Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
RSS articles currently collapse Jev's judgments into a hidden relevance score. This change asks Jev independently whether each article would interest the reader and whether it would help them learn, decide, plan, or do something. Either dimension can qualify an article for selection, while the feed stays in publication order.
Both scores are retained and shown out of 100 in the web and iPhone story readers, with Jev attribution and a limited-confidence note. Existing interests, priorities, and public feedback supply the profile. Provider failures keep basic selection available; private-update scoring keeps its existing separate opt-in.
Scoring versions invalidate cached decisions and source validators, unchanged articles retry after outages, and expanded stories keep their prose and user state when scores refresh. Optional fields preserve compatibility with older feed data.
Validation: 57 feed tests, web typecheck and production build, targeted Biome checks, Swift lint, and all 11 native FeedTests passed. Jev protocol tests use synthetic responses; no live model-quality evaluation or deployment was performed.
Production browser checks passed at desktop 1440×1000 and phone 390×844: both scores and limited-confidence attribution render, unscored stories show no numeric estimates, no horizontal overflow or console errors, and Escape restores focus to the story opener. All browser data was synthetic.