Data Jamboree 2/Annotation Experiment

Curation consistency experiment: Review of Results

Character #2

• difference between depth, length, width
• some failed to post-compose (e.g., increased length)
• need to standardize shape, length, depth, and width definitions
• should then use higher level character (such as 'shape')

Character #3

• some use 'increased count' which is-a count

Character #4

• question of whether basihyal bone precludes presence (or implies absence) of cartilage
• question of how to annotate this looks more difficult than in reality is
• basihyal cartilage absent implies basihyal bone absent (because the latter develops from the former)
  • in fact it can also be that the cartilage is absent b/c it has developed into the bone (completely ossified)
  • hence need to add that basihyal is absent too
• graph view can be very helpful

Character #5

• software should prevent filling in relative entities for qualities that aren't relational
• post-composed relative entity because of uncertainty of whether the full (existing) term would be compatible with the definition the author may have been using (for a different clade)
• difficulty to find the 'bony projection' by auto-completion (because it doesn't pop up when typing the beginnings of 'projection')
• problem of capturing the 'overlaps with' aspect in addition to 'anterior to'
• full text search of the definitions would be very useful

Character #6

• hypural is also contained within the upper lobe of the caudal fin
• difficulty of finding and comparing definitions of relationship types

Character #7

• Software should ensure that entity starts with an entity term if post-composed, not a quality

Character #8

• sphenotic needs to go into the comment field

Character #9

• too complex to express the exact nature of the orientation, so choose just 'orientation'

Character #10

• triradiate versus tripartite