fairtracks / fairtracks_standard

FAIRtracks is a JSON Schema defining a minimal standard for genomic track metadata.

Home Page:https://fairtracks.net

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Various documentation changes

sveinugu opened this issue · comments

  • Document fields to be filled by Track Registry as Placeholders

    • doc_info->global_id
    • doc_info->doc_version?
    • collection->global_id
    • track->global_id
  • Check documentation of 'local_id' fields.

    • All should refer to correct entity. "within investigation/hub/track collection" for all except collection
    • collection should be "of investigation/hub/track collection within source repo/document", or similar.
    • All should prefer curie.
    • Add sentence about unique requirement within collection for all, except collection
    • collection should have unique values within document
  • collection ->source_repo_url: Update description to allow for some collection defined as mix-tapes where this field is not relevant. In that case track->orig_coll_ref should be used. update description of that field also.

  • Go through description of all fields related. Update wording related to new model.

  • Change "doi:10.1186/1471-2105-12-494" and possible other references to URLs

  • Change "FAIRtracks JSON document" and "JSON document" to "FAIRtracks document" throughout. But keep "FAIRtracks JSON schema". Make sure that this makes sense. Also, "the current FAIRtracks document" should be used instead of "this" (or perhaps a better wording can be defined?)

  • Once more, clear up the wording of the description of the @Schema property. E.g.: "The absolute URL of the 'current' version of the relevant FAIRtracks JSON schema within the same major version followed by the current FAIRtracks document (ensuring compatibility). Must match the value of '$id' in the linked schema"

  • "identifier for" -> "identifier of"

  • Rephrase "in CURIE-format, if applicable", specifically mentioning identifiers.org? Or should we go the other way around, as there is a US counterpart N2T.net. If so, perhaps make sure that the manuscript does not push one over the other?