Commons:Structured data/Archive/2014/Issues without structured data

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search
This page is a work in progress page, not an article or policy, and may be incomplete and/or unreliable.
Please offer suggestions on the talk page.

Wikimedia Commons has issues and limitations that could potentially be helped by providing structured data with Wikibase. This list of issues and limitations is not necessarily comprehensive and is in no particular order of rank; feel free to suggest additions that might benefit from structured data.

InternationalisatioN and LocalisatioN[edit]

InternationalisatioN is changing the structure to make LocalisatioN easier.
LocalisatioN is translation of the information into all the languages.

  • Without structured data each file description must be translated independently.
  • Even where descriptions are similar it is difficult to reuse the translations.
  • Even where translations exist it is not practical to show the correct language to each reader and hide other languages.

Upload[edit]

  • English is required - contributing media is difficult to impossible if you do not speak English
  • Free fields in the upload forms leads to inconsistent data
  • Template parameters are not localized, making data inaccessible to non-English speakers
  • Current upload workflows provide limitless overlap of data

Edit[edit]

  • Editing data is a different user experience than upload, with an inconsistency in what data is editable
  • The file page has a confusing mental model - it is not clear what content is editable (i.e. wikitext and templates)
  • English is required for editing parameters and template content
  • Overlapping/redundant templates - data is often repeated and other parameters get dropped and it is not clear where/when, causing a maintenance burden
  • Not clear what, if any, template parameters are standard or not
  • No validation of data
  • Difficult to track derivative works when editing shared data

Search[edit]

  • It is currently impossible to filter or sort search results. In theory, with structured data this could be done (by author/license/resolution/media type/date/geolocation etc...)
  • Cannot search based on assessment criteria to find the best example of media being searched for
  • English-centric search results - For example, compare a search for the English word apple against the German word Apfel
  • Search does not disambiguate concepts, such as searching for "orange" does not differentiate between orange the fruit and orange the color
  • Search does not return related files based on concepts
  • No meta data in search results

View[edit]

  • Inconsistency in viewing...well, everything about data from file to file. This is impractical for human and machines alike.
  • Mobile experience in the File: namespace is inconsistent, lack of uniformity or just plain broken views
  • Design is hard-coded into templates and file pages, limiting data to what is hard-coded
  • No clear distinction between in-file data and other data in the description
  • As with editing, linking to derivative works for viewing is sub-optimal
  • Any alternatives you see to English when viewing a file page are achieved by hacks (like "init.")
  • No clear minimum requirements for data about a file
  • No clear access to related files aside from categorization

Re-use[edit]

  • Ambiguity in license requirements for multiple licenses
  • Ambiguity in attribution requirements for authors
  • Unclear usage restrictions
  • No clear way to indicate other used sources in reuse mix
  • Sub-optimal APIs
  • English is required for a substantial part of reading about re-use