Matt Covalt[1] posted[2] about his brilliant https://hndex.org/, a "full-text search engine of articles submitted to HN" last week. This is basically the same, but the results update as you type, in real time.

Thanks to arc-in-space for the idea[3].

[1] https://news.ycombinator.com/user?id=mcovalt

[2] https://news.ycombinator.com/item?id=24079592

[3] https://news.ycombinator.com/item?id=24089763

Can you share the stack for your version? Its quite fast. Anyone know what stack Matt used?

The best text is the comments. That's a search engine worthy of frequent use.

The links (URLs) in the comments (references) are more important.

Idea for project: Filter the HN comments for URLs, dedupe, retrieve content, and include in hndex.

hndex seems easy enough to copy. There should be mirrors.

There should be many HN articles that are no longer accessible at their original locations. Does hndex use archive.org in these cases?

https://hn.algolia.com already does this and more.

Neat that the top two “all time high” links are from years ago. Does HN not grow as fast as I thought it did or are more recent things more controversial?

Cool concept. I wish it had some filters to only show posts with X votes, date, etc since right now a lot of noise posts are filling up the search.

Just felt I’d share an alternative approach which uses no text search:

https://hnprofile.com/

We only use the context of the discussion around an article. You can search profiles as well.

How is the "morale" calculated? The "how it works" link on the "learn-more" page just points back to itself.

I was pondering last week as to what it would take to build this. Very useful. Kudos :)

What is the monetary model behind this. My reason for asking is what would it take to keep this up long term.

One thing I am missing from both this and hndex.org is to have the date of the post under the results.

Love this product would love to how you created if that’s okay

Does this search text of pay walled articles?

it shows me first the results from 2014. What is the logic behind rankimg the results?