HNRelevant is a browser extension that adds a "Related" section to Hacker News posts, displaying links to similar discussions found on the site. It uses embeddings generated from past HN comments to identify related content, aiming to surface older, potentially relevant conversations that might otherwise be missed. The extension is open-source and available on GitHub.
A novel browser extension called "HNRelevant," brought to the attention of the Hacker News community, aims to enhance the user experience by dynamically appending a "Related" section to individual Hacker News post pages. This extension, leveraging the power of the OpenAI embeddings API, strives to provide users with a curated list of contextually relevant links to other Hacker News submissions. The operational process involves extracting the textual content of the current Hacker News post, utilizing OpenAI's API to generate a vector representation (embedding) of this content, and then comparing this embedding to pre-computed embeddings of other Hacker News submissions. These pre-computed embeddings are stored and accessed through a Pinecone vector database.
The displayed "Related" section presents a ranked selection of Hacker News links, ordered by their similarity score, as determined by the proximity of their embeddings to the current post's embedding. This proximity calculation effectively identifies posts discussing similar topics, offering users a pathway to explore related discussions and delve deeper into subjects of interest. The extension, currently available for Chromium-based browsers, aims to improve the discoverability of relevant content within the vast archive of Hacker News submissions, enabling a more interconnected and enriching browsing experience. It essentially automates the process of manually searching for related discussions, saving users time and potentially exposing them to valuable insights they might have otherwise missed. The project's code is open-source and hosted on GitHub, encouraging community contributions and further development. The reliance on external APIs, specifically OpenAI and Pinecone, introduces potential dependencies and cost considerations for the operation of the extension.
Summary of Comments ( 20 )
https://news.ycombinator.com/item?id=44078024
HN users generally praised the HNRelevant tool for its potential to surface interesting and related discussions, filling a gap in Hacker News' functionality. Several commenters suggested improvements, such as adding the ability to filter by date range, integrate it directly into the HN interface, and allow users to specify which subreddits or other sources to include in the related search. Some expressed concerns about the reliance on Reddit, questioning the quality and relevance of results pulled from certain subreddits. Others pointed out the existing "ask HN" threads as a partial solution to finding related content, though acknowledging HNRelevant's potential to be more automated and comprehensive. There was also discussion about the technical implementation, including the use of embeddings and potential performance bottlenecks.
The Hacker News thread for "Show HN: HNRelevant – Add a "related" section to Hacker News" contains several comments discussing the merits and drawbacks of the proposed tool.
One commenter expresses enthusiasm for the project, stating that they often search for related discussions on HN, and this tool would streamline the process. They appreciate the implementation using the Algolia API and find the results to be quite relevant. This commenter also suggests potential improvements, like incorporating past discussions and potentially integrating it directly into the Hacker News interface.
Another commenter questions the value proposition, arguing that the existing search functionality on Hacker News already adequately serves the purpose of finding related discussions. They contend that adding a dedicated "related" section might clutter the interface unnecessarily.
Further discussion revolves around the technical implementation of the tool. One commenter inquires about the specific algorithm employed to identify related discussions, prompting the original poster (OP) to explain their use of the Algolia API and its reliance on textual similarity. The OP elaborates on how they filter and rank results based on relevance scores returned by the API.
A few commenters raise concerns about potential performance issues, particularly with older or highly commented-upon posts. They suggest potential optimization strategies, such as caching or pre-computing related links.
One commenter appreciates the minimalistic approach and the clean user interface of the tool. They find the presentation of related stories straightforward and easy to navigate.
Another thread emerges around the potential for misuse or manipulation of the related links. A commenter suggests that the algorithm could be susceptible to gaming, leading to the promotion of irrelevant or spammy content. The OP acknowledges this possibility and mentions exploring mitigation strategies.
Overall, the comments demonstrate a mix of positive feedback and constructive criticism. While some appreciate the tool's potential to enhance the Hacker News experience, others remain skeptical about its necessity and raise valid concerns about its technical implementation and potential vulnerabilities. The discussion highlights the ongoing debate within the Hacker News community regarding the balance between adding new features and maintaining the platform's simplicity.