Create <IGNORE> class

[mathildavz]

As discussed in PR #1232, it might be useful to have a way to EXCLUDE mappings, or refuse FOREVER to map, when things are unclear. @LinguList suggested that one could map to a new class IGNORE or similar in these cases, and we would then not look at the cases again when doing our check for left-over-mappings.

[LinguList]

@xrotwang, do you think this is a good idea? The advantage is that we can re-map more quickly and don't have to discuss accidentally cases again, which we solved a couple of years ago in discussions. Something like an authoritative "don't map this for the next couple of years".

[xrotwang]

I think this could be an option once we move to Concepticon 3.0 - when public data access should go through a Concepticon CLDF dataset, which will not include such pseudo conceptsets.

But I think it would be nice if assigning a concept to the "IGNORE" conceptset always means adding a summary of the relevant discussion as comment.

[LinguList]

So we could do the following:

1. make a list of conceptlist-concept-IDs with the ID and the comment, maybe calling this file ignore.tsv
2. we insert the ID and the discussion result there

By now, this means we don't touch the list, we just have this information elsewhere without disturbing the evaluation workflow. But later, we could integrate it, e.g., into the command that fishes for unmapped concepts, that it reads the file before?

[xrotwang]

Yes, that would be a good way to start.

[LinguList]

@mathildavz do you want to set this up or do you prefer me or @AnnikaTjuka to do it?

[xrotwang]

Hm. Maybe rather store a summary of the discussion in a comment field? Johann-Mattis List ***@***.***> schrieb am Mi., 5. Okt. 2022, 13:48:

…

@xrotwang <https://github.com/xrotwang>, do you think this is a good idea? The advantage is that we can re-map more quickly and don't have to discuss accidentally cases again, which we solved a couple of years ago in discussions. Something like an authoritative "don't map this for the next couple of years". — Reply to this email directly, view it on GitHub <#1235 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAGUOKADARSN2FW2QXIVBM3WBVTJFANCNFSM6AAAAAAQ4JZG6I> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[LinguList]

Yes, I am also fine with that!

[AnnikaTjuka]

Since the file was added in a recent PR (#1240), I'll close the issue.