As a piece of software, nothing. It’s an open source browser, and has an added bonus of having many privacy settings on by default. Not even firefox can say the same, it comes with telemetry, pocket and whatnot out of the box.
But there are some fair criticisms about the company and its administration. For example, there was an incident years ago when you signed on a crypto exchange, it would swap the sign on link for their own referral link. They claimed this was an error and quickly patched it, but I don’t buy it.
You’ll quickly notice that a lot of people on lemmy passionately hate brave. So expect a strong bias and, as a result, truths but overblown, half truths and misinformation. Don’t ignore what they say but double check them.
DDG is so bad in results though it makes not using a search engine seem like a solid alternative. I remember that research from a few months ago where they compared Google, Bing and DDG to do statistical analysis of if and how much Google's search results have been getting worse in recent years.
Result:
Google did get worse.
However, Bing and DDG got worser, faster.
The real conclusion was that SEO spam has found ways to optimize that is no longer easy to exclude for search providers. Hence all search is getting shit pretty quickly.
A lot of people are complaining about Kagi using Brave as a backend but the alternatives aren't much better. Both Google and Microsoft are BDS for example.
The search engine bots are absolutely powerhouse-obnoxious in how many requests they make, and there a ton of them, and Lemmy's not real optimized to cope with the load -- most big instances block all bot traffic simply as a matter of server survival as a result of that. So I would expect not to see any of them in search results any time in the near future.
Search engines currently struggle with the concept of federated posts. My guess is that instead of finding a post's home instance, they accept the first mirror of that post and discard copies from all other instances through deduplication.
I made a crude lemmy scanner, but as I'm using the "http api" I'm a bit stuck at servers and instances. Almost 1800 servers and IIRC (I'm on my phone) some 40.000 communities. Yay!
I know the official api docs, but I don't know how to really use it, just send a json to some server? Would love to have an easy example and I'll go from there (family, work, art, and other gets in the way).
The goal would be to figure out how to make a search engine without storing the whole lemmyverse locally.
It's a little disingenuous so say that "Vivaldi is closed source" and leave it at that. The vast majority of their browser is built on open-source code:
Only a small portion (~5%) of their bespoke UI code is closed. The vast majority of their source is published.
Obviously whether the small portion of the code they withhold is important to you is a matter of your own to decide, but I feel this was important to clarify.