Old stuff

Images scenarios

• Favouriting the image > saving to a user-defined collection
• Editing image metadata
• Geolocating the image
• Choosing as a header image
• Removing a falsely associated image from the Images section of a topic, an act that in reality adds more data to image data object, such as the correct related topic. This data could be further user to disambiguate topics.

Foreseeable future interactions:

• Identifying people, places, and date

See also http://wikidocumentaries.wmflabs.org/wiki/Images

What is saved of the image with this first click?

1. Only the necessary pointers to the original image
   1. Image url
   2. It may be wise to save also the image hash to be able to compare with other images on the web.
2. or all metadata that the project can map?

When the user enriches data by associating textual data with Wikidata/Wikidocumentaries entities, how is additional data saved?

1. Raw and refined values are saved in the same property, one as a qualifier. If the textual comes first, then it must be that as the value, and the linked value as a qualifier.
2. Only when the user links a value, values are recorded. This would allow the same structure to be vice-versa, the linked value as the main value, and the original textual value as a qualifier.
3. Raw and enriched values are saved in different properties (I cannot imagine why)

• • Scenarios of uploading to Wikimedia Commons**

The minimum file info is saved immediately into Wikidocumentaries and uploaded to Wikimedia Commons when the data becomes complete enough

• The file can be automatically saved to Wikimedia Commons, if it meets a certain minimum criteria, for example existing title, description, copyright notice, attribution, year or the lack of it.

The file is immediately saved into Wikimedia Commons at any user interaction

• Commons can accept partially structured data
• It will however have notability policies that are more strict than what Wikidocumentaries allows
• The original image will remain at the original source if the image is removed from Wikimedia Commons.
• The original image can be backed up in Internet Archive to prevent it from disappearing if removed from Commons or the source.

• Synchoronising sources*
• How to deal with varying data of the same image, evolving metadata on any of the platforms, and serving the enhanced data further.

Upload stuff

Where

Vue function that can be attached to different parts of the UI as part of another action or in itself

• Image actions (image actions menu, image section mass actions, image viewer)
  • Save image to a list
  • Promote image to header picture
• Blacklist an image / disassociate with a topic (image actions menu, image section mass actions)
• Use a tool such as geotagging (image actions menu, image viewer)
• Modify any parameter in the metadata (image viewer metadata editing section)

What

• Invoke OAuth
• Check image hash for duplicates
• Upload image file
• Write original string values to SDC properties that accept string values. Note on source language.
• If there are automatically mapped item-values for properties, upload those to respective properties in SDC. (*
• Properties in SDC that expect items cannot be imported. They will display as not resolved (red) in the metadata viewer. They need user interaction to be resolved and updated.
• The linked Wikimedia user will be noted as the uploader. Generally, the originating GLAM, the distribution platform and the image license can be recorded.

• ) Every user interaction could contribute to a match database on Wikidocumentaries. This data should include the combination of
• the string value
• the Wikidata/Wikidocumentaries property
• the linked Wikidata/Wikidocumentaries item
• enough context, TBD: The originating institution (for example, a certain GLAM uses a character combination to define a certain location. This string is connected with a Wikidata item, but it is only valid in the metadata of this GLAM or their collection.)