Thanks to @calliem's work, a curator will have the ability to pre-fill EQ annotations in Phenex from a named entity recognition service. Many times, these will need subsequent manual adjustment by the curator, such as deleting redundant or incorrect E and Q combinations, or further tweaking of Es and Qs. A curator may not be able to go through all characters in a single session, and when resuming from file later, it may not be entirely obvious where they left off. As a workaround, comments could be left for the character. However, a visible hint (such as by font color) to which characters have been manually adjusted and which ones have not would be very useful, and would need to be persisted between sessions.
The question is how should such a marker best be added as metadata to the in-memory model so it gets properly written out to the NeXML file and read back in upon reloading the file? @balhoff could you advise?
Thanks to @calliem's work, a curator will have the ability to pre-fill EQ annotations in Phenex from a named entity recognition service. Many times, these will need subsequent manual adjustment by the curator, such as deleting redundant or incorrect E and Q combinations, or further tweaking of Es and Qs. A curator may not be able to go through all characters in a single session, and when resuming from file later, it may not be entirely obvious where they left off. As a workaround, comments could be left for the character. However, a visible hint (such as by font color) to which characters have been manually adjusted and which ones have not would be very useful, and would need to be persisted between sessions.
The question is how should such a marker best be added as metadata to the in-memory model so it gets properly written out to the NeXML file and read back in upon reloading the file? @balhoff could you advise?