Hacker News Reader

Comment by andreareina | Hacker News Reader

The problem with curation is that it doesn't scale. I'll take a list of links curated by a subject-matter expert over what Google thinks I want, the problem is (a) finding said expert and (b) convincing them that creating and maintaining said links are a productive use of their time.

View on HN

I've been wondering about 'scale' too, but is that really the problem? Sure, we hear 'that isn't scalable' constantly in startup world. But if you look at Google's gazillion links, users hardly ever get past page 1. Most of them don't go beyond the 3rd search result. Shouldn't Yahoo have positioned themselves as the anti-Google, human powered search engine?

It's a problem of breadth as much as depth. Finding the people to manually do this for then thousand, a hundred thousand, a million subjects is going to be a big strain.

Another way of looking at it is that it already exists, as Reddit. The subreddits I'm subscribed to typically have a small collection of high-quality links that are germane to whatever the subject matter is. And when the link to what I'm looking for doesn't exist, asking typically gets me the answer in short order.

That's true that its hard to scale human powered search engines. But on other side we've platforms like Wikipedia and Quora, who are using a right balance of tech and crowdsourcing for content curation.

The problem, I find, is that Google results tend to be OK until there is a commercial intent. Then you just get the biggest advertiser, biggest brand (sometimes 3 links for the same thing) and most seo optimised results.

A human curated alternative would, likely, end the same way as the urge to monetise such a resource would be overwhelming.

That's all about algorithms and platform rules. Every platform needs ads for monetisation. But, by giving power to humans like in hackernews and reddit we can remove that SEO optimization part. Crowd will get power to like/unlike or promote the content.

I agree. Also, other people have tried and failed. I think the biggest similar project was https://en.wikipedia.org/wiki/DMOZ

Reply on news.ycombinator.com