Hacker News new | ask | show | jobs
by ivyabc 2022 days ago
(Disclaimer: Self Promotion)

Today, Google accounted for 92% of all Internet searches. Google and Bing together occupy 96% of the search market. Most other search engines, including the popular privacy focused ones, simply get the search results from Google or Bing and reorder them, because developing a search engine is neither easy nor cost effective. Consequently, the few big tech companies control what people see.

Right Dao is different. We are a fully independent search engine, and we have the infrastructure and build the technology from the ground up. That enable us to show the search results free from search engine monopoly's manipulations.

We invite you to have a try, while we are constantly improving the quality and adding more features.

https://rightdao.com

FAQ: open source?

We think search engine's code is a bit sensitive in general. The engine depends on many subsystems, including storage, scheduling, indexing, etc. If the ranking code is seen by the spammers, they could push their spam websites. We are not sure if there's a viable way to open source a search engine.

3 comments

Looks promising. But how do we know you are doing your own indexing and not simply buying in the results from Google or Bing like pretty much all of the other search startups? And if you are doing your own indexing, I see you include some very large sites like wikipedia which is going to cost a lot of money to index on a regular basis, so how are you going to pay for this in a sustainable way?
Our results are different from other search engines. Wikipedia has regular database dumps (https://dumps.wikimedia.org/), which is a relatively low cost to index. Overall, our scale is currently small and the cost is manageable.
Looks very decent. Make it configurable and I'll pay for a subscription.

What I mean is instead of indexing the entire internet with an adhoc ranking, trying to guess what I want, let me whitelist and blacklist domains and let me configure the ranking. I'd begin with stackoverflow, hackernews and arxiv and probably blacklist pinterest and other paywalled gardens.

From time to time your search engine could suggest search results from other sources, so I could update my whitelist.

It could index pdf files or even show their summary with some ml model, perhaps for an additional fee.

Another idea that's been bothering me for a while is searching for movies or songs. If you figure how to show me the most interesting (according to my filters) movies in 2020, I'd pay for that. Even more so for music.

This comment made me envision a tabbed results page with my results from my curated, first choice sources, and a second tab of general results based from the search engine's best guess algorithms.
This is super snappy! How are you funded? Is it right that you don't collect any user information, including ip address?