The Future of Asset Management (Neos Sprint November 2021)

This post should server as condensation point for the things discussed today in Slack.

Topics we talked about

  • Asset organization in Neos and other systems
  • Permissions (in Flow code, for editors and (future) API/GraphQL use) and how they possibly relate to asset management
  • What asset sources should be able to do
  • Read-only and writavle asset sources and their right to exist
  • Allowing arbitary metadata on assets
  • Translation/localization of assets and metadata
  • Using the CR to store asset metadata
  • Introduce “tag sources” that can be selected for asset sources to mix’n’match
  • Allow asset sources to expose tags/folders/… from their side
  • Asset sources can be a “simple source” (import from unsplash, enrich in Neos, never look back) or a “complex source” (manage assets externally, “use” rather than import, keep in sync with external source), with the distinction being somewhat fluid
  • exposing metadata from asset sources for structured search in the media browser
  • using interfaces for feature discovery on asset sources

Asset organization

  • Folders, can often be nested and contain either assets directly or albums (which contain assets)

  • Tags, can be assigned freely by users, just text, low-level (specific context)

  • Keywords, to be chosen from a defined set defined by someone else, high-level (broad context)

    The documentation of Canto e.g. says:

    It is recommended to keep the number of categories as limited as possible.
    Keywords, on the other hand, you can create and use as many as you want.

  • Relations between assets, as offered by some systems to “group” assets

  • Custom fields / metadata, as e.g. offered by Canto allow for arbitrary search and discovery of assets. These can be mapped to existing concepts in Neos (see and or be exposed to some faceted search

In Neos we currently have Asset collections and Tags as user-facing tools to organize assets.