non-cognate forms

[thiagochacon]

A few cognate sets have(might have) a word from one language that is actually not a cognate to the whole set.

How should we handle that for the purpose of this project? Should we simply eliminate that toke from the alingment database?

Here is a list of entries
COGID # 15, 22, 38, 62, 96

[LinguList]

Luckily, we don't need to eliminate it, it is enough to change the COGID by just inserting a question mark in the field and then pressing enter. Entering a question mark will force the application to search through all existing COGIDs and take one that is not yet used. So these words will then be separated in future analyses and judged as non-cognate by all algorithms.

[LinguList]

Re-opening this issue for consistency. Please close, if the non-cognate cases have been resolved in the data.

[thiagochacon]

where should I place the q mark "?" ? is it in the CGID column in the exact row where that word is located?

[LinguList]

Yep! Just clear the COGID in that exact row you want to change and
replace it with a question mark (or another cognate ID, in case you want
to assign it to another set).

Just try playing a bit with this feature. In case you want to go back,
you can always use the "undo"-button which appears on top-right whenever
you change stuff by typing.

Ah, and don't forget to hit enter after placing the question mark (or go
down with cursor), otherwise it won't be saved.