https://bugs.documentfoundation.org/show_bug.cgi?id=167649
--- Comment #17 from Marco A.G.Pinto <[email protected]> --- (In reply to Kevin Atkinson from comment #16) > Hi, > > Sorry if this is off-topic, but I am the current maintainer of the US/CA/AU > Hunspell dictionaries, and there were a few things I thought needed to be > cleared up. > > I want to clarify first that the project is not abandoned, nor have I > stepped down. I realize that the lack of updates may make it look that way, > but I have been maintaining SCOWL for over 25 years now. Even after long > periods of inactivity, I always manage to swing back around and create new > releases. I also realize that many users are frustrated by my lack of > responsiveness when requesting new words; however, I now have plans to > address those issues. I hope to create a new release sometime in early > 2026. For details on my future plans, see: > https://github.com/en-wl/wordlist/issues/394 > > The next release will move the project to a new format. This new format > stores words as lemmas with POS and all derived forms (including irregular > ones). This will take some work, but the new information can likely help > provide better-quality morphological data. I am actively looking for > someone to help with this: https://github.com/en-wl/wordlist/discussions/432 > > The new release will also make it possible to provide better-quality word > lists with hyphenated compounds and proper abbreviations (ending with a > period, rather than having it stripped off), if that is something you could > use. > > I hope we can work more closely together in the future. The best way to > reach out to me is via the GitHub project: https://github.com/en-wl/wordlist > > Thank you, > Kevin Atkinson Heya, Kevin, For one year that I have been working on my versions of US+CA+AU. I have created/added thousands of patterns/words in my own tool to properly handle them. GB+US+CA+AU will each have around 280 000 words. The ZA one will have 300 000+. I would like to ask if you can release a script in 2026 that will create a static file with all morphologic information that could be hardcoded into LibreOffice and work with the five English dictionaries. Having non plaintext in the dictionaries files will probably only work with LibreOffice and OpenOffice, and it will break all (or almost all) other apps that will use the files. This is the suggestion I made in this ticket or in others (I can't remember where I wrote it). Thanks. -- You are receiving this mail because: You are the assignee for the bug.
