Show HN: Runnaroo – A new web search engine
https://www.runnaroo.com
Why: I started to not be able to easily tell the different between ads and organic search results in other search engines (even on DuckDuckGo), and I have been disappointed with recent UI changes in a lot of the major search engines. My main guiding rule when adding features has been to ask, "is it better for the user?"
Important to know: I still believe Google currently returns the most relevant web results, so like StartPage I use Google's index as the base of the web results.
Some interesting features:
Deep Searching - The inclusion of relevant results from different vertical specific search engines depending on the search query. For example, if a user searches for 'python jobs NY', results from Indeed.com will be pulled in. The long term plan is to connect into the best vertical information sources for each type of query.
Quick Directs - Automatic redirects for a limited number of commonly searched terms (facebook, google, etc.). This is similar to the Google "Feeling Lucky" button, but done by default. I found that most of the searches I did in Google were navigational. It's simple, but surprisingly has been the thing to resonate with normal (non-technical) users.
Strict Search - Optional selection to force all the search terms in the query to be present in the web results.
Full URL paths - The full URL path of the result is available on the SERP under a toggle.
Privacy - The site doesn't do any tracking. There will most likely be a need to put some things in place at some point to prevent abuse, but user search data will never be used for ad tracking.
The number of things I still have to add to the site is incredibly long, but I have gotten to the point where I have switched over to using it full time as my default search engine.
Well I guess I spoke too soon. You do have a google bang. I’ve gone ahead and switched to you as my main search on my laptop.
API.
I think the only way to monetize and preserve user privacy is context based ads. The challenge is that there are very few privacy focused ad networks [0].
So either I need to built an ad network that fits my use case, or try to do direct ad sales. An important part of either of those strategies is user traffic, otherwise no one would be interested in advertising anyway. For that reason I have decided that I will focus on getting people to actually use the site and provide user value first. That does mean I will have to cover the costs in the mean time.
Right now, I do have a clearly indicated ad with an Amazon affiliate link that does display for a limited number of relevant shopping terms, but that won't even come close to covering the operational costs at any volume.
[0] https://news.ycombinator.com/item?id=22362517
Affiliate links might be a good idea in general, adding a partner tag to all links to Amazon for example. I'm sure Amazon doesn't mind, and it doesn't matter to the user, their prices are the same, though clearly stating it is a good idea. For the large retailers that get a lot of traffic and aren't too picky in signups, that could work, but hinges on you getting a lot of traffic that ends up in clicks to shopping related sites.
Unfortunately, I don't think there an ad network to do this. Everyone that I've seen that lets you link to a large variety of retailers without needing to sign up with each one is shady at best (and targets shady sites that wouldn't get affiliate status), so they won't work with the privacy-first approach.
I assumed that with GDPR (and now CCPA) there would have been a greater overall need for privacy focused ad companies, but I may have been wrong.
Building a privacy-first ad network would be a good thing with regards to harm reduction. We're not going to "fix" the need for all the sites wanting to monetize with ads, but if they had a less harmful option than helping Google track your every move, that would be progress.
https://developers.google.com/custom-search/docs/overview
Just a note, it does not return identical results to the Google web search, it is a slightly different search index. Although now there are several companies that sell directly scraped Google SERP results. I believe that they do operate in potential violation of Google's TOS though.
Why does it say "About 7,690,000,000 results" but there's only 10 results available?
I think you are right though. Google has great organic web results, but I find the most value I get from them today are the instant answers and additional structured information that they bring in. It's to the point where I rarely need to go to any website and can find what I am looking for without navigating away from Google. Long term I wonder of the viability of Google's method of obtaining and displaying the data with antitrust considerations, but that is an ancillary thought.
My goal is to ultimately mimic this with the "deep searching" feature. Today deep searches include Twitter, Wikipedia, IMDB, Stack Overflow, and Yelp results, but I have dozens of sites on my list to add, and I'm sure I am just scratching the surface.