alt Hacker NewsIt really shows you what a bubble HN is.
Every post about Google for years has been people saying it's terrible and dead.
Kagi gets talked about on here constantly, and it's not even on the list (though I suspect there's a reason?)
Even within Google, about a year ago, everyone was saying that Google was dead because of Perplexity, which is barely a blip.
It's kind of shocking to see DuckDuckGo is only about 1%, with everything you hear and how much you hear it within certain bubbles.
Replies
I’d argue that HN sentiment is a leading indicator, not a claim of current reality.
Compared to a year ago, Google has declined from 89.487% to 88.915%. Just half a percent, but IMO it will accelerate.
Meanwhile OpenAI has gone from 0.194% to 0.226% in just three months (they weren’t on previous quarter’s reports).
Sure, it’ll be years before Google drops to 50%. But it will happen.
> Kagi gets talked about on here constantly, and it's not even on the list (though I suspect there's a reason?)
Being at least 10€/month for the only "useful" tier is a powerful reason for that...
Maybe also Kagi being a metasearch engine reduces its visibility? Just speculating, I obviously don't know how it really works.
Google spends something around 30 billion dollars a year to be the default search engine across many platforms. You can spend the same amount and tomorrow your search engine will have 88.9% of searches.
It's not a charity, if people truly preferred Google results over defaults, Google wouldn't give out tens of billions of dollars to be the default.
> Every post about Google for years has been people saying it's terrible and dead.
They are terrible, but that doesn't mean anybody else is good or better. And being better at search isn't enough anyway [1]. Also, when you give Google less of your searches, personalization drops off and it gets even worse, but most people give all their searches to google so they see the benefit of personalization if they compare.
[1] When Yahoo did user research on search, one of their findings was that if you asked users which results were better, there was a strong and consistent preference towards results that were shown as Google results, regardless of the actual results. It's been forever since I saw those reports, so I don't remember the numbers, and the numbers are likely different today anyway, but that's a huge barrier to adoption that you have to manage.
Is it a though? Just bc 99% of population don’t notice or care about misinformation, does it mean the majority consensus is right?
Which do you think is more abundant, lies or truths?
> Kagi gets talked about on here constantly, and it's not even on the list (though I suspect there's a reason?)
Not that I'd expect them high up on the list, but Kagi sends the following response header:
As a result the browser won't send a Referer header with outgoing links, completely excluding them from this report.