Hacker News Reader

>Whoever solves this problem will probably get to be the new google

If it's solvable. There are many problems that aren't.

Perhaps the new Google would just be a 1996-era Yahoo! human-curated catalog.

I've often wanted something like that yahoo catalog. I miss the days of being able to browse categories of sites.

My hope is for something like YaCy with a web-of-trust overlay. I'd want it to only show results from peers that I trust, and if garbage starts showing up in my search results, I want to know from which peer it came so I can un-trust it.

Which in my mind is somewhat of a middle ground between a search engine and a directory: what I'd want is a whitelist of curated directories that contain domains to be crawled by my search engine.

Ditto. I would like a fuzzy web of trust: I specify connections that I trust explicitly, and it provides results from that network out N steps. Each hop reduces the confidence value.

And yeah, cited sources so I can un-trust sources of junk.

You'll probably enjoy Curlie (created by the former DMOZ maintainers) and Neocities (try out different tags for better results)

https://curlie.org/en

https://neocities.org/browse?sort_by=last_updated

I sure hope so. It's part of the approach we're taking for e-commerce search with https://shopdeft.com

Reply on news.ycombinator.com