[flagged]
Two upstart search engines are teaming up to take on Google
21–30 of 299 posts
Re: Two upstart search engines are teaming up to take on Google
#22It shouldn't be too hard to achieve what Google were good at before.
Their recent search results for me (last 6-12 months) have been so far removed from what I'm searching it felt like a meme.
Even after rephrasing things, more details, special quotations etc that everyone knows as the 'search tricks' the results are terrible.
Re: Two upstart search engines are teaming up to take on Google
#23It shouldn't be too hard to achieve what Google were good at before. Their recent search results for me (last 6-12 months) have been so far removed from what I'm searching it felt like a meme. Even after rephrasing things, more details, special quotations etc that everyone knows as the 'search tricks' the results are terrible.
I'm using Kagi for quite some time. It's invisible to me. I search, get 30ish high quality results per search, and I'm a happy camper. No ads, no seo grafting, nothing.
Moreover, I can block sites and customize my own search results. This feels good.
When I first started using Kagi, it felt like leaving a closed building and stepping out to open air.
Re: Two upstart search engines are teaming up to take on Google
#24Earlier quoted context omitted.
Because it is super expensive and difficult to keep an index up to date. People expect to be able to get current events, and expect search results to be updated in minutes/seconds.
Some sources update faster than others, you could index news sources hourly and low velocity sites weekly. Google does that. CommonCrawl gets 7TB/month, indexing and vectorizing that is quite manageable.
This reads like the “I could build Dropbox in a weekend”.
Re: Two upstart search engines are teaming up to take on Google
#25This is somehow not about Perplexity. Like many, I tried many other search engines, starting with DuckDuckGo way back when. I always ended up Googling (or !g… -ing). Perplexity is the first one that consistently works for both code questions (what’s this error message) and local questions (where’s my nearest store X and when do they close). Now they just need to speed it up a bit - Google queries are effectively inst…
I would strongly prefer a search engine that searches and doesn't attempt to proactively answer questions.
Re: Two upstart search engines are teaming up to take on Google
#26Earlier quoted context omitted.
I’m my second year into Kagi and loving it. I actually just upgraded to get Kagi Assistant (basically, cloud access to every LLM out there). But the search alone is worth every penny, and it’s built/operated fully in-house as far as I know. https://kagi.com
As much as I like kagi and wish it success, it's not a search engine from scratch. Kagi uses other search engines (Google and Bing) wraps them and does a light reranking
Kagi is also building their own index at the background, and mixes these indexes as you search.
When I search Kagi for "Hacker News", results start with this fine text:
65 relevant results in 1.09s. 47% unique Kagi results.
Re: Two upstart search engines are teaming up to take on Google
#27Every new search engine I've seen was a Bing wrapper with sometimes light reranking. I understand that competing with Google was borderline impossible a decade ago. But in 2024, we have cheap compute, great OSS distributed DBs, powerful new vector search tech. Amateur search engines like Marginalia even run on consumer hardware. CommonCrawl text-only is ~100TB, and can fit on my home server. Why is no company buildin…
[dead]
Re: Two upstart search engines are teaming up to take on Google
#28This is somehow not about Perplexity. Like many, I tried many other search engines, starting with DuckDuckGo way back when. I always ended up Googling (or !g… -ing). Perplexity is the first one that consistently works for both code questions (what’s this error message) and local questions (where’s my nearest store X and when do they close). Now they just need to speed it up a bit - Google queries are effectively inst…
I'd rather hear the word from the horse's mouth rather than a hallucinating parrot which paraphrases stuff as it "understands/thinks it knows".
Re: Two upstart search engines are teaming up to take on Google
#29every new search engine i have tried so far, failed to do regional searches properly. for some queries i need regional index without adding my country to the query.
Re: Two upstart search engines are teaming up to take on Google
#30I found this article much more informative:
https://techcrunch.com/2024/11/11/ecosia-and-qwant-two-europ...