in search results not explained
[ ] Idea: add links to public domain images to csv export
[ ] FAQ - entry on export missing to do @mduering , mention that BNL data not part of the export
[ ] FAQ entries: NGrams needs updating, to do @mduering
[ ] Make FAQs searchable
[ ] Missing: information about system architecture and processing pipeline, to do @mduering

NGRAMS

[ ] Missing: Warning that the corpus is too small and/or display of current corpus size based on which ngrams are visualised. (how to do?)

impresso / impresso-frontend