Comments & real-time editing
Collaboration is real-time. When someone’s editing a block, you see it as taken; cursors, comments, and resolved threads update live.
When should I use a comment instead of chat?
Comments anchor to specific text — a thread on block 12 about that sentence. Chat is document-level (“let’s talk about this doc”). In both you can @-mention teammates or Literess; mention her and she replies inline in the thread.
Resolve a thread with the tick and it moves to the Resolved view; reopen it anytime. React to any comment with an emoji. The Notifications panel is your inbox for mentions and system events — it doesn’t duplicate every chat message.
Will the AI undo something we already settled in a resolved comment?
No. Transept’s AI actions — translate, regenerate, proofread — still see resolved comments as historical context, surfaced to the model with a
[RESOLVED]marker, so a decision your team already made doesn’t get reverted on the next run.Can I stop the AI from acting on reviewer comments?
Yes. The Ignore Reviewer Comments toggle works on every flow (translate, regenerate, proofread). With it on, the AI ignores comments from human reviewers — but Literess’s own commentary still flows through, since she’s part of the AI loop rather than an outside voice.
Why don’t I get a notification when Literess leaves a comment?
That’s deliberate. Literess’s comments — her own notes and her Smart Proofread flags — appear in the comments panel but never trigger a push notification, because you’re already talking with her in real time. Notifications stay reserved for @-mentions of you and system events like a finished translation.
Do comments cost words?
No. Words are spent only when a block is translated. Leaving comments, replying, resolving threads, reacting, and editing by hand are all free.
Still stuck? Ask Literess in the app, or write to [email protected].