(For anyone who's ever tried to modify a reservation, the difference is astonishing. If you're booked with the hotel, all kinds of adjustments are at least possible. If you're booked with Booking, Travelocity, Expedia, etc., it's somewhere between hard and hopeless.)
Brave gets it right. Bing, Google and even DuckDuckGo do not.
That's utterly backwards IME. On Booking.com, most modifications are the push of a button (that, crucially, is right there in the UI that I already know) and if I need to do something complex I can send a message and they'll sort it and get back to me. On a boutique hotel website, maybe I'll get a phone number that I have to look up the international dialing code for and then call in that country's business hours, if their website isn't just completely broken.
(Ad) and (sponsored) were ok at first, but then they became the first screenful of results on most devices.
That’s when my childhood BFF google went off the rails.
“Best way to modify the xyz on a raspberry pi 4”
I stopped using DuckDuckGo because it always failed at searches like this compared to Google. Brave was ok
I just put "the 252 boutique hotel" into Google (without quotes) and the first non-ad result is the hotel's own website. It's also first in Brave. And first in Bing. So no difference here.
Generally in my experience, I've never had a problem finding a hotel's website via Google.
haskell megaparsec operators -> should return [0] somewhere high in the result set. Found it on the third page of Brave Search. For comparison, it's the very first result on Google, second page for Bing, and the third page for DuckDuckGo.
Sure, it's a bit obscure, and I don't even write that much Haskell these days. But it's calibrated to determine whether I can rely on the search engine to quickly surface what I need to be productive, and whether I can trust its 'negative results' (I don't see what I want on the results page -> I need to refine my query or take a different approach). The version number returned in the URL also shows how well the engine handles keeping up with versioned documentation pages and aggregating 'link juice' between them.
I try this on every search engine alternative that pops up on HN. Very, very few pass. I find it often works like an adversarial example, yielding completely nonsensical results.
[0] https://hackage.haskell.org/package/megaparsec-5.2.0/docs/Te... (or any other megaparsec-#.#.# version)
That's simply not true. I just did a search now to see if anything has changed. I randomly searched for hotels in the Seattle area for a 5/18 reservation. The 3rd item in a hotel's result [1] is the hotel's direct booking price. It matches the same reservation on the hotel's website [2]. Another hotel's result [3] has the lowest price as direct booking. Again for another hotel in NYC [4], the 3rd item is the hotel's direct price.
[1] https://i.imgur.com/XtDkLZj.png
[2] https://i.imgur.com/gHJgB3I.png
I stay in hotels a lot though, if you travel occasionally and on special occasions I can absolutely understand.
> In continuing our mission to offer alternatives to Big Tech, Brave is planning to release the Brave Search API. Through it, developers and companies will be able to build search experiences that compete on quality with Big Tech. Those interested should stay tuned for more details, or contact us at bizdev@brave.com.
That's going to be very important for search engines like phind which rely on the bing index service.
Also I switched from StartPage to Brave search, but I do wish Brave search had a translator feature. Like on StartPage I just search "translate" and get an input box. I find it better than most other browsers' translators.
I wish Firefox was like Brave, to be honest. Until that happens, I'll stick to Brave for both mobile and desktop.
For the next 20 years I used Firefox. But I could feel the performance difference from Chromium browsers.
I finally threw in the towel and switched to Brave out of the collection of them now. I hated switching because it meant yet another user lost to one of the remaining handful of unique engines for browsing the web. I like competition and standards, and the more the better with browsing engines. Even if it may bring developers some pain.
I've been very happy with it. People gripe about ads, crypto integration, etc. but it can all be avoided and/or turned off. It has the speed of Chromium but with additional features and is privacy-focused.
But to get to the point, regarding the search ...
I'm happy to see they (somehow!) have been able to build their own search engine now. Obviously that's no simple task, I have no idea what goes into powering that but it has to be a major investment.
It seems pretty good. If they can sell API access for cheaper then the others and still return quality results, that seems like it has potential to be a good revenue stream.
They are very clear about the Web Discovery Project, what the purpose is, what it collects, etc. And obviously it's opt-in.
I really hope they succeed with this. Another example of where competition is definitely good.
It could be that Brave is getting better, or Google search is getting worse, or both.
When I was on DDG, I often had to use `!g` to find better search results. That is extremely rare with Brave Search.
The results are accurate, the UI is polished, and the widgets are extremely helpful. It keeps getting better with time too!
What's the business plan though?
Have been very happy with the search results, and for people who don't like the simpler programming tutorial sites you can even make a custom "goggle" to block those from the results completely.
For example I tried brave to search for watch I'm currently considering buying. When using site:sk it gives me 3 results... Same google search returns thousands results...
I also hate how it's not full width.
Google's: https://www.google.com/search?q=ufc+schedule&rlz=1C1GCEU_en&...
Bing's (doesn't have UFC Schedules but has "NFL Standings"): https://www.bing.com/search?q=NFL+standings&qs=n&form=QBRE&s...
Very rarely I will need to hit the Find Elsewhere 'Google' button. This is usually done for niche technical searches where Google prioritizes some forums dedicated to the topic like Reddit or Stack Overflow. I _could_ re-search with the site operator, but after scrolling down with the Google escape hatch there, the flow just seems more natural.
Just as an aside, I have also been experimenting with SearX searches. The experience isn't as streamlined as Brave Search, but I can incorporate Brave Search into my results. I find the value proposition interesting for SearX, but implementation still lacking.
DDG seem to do quite well in that a lot of their users will deem the relevance good enough, perhaps not aware of its 100% reliance on Bing. More often than not new search engine skins with comparable results to Google and Bing do tend to be the actual results of Google and Bing. Apparently the average searcher doesn't know nor care.
If everyone 'donated' at least a few searches a day to true alternative engines, it'd help diversify search, surely. The fact that Google has such a high amount of revenue per search has helped them price out competitors e.g. being defaults on browsers and devices. Can see why Brave would launch a browser to assist/complement search.
Weird how a company can experience NEGATIVE growth for something like browsers and search
Just today, it told me to use Bing or Google for image search.
I understand the reasoning, but it felt a bit like "whelp, we give up"
The "BraveAI" result was halfway decent, recommending a place I've never heard of, but not listing any of the other top ones I know of.
On the sidebar map, it listed a restaurant in New Hampshire. Hilarious, but not what I was looking for.
For searching every day things, brave search is great. For searching technical things, such as scientific articles or programming errors, the results are not just badly ordered - they’re often missing entirely.
At least, that’s how it was the last time I used it, which was admittedly about a year ago. If things have changed, let me know.
Have we forgiven them now?
I get that they need to pay employees.
But still very happy with how Kagi works