search

glossarist / iev-data

1 stars 1 forks source link

Parse the new REPLACES column #78

Closed ronaldtse closed 3 years ago

ronaldtse commented 3 years ago

The new REPLACES column indicates which term a term supersedes.

The value is typically in the form of “XXX-XX-XX:YYYY-MM” where X represents an IEV ID, YYYY the year and MM the month.

We should enact this as a “relationship” just like how the SOURCE gets parsed as relationships. This should be encoded as a “supersedes” relationship.

skalee commented 3 years ago

@ronaldtse Just want to double-check: this is relationship between concepts, not their translations, correct?

ronaldtse commented 3 years ago

Correct - concepts, not translations. Thanks!

skalee commented 3 years ago

Thanks! More questions ahead @ronaldtse.

Concept 102-01-10 supersedes 102-01-10:2007-08; concept 102-01-20 supersedes 102-01-20:2007-08; concept 102-05-16 supersedes 102-05-16:2007-08. Term ID stays the same. Is it okay or some bug in the spreadsheet?

How to represent these relationships? Is following okay?

  - type: supersedes
    termid: 102-01-10
    content: 102-01-10:2007-08

Or maybe following would be better:

  - type: supersedes
    content: 102-01-10:2007-08

Or maybe something else?

Our diagrams distinguish ref and content, but honestly, I don't understand the difference.

Should I write these relationships at root level (which seems logical) or in the English term (like main authoritative source)? I mean which of following YAML examples is better? This:

termid: 102-01-10
term: function
relationships:
  - type: supersedes
    content: 102-01-10:2007-08

or maybe this:

termid: 102-01-10
term: function
eng:
  relationships:
    - type: supersedes
      content: 102-01-10:2007-08

?

skalee commented 3 years ago

BTW, isn't it a duplicate of #34?

skalee commented 3 years ago

It's clearly a duplicate. I've copied the conversation to the other issue, and I'm closing this one.