Hacker News Reader

Comment by BitwiseFool | Hacker News Reader

I wonder if the solution is a human curated Index. Rather than scanning as much of the internet as possible, manually add domains based on perceived value and usefulness in search results. Personally I'd be happy to use a search without including Livestrong, Pinterest, and USA Today in it's results.

View on HN

> I wonder if the solution is a human curated Index.

Yes, for people who are willing to pay a lot that’s the solution. There’s a reason that professionals in law, etc., pay for and use reference tools that combine search features with human curation/annotation.

But quality human curation/annotation doesn’t scale cheaply to a growing corpus or number of axes of annotation, its at least linear (possibly superlinear, because of supply constraints on quality labor) cost in each.

They'll just corrupt the gatekeeper. The incentives are too strong.

We saw that with DMOZ, which sounds cool at first, until you realize that many categories had an administrator/curator who happened to have a web site in that space.

And somehow he wasn't interested in having other good web sites listed.

I've experienced that multiple times when I submitted topical web sites (that were not my own).

That's a shame. I was hoping there was a space for a programming-only search engine. I'm getting tired of searching for things like "C# ?." and getting the most basic of results about C# itself, Udemy courses, and news about programming.

Federation goes a long way towards solving this. In a federated model there's no single gatekeeper, just a bunch of curators providing links that are indexed according to a shared taxonomy/ontology. You choose what curators to trust.

I'm not overly optimistic about federation. I don't think there's much money in it and who will develop a mass market solution if they're not going to make money out of it?

That was DMOZ.

Reply on news.ycombinator.com