So if a spelling or grammar error gets missed, the message can't be fixed without breaking (possibly printed) documentation? That doesn't sound like an improvement. That said, if there is a way of getting rid of the IDs entirely it might be worth pursuing. Not having to maintain message IDs is one of the reasons why gettext won over competing message translation frameworks.
Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds