-
-
Notifications
You must be signed in to change notification settings - Fork 182
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Is it feasible to use these self-referential pronouns in other languages? #64
Comments
Possibly relevant: https://seo.ai/blog/how-many-languages-does-chatgpt-support @brianpetro I think feasible is a possibility, but supportable? Probably not. At least not now, and probably not for a long while. Reason being is that all of the contributors on this repo are open source developers that speak English. We may speak a handful of other languages, but this isn't a fortune 500 company with resources to support localization. |
Not saying that we shouldn't support this. Just that I don't think we can, and I don't think we can for a long while |
This will be possible soon, even if not supported directly. For example, the self-referential pronouns simply trigger the vault lookup, and I plan to enable adding custom words to do this. So, at minimum, users will be able to their own trigger words soon. Additionally, if someone wanted to contribute a list of the relevant words in their language, then I can implement a "default language" setting. However, I would probably wait for a minimum of 3-5 of these lists before implementation. Thanks for the suggestions @wwjCMP and thanks for looking up that resource @smartguy1196 ! |
Custom words may be a great place to start. This would move the support for multiple localizations to the community. I recommend writing the infrastructure for custom words/localizations in a way that supports importing more than one custom word at a time (like a manifest of custom words). Perhaps writing an import function that parses a json file to load all these new words at once. This would give the end-user community the option to write their own localization files. An end-user could even then host these localization files on GH, so that even more open-source contributors can help with supporting multiple languages/localizations. Sample word_manifest.json for adding support for the Spanish language: [
"yo",
"mi",
"mio"
] |
That would require a local manifest for caching all of the currently added/imported words. |
My idea :
|
@brianpetro recommend renaming with |
The following object adds support for additional languages. obsidian-smart-connections/main.js Line 28 in b515553
This should make it easy to contribute additional and improve the existing languages. |
Problem, in french, you forgot "m', j'" Also, my way to implement this allow any language... I recommend to use a json for the translation, if you don't want to use a text area. |
Just now saw this comment. I did a code review proposing using
Using .json could also potentially resolve that problem. |
Thanks for adding support for new languages in smart chat. Indeed, the expression "que peux-tu me dire sur" is an element that is often used in French. I'm also making for Elliot Meunier (a French content creator, and Obsidian expert) a whole series of teaching videos about the most interesting Obsidian plugins. The course will be released in 2 days. Finally, for external connections, it would be great to be able to add to the embedding, Google Workspace and Microsoft 365. I have requested access to your Beta. And I look forward to covering the developments of your work for the French speaking public. See you soon. |
@mickahouan added "me" to "fr" in version I am ecstatic at getting to see Smart Connections in the upcoming videos! Thanks for that!
I'm still reviewing OpenAI's ChatGPT retrieval plugin. It seems like it builds the bridge from the ChatGPT plugin to vector databases. But there is still a missing link: parsing and syncing the content in your Google Workspace and Microsoft 365 with the vector database. However, my guess is that very soon, especially with the more popular platforms, like the two you mentioned, open-source solutions for parsing and syncing will be created.
Thanks for signing up for the beta. And thank you for covering the plugin! I look forward to seeing how you are using it and further helping your audience make use of it themselves! Brian🌴 |
No description provided.
The text was updated successfully, but these errors were encountered: