You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
... the |-sign [which is added to the end of incomplete words as an indication of such] is treated as a word-separator, rather than as a part of the word (though the error-model deletes all |-signs before the end of a string, so something should be done about that .... Originally posted by @aarppe in #10 (comment)
Generally, I've thought that the marker of incomplete words could alternatively be the tilde (~), as it is similar to a hyphen and not generally expected to occur in words. Or then one could make use of the hyphen (-) which already doubles as a separator of preverbs/prenouns and reduplicative elements, though that could be confusing to the end-users. In the case of the hyphen, one could either always add it at the end of incomplete words (e.g. ni- or nikî--) or append it to an incomplete word only when the correct version of the string would not end in some aforementioned prefix (e.g. ni- or nikî-).
Whichever character we will use, that should be treated as a word-internal character.
The text was updated successfully, but these errors were encountered:
Furthermore, selecting a suggestion that ends with a character marking an incomplete word should not be appended with a space character (as the word yet needs to be completed) (again, warranting a separate issue).
Swapped marker for incomplete words to tilde ~, as that is more similar to but not the same as the hyphen, as a possible word-internal character. Nevertheless, we would need to specify whichever incomplete word marker as a word-internal character.
Generally, I've thought that the marker of incomplete words could alternatively be the tilde (
~
), as it is similar to a hyphen and not generally expected to occur in words. Or then one could make use of the hyphen (-
) which already doubles as a separator of preverbs/prenouns and reduplicative elements, though that could be confusing to the end-users. In the case of the hyphen, one could either always add it at the end of incomplete words (e.g.ni-
or nikî--) or append it to an incomplete word only when the correct version of the string would not end in some aforementioned prefix (e.g.ni-
ornikî-
).Whichever character we will use, that should be treated as a word-internal character.
The text was updated successfully, but these errors were encountered: