Brave Search Discussions

I was initially planning to post this on the original thread but since @Mattches is censoring people again, I had to open a separate thread for this.
Where is the “discussion worthiness” algorithm explained in details?

1 Like

‘Censoring’ might not be the right word.

It is known that they lock threads for new updates posted on this forum like a new browser update release. Steeven creates a post for the update release but the thread is locked. But, on reddit and twitter the official posts are not locked for comments.

Personally I never liked that they continously lock threads for updates made from brave team.

Nobody is censoring anything. Posts like that are intended to be an announcement to bring attention to a new feature or update, etc. If the thread is left open, as it has been in the past, many users pile on in one thread, all with different questions, issues, bugs, feedback, etc and it makes it very messy to try and respond to everyone’s comments in the same thread.

Instead, it is preferred that users open their own threads with their own individual issues/concerns/feedback/bugs/etc. so that we can address them more directly. And if I’m not mistaken, that’s exactly what you did.

I don’t know of one but I have reached out to the appropriate team members to confirm.

1 Like

For now, what’s written in the blog is all we have:

To serve these Discussions, the Brave Search ranking algorithm detects queries where a discussion forum might give an alternative or complementary viewpoint to the search results. This “discussion worthiness” score is based on a variety of signals, including:

  • Freshness (or recency) of the topic
  • The popularity of the topic on a given forum
  • The quality of the conversation (as measured by user engagement, such as upvotes or responses)
  • The search quality score (which measures how relevant the discussion is to a query)

For now Brave Search Discussions come primarily from two sources: Reddit and StackExchange. Brave will be adding more sources for Discussions in the near future.

1 Like

The data given in the official blog seems enough for the end user to know.

The actual data will be in tech terminology, in programming languages, various mathematical algorithms, use of scripts etc

1 Like

@talgeeze ,

Brave Search (beta) in transition.

Reddit was nice 7 years ago, but now it has become a cesspool except from a few select small communities. I only go there when I couldn’t find answers anywhere else, and even then I get thrown up the same useless suggestions I already found from other places. I do my best to avoid it.

Yes this is what I want. Unless they don’t care about transparency and accountability.

So you’re saying your Discussions algorithm doesn’t actually exist and this new feature is just an embedded subquery with “reddit” added to the search terms.


There is also an algorithm working for brave news feeds to fetch the news articles, that too should be open.

The actual brave search algorithm?? Is it open?


Here are my suggestions for a Brave Search Engine:

Slavic languages, for example, that use the Latin alphabet have specific letters that are so to speak identified as diacritic characters in Unicode, but are Brave Search mistakenly recognized as spelling errors, which can often result in poor search results. The solution for this would be to add new regions, but also rely on other search engines such as Yahoo, Yandex, Seznam, Baidu and Naver; not only for Slavic languages. (I don’t like Mojeek.) DuckDuckGo had already solved this problem, and Presearch didn’t even have one.

The new search category could be “Documents”, which means that in this case only documents in formats such as dOC(X), PDF, EPUB, etc. will be searched, regardless of the language.

New options could be opening sites in proxy mode (as Startpage, Meta-Ger or Torry already have) and including search results from Onion sites.

Like Presearch, Brave Search can also allow users to make money by searching, which would be quite independent of a Brave Wallet and Brave Rewards.

Of course, all of this would also be available to users who use the Brave Search Engine on other browsers, which can further motivate them to use it, and then move on to the Brave Browser.

Censorship of search results should not be at all, regardless of what someone is looking for, because only the users themselves are responsible for this, and by no means the search engine. The right to privacy certainly implies responsibility, doesn’t it?

1 Like