I switched from ddg which I did ASAP whenever that was. People seem to forget that Google's heyday at search was during a period where they proudly had no business model beyond better search. The response to what is the business model question I believe was "do no evil".
Once you look up who's behind Kagi, language support is no surprise though.
Edit: the only thing I dislike is that logo with "g" dominating and reminding you of Google.
It does take a little getting used to vs Google, but I'd say 95% of my searches are on Kagi now. Occasionally if I can't find something, I'll use Google as backup, but for most things, Kagi is good enough.
For the downsides, it requires linking a payment card so it's impossible for them to guarantee privacy (cryptocurrencies could save the day here) and it completely relies on other search engines like Google and Bing for search results which is a questionable long-term strategy in my eyes.
Got a quote on that? I was pretty convinced they (Kagi) did their own indexing, and results are significantly different from both Google/Bing.
https://kagi.com/faq#Where-are-your-results-coming-from
I meant that they were "completely reliant" on these APIs in terms of utility, not that 100% of search results are external. I'd love to be proven wrong on this but my impression is that without Google, Bing, and other APIs Kagi wouldn't be a functional and competitive product.
I still use Google for Google images
In the beginning I went to Google quite often, but now I'm down to <5%.
Now I use Google only if I want to buy things, or do image search. I look forward to trying Kagi.
Google retrieved content spam and mailing list entries about previous versions of that framework that have been outdated for years. There was no code in the first page of results.
Kagi found commented scripts in a GitHub project interfacing with the framework, which basically did what I wanted to do.
While I understand that SEO is a problem that Kagi does not have yet, I don’t understand that Google doesn’t allow me to simply blacklist Pinterest, Quora and annoying content spam and StackOverflow mirrors. Also I never want my queries to be interpreted for me, by ignoring search words or using synonyms so that my exact query is always a few more clicks away. If you want to appeal to the lowest common human denominator then techies are probably not in that target group anymore. Are Googlers still using Google?
Kagi also seems to be ahead of their time in thinking about search. In a blog post they were talking about personalized AI very early and I only really understood what they meant after encountering ChatGPT.
I split tested Kagi for every result for a few days. I used an Alfred workflow to open results in both engines at once.
Then I switched to Kagi full time and never looked back. Been using it for a few months now and I’m still super impressed.
Sometimes I have my doubts about the results and open google. In the majority of cases the results are not better.
I'm really confused as to why people still state Google as the standard to beat. IMO, that's no longer true.
Edit: I should explain why - a) you search for something, you find that thing. This should be table stakes for a search engine, but it's just not true of any of the others any more. b) No ads, no tracking. c) The lenses are excellent, mine defaults to the programming one a lot and the results are consistently great. d) You can up/down rank or block specific sites! Bye bye w3schools and pinterest! e) Did I mention that if you search for something you actually find it?
There are release notes! With RSS! https://kagi.com/changelog
Experiments (remember Google Labs‽) - https://labs.kagi.com/ai/sum
I hadn't encountered Kagi's summarizer yet - thanks for the pointer.
Their "context search" was interesting given the hype around ChatGPT, and I'm excited to see where that goes. I see that's moved out of "labs" and will be added this month: https://labs.kagi.com/ai/contextai
A bit of self-reflection:
https://labs.kagi.com/ai/sum?url=https://news.ycombinator.co...
Many people have replaced Google search with other search engines such as DDG, Kagi, Brave, Neeva, and SearX. These search engines offer features such as manual control over upranking and downranking domains, the ability to lock search results to a specific region, and the ability to exclude or include certain sites in the results. Additionally, some search engines such as Kagi and Neeva offer the ability to pay for a subscription to access additional features. One interesting feature is the ability to use keywords in autosuggestions automatically, which can be found in Neeva.
For some reason I don't really see other people talk about it on HN, but I absolutely love it. You don't need to use a specific browser (I use firefox on desktop and iOS) or use any other Brave products/features, it's completely standalone.
I also have r! which is google but add site:reddit.com to the end.
1. A very specific error message nodejs was giving me on one platform about ssl. I usually end up doing site:stackoverflow.com
2. A firmware upgrade for an NVR system (hunting through the manufacturer's website was futile, so I did a search site site:manufacturer.com)
And in both of these cases, while I was using ddg, I was only using it as an index into a specific site, rather than searching the internet.
Other things that I search for, I tend to go to a more direct search engine. If I want to know about a thing/person/event, I just search wikipedia. If I am looking up a band, I just search metal-archives or bandcamp. For movies/cast, straight to imdb. For computer products, straight to newegg. For news, I go straight to my preferred new sources. For Elixir questions, straight to elixir forums. Restaurants, straight to maps.
So, what is it everyone is actually using Google (or other general search engines) for?
But for general searches, I'll hit google.
A few months ago I was looking to buy a NVMe drive, so I hit google looking for product reviews, etc. I don't do this often enough, so even though I know some of the usual review sites by name, I don't usually think of going there right away.
I'm looking to set something up on windows at work. I'm not a windows guy, and it's not something very common, so my colleagues can't help. Windows docs are questionable on this subject. Hit google up looking for this and found some random blogs.
For tech searches, I often end up on either some stack exchange site or that product's forum. But not always, and I don't tend to look there in priority. I sometimes get reddit results, but I absolutely loathe the experience, so I try to avoid going there as much as possible.
To your point, I don't do it all day every day, but it's pretty much my go-to way of looking "new" stuff up.
- well-written blog articles
A single article, describing a real dev solving a real business challenge with real code, is the best way to answer technical questions. They provide a lot of context, and reasoning, and have real-world advice which is invaluable to me.
I use Google for this, but it requires care. A lot of sites are poorly written, or copied from other sites, or are gibberish. With experience you can quickly tell if a blog post is going to be of high enough quality or not.
- short answers, or reference documentation
Google/StackOverflow is okay to answer the question "I have tech X and want Y using library Z". Just recently I've switched to asking GitHub Copilot for answers to small questions. It's not perfect but incredibly valuable. The big downside is since Copilot sends code to a server, it's not okay when working with client code. Copilot is great for learning, though.
For reference docs I use Google to find manpage sites, then I setup custom searches for them. Example: https://man.cx/ for Linux docs, and https://ss64.com/osx/ for macOS.
There's also https://devdocs.io/ which is incredibly fast but too webdev-centric compared to my normal work.
- example code
This site is magic! Incredibly useful source code search engine: https://grep.app/
Example: Kubernetes manifests can be complex and mysterious. The following produces 250,000(!) Kubernetes YAML manifests for study. Enjoy!
https://grep.app/search?q=apiVersion%3A&words=true&filter[la...
It also supports !g , so if a search doesn't return expected results I just head to google.
It’s still got room for improvement, but I’ve personally found to to be a marked improvement over ddg.
Thanks for your feedback, we've fixed the !i image search bang.
I started using it as a Beta product, and then continued with a paid subscription.
You can bump search sources up/down so filter out website you don't care about or emphasize those you do like.
They just launched a ChatGPT (Neeva AI) search result integration, which is as good as can be expected - it includes citations, which is nice.
I found more value in its results than Google. And I don't even like Microsoft.
I've made some simple tests, and it works perfectly! congratulation!
Google results are usually only slightly better than DDG, and this is aggressively worsened by Google's extremely user-hostile experience (for this particular user), but sometimes that's sufficient to get me where I'm trying to go.
Rapidly reaching a point where I'm best off searching Reddit and hoping what I want was mentioned there at some point and turns up in a result, but they also have a user-hostile (for me) experience, subpar search, and an ever-increasing amount of plain gibberish noise to sift through.
I'm basically at the "The web was better before Altavista" point.
Also https://crowdview.ai/ for 100% organic, forum search.
I don't use kagi.com because I don't like idea of linking my credit card to search queries.
You're fully capable of answering that question on your own.
To clarify why I’m asking, it seems that linking your credit card to a search engine is no different than linking it to your Netflix or using it for any purpose online. Could your history be linked to you? Sure, but no more so than when using free search alternatives which build ad profiles on you then actively sell that history to anyone willing to pay. Maybe DDG and Brave refrain from this practice, but that just leaves them still searching for a reliable income stream.
I personally pay for Kagi and have been very happy with it. The search results are great and the customization features are a great selling point. So, as someone who pays for search, I was asking for insight on why someone would be hesitant doing just that over and against paying for any other online service. I may be missing something. I probably am, but that’s why I posed the question.
But this search tool intrigues
I also use one extra thing, an extension called 'highlight or hide search engine results'[1]. I have first party documentation highlighted in blue, good community sites in green, lower quality sites greyed out, and content farms hidden completely. I've also been storing page highlights in raindrop.io[2]
Between the two of these I spend less overall time actually reviewing the search results and it is easier for me to return to good results that I found in the past.
Here's the form to do exactly that:
Create a Lens
Name of the lens
Describe what the lens does, up to 300 characters. (optional)
The style of how results will be rendered
Lock your search results to a specific region
Choose Region
Include only these sites in results (up to 10, comma-separated)
Exclude these sites in results (up to 10, comma-separated)
Include pages containing these keywords (up to 5, comma-separated)
Use keywords in autosuggestions automatically
Automatically use your supplied 'Included Keywords' in autosuggest when the lens is enabled.
Exclude pages containing these keywords (up to 5, comma-separated)
File type
Include results before this date
Include results after this date
Yesterday
Show results from Yesterday to Today (UTC)
When checked, before and after times will do nothing.
Why not load it into OpenSearch or some such?
https://blog.mojeek.com/2022/08/mojeek-focus-search-the-web-...
As most of the search alternatives rely on G/Bing in one form or another, there's not many in the position to offer this kind of customisation using an independent index. Disclaimer, I am an employee at Mojeek.
One of the many SearXNG available, see https://searx.space
I switched from DDG when I learned that DDG censored search results during the initial invasion of Ukraine.
But their latency was too high compared to DDG, so eventually I went back.
I do make frequent use of !wen to get a Wikipedia article, if I already know that's what I want.
Google used to be so good that other engines were noticeably less usable. As Google keeps getting worse and worse, the alternatives get relatively better. Even if brave and DDG aren't on the same level as Google ten years ago, they're miles better than Google today.
YaCy is a really cool open source search engine and web indexer, and features p2p index sharing. Im hosting my own crawler, and feeding my browser history in, so it’ll index my history, and make finding things I’ve already seen (eg documentation) easy.
All of what are sold as search engines today that are known to me are just chatbots of varying quality. As far as I know the last search engine was Altavista.
DDG results are usually good on their own now.
With bangs I can use !s for anonymized Google results, !g for Google, !yt for Youtube.
I use Bing at work for Edge browser due to the Office 365 search and also get rid of the nag screens... But I REALLY like the layout of Bing.
Using it for years already
site:reddit.com OR site:news.ycombinator.com %s
It's not free but they aren't in the business of building marketing profiles, only for improving the science of search.
I am a paying subscriber to neeva since the start.
This is the about page:
Neeva was founded by Sridhar Ramaswamy (ex-SVP of Ads at Google) and Vivek Raghunathan (ex-VP of Monetization at YouTube). They met in the early days of search ads at Google, and came up with the idea for Neeva over hikes and coffee.
Both of those jobs are about marketing profiles. The "visionary investors" include the master of all business profiles:
Reid Hoffman - Greylock partner. Investor, entrepreneur, strategist. Co-Founder of Linkedin and founding team PayPal. Investor: Airbnb, Facebook, Zynga. BS Stanford, MA Oxford.
Initially, it seemed evident they were optimizing search for products. It seemed evident the business model was to empower consumers to find products and trust the search that helped them do that.
Now it's not clear to me what they're up to, and I can't tell what the business model is. I'm not really buying the privacy marketing angle, like, at all.
Still the best general engine for product search though I rarely use it over Kagi which I also pay for.
there is no way you rage quit against ads, as an SVP ads at Google, and also bring a VP of monetization with you.
it's not that they are evil in disguise. i very much doubt that. but there is also no way they "fight for the user". When your mission statement is a lie ("Our mission is to serve our users, and only our users."), you have a problem.