1. 4
    feedback on a new kind of Search Engine ask web

Wire frame:

[Enter your query] (Search All Sites) (Search Only Ad-Free Sites)

Buttery delicious or barnacles: What do you think?

The problem it solves for me is when there’s a story that 100 people have written about and I want to find the best one to share. It takes me many many minutes to find the one with no ads, which is generally the best one.

Google is very broken for this use case, because they seem to prioritize sites with ads.


  2. 3

    Google is very broken for this use case, because they seem to prioritize sites with ads.

    I was going to say “If you use ad-blockers, many sites don’t have ads”…but then I read this and I see what you’re getting at.

    It might be interesting, I think. You could maybe use like an reverse PageRank or something, because the traffic of a site is pretty proportional to how profitable it is to serve ads. The problem with this is that lots of good resources get a lot of links in, and still don’t really have ads–and a lot of crank websites don’t have SEO either.

    Maybe index sites by how much third-party nonsense they pull in? I’m not sure anybody’s done that, actually…and then use that to modify the search ranking.

    1. 2

      Interesting. An VMP sounds fairly easy — have a “search page”

      • forward the search to google using the API
      • fetch the top 50 results in parallel
      • count the number of js dependencies
      • sort in ascending order
      • return the result to the browser.
      1. 1

        This is a good idea! Thanks!

      2. 2

        I’ve been plugging every link into https://outline.com/ to generate a pure text version, but it has some drawbacks (and, of course, NYTimes, etc. won’t work with it).

        But I can see what you’re getting at, since there’s no easy way to search for something like this. I would wonder, given a sample of relevant websites for a query, whether the ones without ads will still be of the same quality / relevance as the ones with ads.

        1. 0

          Market research is off-topic.

          1. 3

            This is not a startup. It’s a technical idea that would be open source to make the web better.