Closed johanneswilm closed 9 months ago
@aphillips This PR adds the link as you suggested. Note though that it introduces a ReSpec warning: The definition in the i18n-glossary is only informative, but we need it here in a normative context. Would it be possible to change the definition in the i18n-glossary to become normative or would you advise something else?
@aphillips To me, actually your initial recommendation of linking to the definition of "grapheme" feels more accurate as we are referring to as something that is perceived as being a character by the user. But I'll be happy to replace "grapheme" with "grapheme cluster" if you think that is more accurate.
@aphillips I have asked the Respec experts for help on this. Hopefully they have a solution.
@johanneswilm I would prefer the grapheme
definition, but your original text was more specific (via the character
definition) in referring explicitly to UAX29's definition of extended grapheme clusters. Our grapheme cluster
definition provides that level of specificity whereas our grapheme
one isn't quite so clear. We might add some text to grapheme
to make it clearer and that's probably a good idea. I'd prefer to reduce the level of opaque jargon overall and avoiding the word "cluster" would be good.
On the Respec thing, let me know what you find out (or copy me!! I'm obviously super interested)
Preview | Diff