Kagi and Wolfram | Hacker News Reader

Kagi and Wolfram

blog.kagi.com

810 pointscwmartin332 comments

> To ensure result quality, we automatically downrank pages with advertising and tracking, which are often associated with low-quality or machine-generated content.

That is one of the most compelling things I've ever heard about a search engine.

How well does kagi work for niche "reddit queries" like "best waterproof midi synthesizers reddit"?

There are built-in bangs for reddit, and users can prioritize reddit.com as a domain if they'd like.

Not only that, but you can make it rewrite all reddit links to use old.reddit.com

Yes! Instructions here (old.reddit.com is one of the main examples): https://help.kagi.com/kagi/features/redirects.html

I do the same for Youtube links to Invidious.

Interesting https://invidious.io/

You can also change your Reddit settings to use the old UI for www.reddit.com.

Only if you're logged in, though. (I believe?)

It doesn't work on mobile though.

What do you mean?

Even if you set your preferences to prefer old.reddit.com style, the reddit website will only respect that when accessing from a desktop device. On a mobile device you always get the "new" reddit.

Ohh okay. I skipped a comment in the chain and thought you meant either the redirect on Kagi doesn't work on mobile, or the old.reddit.com site doesn't work on mobile.

> How well does kagi work for niche "reddit queries" like "best waterproof midi synthesizers reddit"?

I’ve started using their quick answers to sort through the crud. In most cases, it catches and filters out the obviously-bought Reddit recommendations, surfacing bloggers and niche industry publications that did their own lab work.

It has "lenses" that you can tailor yourself. Just pay and try it. Cancel if it's not your goto in a month.

They have a free trial as well. I didn't even finish half my 100 free search queries before I decided to pay.

In my case 100 was not enough. I almost didn't sign up for Kagi because I hadn't yet felt the value and configuring a browser to use Kagi as the default search engine isn't a zero effort task.

It probably took me about 200 to 300 searches before I was firmly decided that paying for Kagi is worth it. If I hadn't been freshly mad about Google's declining search quality, and displeased with DDG for something they had recently done (can't remember at this point), after the trial I probably wouldn't have paid and just would have gone back to DDG.

The latency of Kagi is so much better than Google or DDG. Just for the responsiveness and ball of "stuff" that Google tacks onto the front of search is an abomination. Worth if for the latency and the kruft ball removal alone.

> I almost didn't sign up for Kagi because I hadn't yet felt the value and configuring a browser to use Kagi as the default search engine isn't a zero effort task.

It's fairly easy to set up now.

I configured Reddit as a "Lense" (similar to what Kagi uses for things like searching across Forums, or news). With that, now I have a simple toggle at the top of Kagi which allows me to immediately turn a search into a Reddit search.

i just pinned reddit, wikipedia to top so my searches will usually display them first.

Thats the best way, because then what you're looking for gets surfaced rather organically, and its right there, so you don't have to go out of your way to find it.

Generally I've started using bangs and lenses for changing searches in a wholly categorical manner. I.e. the !i bang for image search.

I did change the !p bang from podcasts to activating the programming lens, because programming tends to have a lot of terms that overlap with more general language, and so sometimes its nice to swap in and out of that mode.

I did the same and added a custom bang so I can use it from the address bar directly (!r pointing at https://kagi.com/search?q=%s&l=8 where 8 is the lens id).

Probably least a third of my queries are preceded by an !r now. A third of the rest are now question mark queries that activates their AI fast answer. It's like the google info box on steroids since it can answer any query and it works with lenses to restrict the fast answer to specific domains.

By the way, you don't have to manually add a custom bang and point it to your lens id, you can configure a bang for the lens directly in the settings of your lens

You can also define r as a “quick bang” which means you can just use “r” without !

oooh, I didn't know I could do this. Doing it now. Thanks!

I've been using site:reddit.com for ages now on ddg.

It took me way too long to start using Lenses. I've been a Kagi user for a while now, but lenses never really seemed that useful. The unlock for me is that I'm often looking for 3D models, so I added one for all the usual 3D model suspects (Thangs, Printables, etc).

I similarly added one that is for the recipe websites I like the most. It was a game changer!

Do you find that works better than using Thangs directly? It’s a meta-search engine itself.

I have !sgn to search sourcegraph for nix code.

Forgive my ignorance. I'm curious as to how that works.

For extra fun, add a rewrite rule to change every result from reddit.com to old.reddit.com: https://help.kagi.com/kagi/features/redirects.html

can you share your reddit's lense?

I think quite well. I've been using kagi for a couple months and I'm super happy with it, esp for programming-adjacent searches. You should give it a try! (not an investor, just a happy user).

Honestly, I don’t think I really notice the difference, but I’m just happy to pay for anything not google.

Really well, I just have Reddit pinned so it’s (almost?) always one of the top results, without having to add “Reddit”. HN is another. Similarly, I have lots of domains that I’ve always hated seeing in results, nuked, so I never see them anymore. Kagi is amazing.

Just entered this query as is, without configuring any lenses, and got a bunch of results from /r/synthdyi. I suppose it's good? I imagine if you use reddit a lot, you should bump its relevancy and add a lens for it, as other folks here described. But from that I see, it does decently even without that - even for topics where there's a lot of garbage contents, like reviews, it returns mostly the legit review sites, even though choosing which one do you trust would be a challenge. If you have your preferences, then bumping the relevance for these sites is easy.

Kagi has a built in 'Forums' search that limits results to Reddit, HN, and other high-quality user discussions. It's probably my favorite feature.

Google had that and I used it constantly. Then they removed it of course. I can't remember if it included small blogs as well.

Yes you're absolutely right, I used it too, it's a shame they got rid of it but Google will be Google

Reddit has ads and tracking (so does facebook, twitter, Instagram, etc) not to mention your local paper.

Do they filter out reddit or just small ma and pa sites with adsense?

I never realized before that sentence, but the presence of ads and tracking on a webpage is correlated really well with me never wanting to read its content. I might sign up to Kagi just for that.

I’ve been using Kagi for a while now. It’s worth a shot. I’ve been very happy with the quality of the results and I never find myself feeling like I should check google too.

Please do try it, it’s great. I’m just a happy user, no affiliation.

My favourite feature is the customised down ranking I can do. No Pinterest etc for me.

To me Kagi is as much a life changer on the web as ad-blockers are: Whenever I use a computer without them, I realize how poor the user experience is and how much time and energy is wasted just to filter noise from ads and poor content.

it seriously is. I have grown to resent advertising and mistrust sites that depend on advertising since "What is the ulterior motive here?" is a question that I have to ask myself.

Zero ads. User centric features being rolled out. Kagi has improved more in the last 6 months than google did in 10 years that I used the product.

Search definitely sucks, but I don't know if it sucks enough for $10/mo. I'll try Kagi and see, though.

Happy Kagi user here.

Try it! After you'll be on a site that is just generated content, click backwards in the browser and remove the site from all your future kagi results. That's how search should work (and it should _not_ provide you a top 30 of random AI content).

That just convinced me to sign up

The fact that people who are not affiliated with the company go out of their way to suggest people to use kagi and pay with their hard earned money is a testament to how much people enjoy and value the product.

Personally, if I could own Kagi stock, I would.

Also, I'd love to work for Kagi.

wink wink, nudge nudge.

> Personally, if I could own Kagi stock, I would.

Google stock would be worthless without its ad empire. Without ads, they would have no need to acquire Youtube and Android.

$10/mo also would not be enough for Wall St. Ask Netflix, YouTube, Spotify, anybody really. Ads will always be just around the corner.

Yeah. I chime in all the time to recommend it. It's a great product that I use many times every day, and I don't want to see it go away.

> Personally, if I could own Kagi stock, I would.

I hope it never becomes possible to own Kagi stock. They seem like a great company, I'd hate to see them become worthless chasing eternal growth because shareholders demand it.

They're already worthless then, because they took a round of investor money last year.

> the presence of ads and tracking on a webpage is correlated really well with me never wanting to read its content

It's not just you.

If there are ads on a page, it means the advertisers own the author. He is not free to write what he wants, he is free to write only that which the almighty advertisers will tolerate being associated with. He will not write things which bite the hands that feed him.

I'm convinced writers like that will never write anything truly genuine. Chances are if you see ads anywhere you're reading self-serving generic clickbait content meant to attract attention and drive up ad impressions. It's not real, it's just "content", a generic square around which ads congregate like parasites.

Where do you use a MIDI synthesizer that it needs to be waterproof? In the rain, the ocean, the bathtub? Inquiring minds want to know!

In case your in the bath, poked holes in the bar of soap and want to oscillate the bubbles in to midi notes so you can code your new cloud platform with Velato?

Beach raves. Pool parties. Orgies. Anywhere there's booze/fluids next to electronics.

Your username makes more sense now.

you guys are living in a dreamworld. I'm totally fine with a site that has "tracking" like Google analytics to help content makers and businesses alike make decisions. And have visited many a site with valuable content that might have been tracking me at the same time. This eliminates tons of valuable content.

> eliminates tons of valuable content

They aren’t eliminating it, just downranking it. Given Kagi’s search quality, ad and tracking density seems to negatively correlate with site quality.

And the type of tracking they’re referring to is more like selling-your-data-to-third-parties tracking, not Google or New Relic or something. Maybe sales funnel tracking and interstitial popovers. Your average Adblock doesn’t distinguish between tracking methods and goals but trying to block all pages that load Google Analytics would probably lead to an incredibly small pool of websites, which would worsen the results.

Anti-tracking has gone mainstream enough among America’s elite that I’ve seen executives who are otherwise not tech savvy find themselves unable to load a page (or follow through a tracked link in an email), figure it’s because it’s infested and—on that basis—move on from that vendor.

Honestly, I don't care to visit those businesses. Their sort of maximize-engagement attitude is the exact opposite of what I want.

I'd much rather get information from an individual or small group who is intrinsically motivated and is not just looking for the lowest bar of quality that won't make folks immediately bounce.

I also want to note that just putting Google Analytics on there doesn't kill the site's ranking. There are some sites that are just infested, and those are what get dramatically downranked.

The valuable content is still available in the Google dreamworld for anyone not yet ready to take the red pill.

Businesses can make decisions parsing their web server logs. There is no need to involve a third party data broker (or a half dozen) in the process.

It doesn't eliminate it silly. It downranks it and upranks things without a lot of tracking relative to Google.

So on Google you miss out on things that might otherwise be hidden gems because Google wouldn't rank it as highly.

There are tradeoffs on both sides.

I think this is a solid point. Google Analytics is a killer tool for website admins, and using _some_ kind of enhanced logging/parsing tool like this saves a ton of time over building your own solution or manually reviewing logs.

That said, I am still quite opposed to js that tracks users across the internet for advertisement purposes. I do use ad blockers despite the fact they block a lot of less-harmful tracking by default, which I don't love, but it's too much work to differentiate between. (At least adblock users are the minority of internet users in general, so hopefully Analytics users still get enough data to be helpful for their purposes.)

Kagi has a !reddit bang, but it sends you to the terrible reddit search. I added a custom bang !reddit with "search?q=%s site:reddit.com" as the URL, which just adds site:reddit.com to the search

I have been using Kagi for quite a while. When I’m looking for code stuff at work, I will get results from Reddit that come to the top or close to the top. It doesn’t throw all of Reddit away, even when I’m not specially looking for Reddit.

I just tried your search, without even adding “Reddit” on the end. At the very top was a “Discussions” section, which had a tile for Reddit. The first result was also Reddit, with a few discussions nested under it. The the gearspace forum, followed by YouTube, then it bunches up a bunch of those 10 ten lists that pollute Google all into a section that is easy to use or skip, then sweetwater music, and then funny enough, your comment here on HN. It keeps going, but yeah… Reddit isn’t deranked to the point of not being used, Kagi sees the value in discussion forums when looking for the “best” something.

This comment is now the second result for me, below https://www.reddit.com/r/synthesizers/comments/uhga1j/usb_mi... and above https://www.reddit.com/r/synthdiy/comments/g6cm6q/waterproof...

I've been on Kagi for some months now (six-ish?). Best subscription I carry.

I originally signed up purely out of spite for the SEO scam that is Pinterest (Kagi lets you blacklist domains), but have since been repeatedly pleased with other rankings.

I love that MDN tops the list for DOM ish searches and w3schools is not even in the results.

Using Kagi often makes me forget how awful the Internet can be.

It's OK. One thing it does is rope "reddit" into the "Forums" Lens, which on face seems good but I search mostly for car stuff and forums are a trove there.

Feature request: block a domain?

"saab 900 power steering rack -reddit"

still returns a ton of reddit results.

This is actually already possible!

https://help.kagi.com/kagi/features/website-info-personalize...

You can block specific domains, as well as make them rank lower. And if you ever want to ignore those rules, you can easily do that too!

Not affiliated with Kagi in any way, just a very happy user.

Seems like they are asking to block a domain for an individual query, or to remove it from their forums lens. It is weird that "-reddit" didn't work for them, seems like that should do it to me

Well that or let me pick sources for my lens default! Because reddit just overwhelms it with low quality content

You could create a custom Forum lens and disable the default one in favor of your own. I think a great way to do this would be a "derivative lens" feature, that way you could use all of the sources for the Forum lens, minus the ones you want to remove. I will make feature request for that probably.

-site:reddit.com works for that.

I think it might also be possible to create a new lens from an existing one while blocking a domain.

Im sharing some saved Kagi searches:

best waterproof midi synthesizers reddit https://kagi.com/search?q=best+waterproof+midi+synthesizers+...

best waterproof midi synthesizers https://kagi.com/search?q=best+waterproof+midi+synthesizers&...

Wait, why do you need a waterproof midi synthesizer?

With everyone saying how great the "put down websites with ads is" there are exceptions (few probably)

One of them being vietnamcoracle.com, which is, without a doubt the best travel website for Vietnam (if you like to really go deeper.) None of the content is sponsored, just relevant ads (here you can rent a bike, book a tour etc) and they are discrete in general.

Between him researching new destinations and writing / updating/ replying to comments he just doesn't have a lot of time for the classic "money making job + sideproject".

I just imagine (hope) he's not the only website like that out there.

I don't currently use kagi but would prefer if there was an option to filter by ad type (everyone hates pop ups) and amount.

Anyway just my 2c.

With wolfram and the llm + more users I'm hopeful that in the longer term the price will go down and/or stay at 10$ over the longer term.

Hopefully the defaults handle nuance like you mentioned, but one of the other benefits of Kagi is you can boost/de-boost domains, so you can correct for things even if they're using an overly blunt approach. If enough people do that, it'll also bubble up on the leaderboard: https://kagi.com/stats?stat=leaderboard&k=1

I’ve spent money on things because I had to, I’ve spent money because I wanted to. I have rarely spent money on something because the idea just sounded good and been so consistently pleased with what I received for my investment. It’s not free, but Kagi is a small investment for a vastly improved (and customizable) search experience.

Best $10 I spend every month.

I'd say food and water would be your best $10 investment

I'd rather cut $10 out of my food budget than my search engine budget. As it stands, I spend a lot more than $10 each month on food.

But how will you live with yourself when every search for library documentation or a programming language doesn't flood the first page with expertsexchange, w3schools, geeksforgeeks, favtutor, freecodecamp, and a pile of other tissue-thin content covered with popups?

medium is the 2024 expertsexchange

You haven't lived until you searched and found something useful on expertsexhange. The euphoria is unparalleled.

It’s like spotting a beautiful unicorn!

I should boost expertsexchange.com in my kagi search results just for the joy.

I'm so old I remember the times when it was useful...

I went there, and it looks like it's a pay site. It wants me to sign up for a free trial. Not sure how I'm supposed to search it.

I don't know how it works in 2024, but it used to work like this:

- Google would show the result, on Experts Exchange.

- You'd go there and the result would be cut off, and it'd tell you to pay to see it.

- Since Google required the results to actually be there somewhere, you could still scroll waaaay down past all the nonsense to the full text right at the bottom.

TBH with a uBlock Origin + PiHome and some ctrl+f you can even find useful information in those SEO/spam sites ;)

Sex changes aren’t just for experts anymore!

> welcome Stephen Wolfram to Kagi’s board of advisors

This is interesting. I guess we will soon read a longwinded post about it from the first person perspective.

A new kind of search engine

Sir, u r brilliant!

..powered by cellular automata!

Be careful with rule 34, though.

Kagi-Wolfram derangement syndrome here we go!

"Kagi" goes first to get under his skin just a little bit.

Complete with how he played a major role in the early history of search engines.

Hahahaha also sooo true. Little Sergei Brin interned at his startup. And how Apple Steve job and him had the many discussions about search engine before search engines were a thing. Wolfram practically invented search. U will see on the follow up blog post

Hahahahahahaha sooooo trrruuuueeeeeeee Let’s upload also a 2 GB paper with lots Of graphics and Mma code on ArXiv because everyone on ArXiv is looking forward to it

Facts folks no need to keep downvoting. These are facts. But feel free to show ur ignorance and hatred if u must.

Please refrain from such inane, low-effort comments, regardless of how annoying Wolfram is. If you want to contribute to the discussion, say something about how this will impact Kagi.

I'm excited for this. I've found Kagi to be very useful having widgets like this, in addition to the search results being good. I still have DDG as my search engine for ephemeral searches[1] on my phone and I'm getting sick of it. I don't know if I'd say the result quality has been getting worse, it's just not good and hasn't been for years.

Kagi on the other hand feels like an ideal subscription service. It feels like what Google search wanted to become.

[1] To force me not to leave open tabs I don't really care about, my default browser is FireFox Focus, which is strictly "incognito" mode, which means I'd have to manually log in to Kagi every time I did a search.

You can use the session link [0] in incognito so you don't have to login each time.

[0] https://kagi.com/settings?p=user_details

On mobile incognito you'd need to access that session link and go from there. You cannot just use the address/search bar, as mobile browsers like Chrome do not include a config to put the link.

Mobile Firefox allows it.

you have to love that the GP says "browsers like Chrome don't allow" and the two responses under it are "Safari does" and "mobile Firefox does" -- literally the only other browsers in the mobile ecosystem

maybe this limitation isn't "browsers like Chrome". Maybe it's JUST CHROME with the limitation, because Chrome sucks, and has for awhile now.

It's almost like the developer of Chrome doesn't want you to use a different search engine. Funny, that

It's true on Brave, too. When I said "like Chrome" that's what I meant.

On iOS Safari it works with the Kagi extension.

Works even without it and also in incognito tabs. Use your session URL. The config documentation needs some work, it's very unstructured and difficult to follow to be honest.

https://help.kagi.com/kagi/getting-started/setting-default.h...

It is also open source (just click edit). We welcome improvements.

Woah, what an awesome feature!

You can get something somewhat similar to Firefox Focus in regular Firefox for Android by enabling the option to open in Private tabs by default and turning on tab auto-closing. That's what I did for a while when there were indications that Firefox Focus was about to be abandoned (which never actually happened).

Even better though, now that Add Ons are opened up on Firefox for Android you can install Cookie AutoDelete. Then you just use regular tabs and can keep your Kagi cookies (and any other sites you regularly log into) but have it nuke every other site.

There is also a Kagi Add On. It didn't do pretty much anything when I first installed it last year but it might work these days.

Oh, and finally, there is the Kagi Session Link you can use that embeds a token in the URL so that you can more easily use it for something like a search provider in Incognito/Private tabs.

> [1] To force me not to leave open tabs I don't really care about, my default browser is FireFox Focus, which is strictly "incognito" mode, which means I'd have to manually log in to Kagi every time I did a search.

I strongly recommend the Tab Wrangler extension. I set it to close any tabs that have not been visited in the last 6 hours. Of all the methods I've tried to deal with too many tabs, this has been the most effective.

On my phone, my solution has been working well for a few years.

FF Focus is my go-to browser, so most things I look up or do on the web vanish.

I have a second browser (Vanadium) with ~20 tabs open in different groups which I keep open perpetually. Things like my bank's site, my healthcare portal, other things I want to be able to stay logged into. Then there's one group for different articles I'm reading; these I close once I've exhausted the conversation and opened all the links I'm interested in.

(My desktop computer is a different matter entirely.)

Imo ddg's mission is incompatible with today's web. The mission is simply to serve you regular Google/bing search, but without tracking and they do that well, but it is missing the absolute trash fire that those results are currently (yes, they have moved somewhat from that but clearly they are still heavily reliant on Google ranking). Kagi on the other hand, I spend 5$ a month on search that doesn't track me AND gives me great results. The open-source stuff (think of me when a bang now gives you https directly :D) is just the cherry on top.

> The open-source stuff (think of me when a bang now gives you https directly :D) is just the cherry on top.

Oh that was you, thank you!

> but it is missing the absolute trash fire that those results are currently

I don't understand your answer. Isn't that a good thing? Why are you missing trash on DDG?

IIUC, "it is missing" was used in the sense "it is missing the forest for the trees" -- as in, the Big Picture is that search results are turning to garbage steadily, but if you shrink your perspective, hey at least you're not being tracked ¯\_(ツ)_/¯

What I meant is that DDG focuses on privacy while promising the same results as Google. This used to be desirable, but now DDG ['s leadership] is seemingly unaware that Google's results are no longer good

So.... this is about a pay-subscription search engine called 'Kagi' out of Palo Alto, founded 2018.

Made me think and have to do some digging for the 'Kagi' out of Berkeley founded in the 90s where you could register your shareware purchases in the days before PayPal. Which seems to have been largely scrubbed from history outside of some WayBack snapshots.

https://www.macrumors.com/2016/08/01/kagi-shuts-down/

https://tidbits.com/2016/08/04/kagi-shuts-down-after-falling...

The two are unrelated, but they do reference the shareware platform in their FAQ.

https://help.kagi.com/kagi/faq/faq.html#are-you-affiliated-w...

Technically they're tenuously related as apparently they bought the name (and maybe the domain?) from whomever got it from the bankruptcy.

Wow, that was a blast from the past, I had completely forgotten that's why the name seemed familiar.

Wolfram Alpha still has the same old bug:

https://www.wolframalpha.com/input?i=2016-04-04+to+2019-01-3... = 2 years 9 months 27 days

https://www.wolframalpha.com/input?i=2019-01-31+to+2016-04-0... = 2 years 9 months 26 days

The duration should be identical but is off by one day.

One day a space mission will fail due to this bug.

Reported several times, they never cared.

Both show 1032 days. Do space missions depend on such fuzzy defined time stamps as (calendar?) months? I think time dilation would be a much more serious factor.

Is that a bug? I interpret it as from midnight 2016-04-04 to the first time which has the date 2019-01-31. That includes the full day of 2016-04-04 but nothing from 2019-01-31. This is contrasted with the opposite direction which is midnight 2019-01-31 going backwards, thus not including 2019-01-31 and all the way up till the first time which is 2016-04-04 non-inclusive.

Date time arithmetic is weird ;)

OK, I see, but it shows a different value for the absolute number of days, that's weird.

If that were true, then this would be false:

https://www.wolframalpha.com/input?i=2019-01-31+to+2015-10-2... = 3 years 3 months 10 days

https://www.wolframalpha.com/input?i=2015-10-21+to+2019-01-3... = 3 years 3 months 10 days

This is fascinating. Even sillier example, https://www.wolframalpha.com/input?i=2019-01-31+to+2016-04-0... and https://www.wolframalpha.com/input?i=2019-01-30+to+2016-04-0...

both show an identical duration of 2 years 9 months 26 days (edit: despite reporting 1032 and 1031 days respectively).

Counting months and days are two very different metrics.

Months are lunar. Days are solar.

Depends on the calendar system, as far as this discussion goes, the Greogorian calendar system is solar even for months.

If you use "from" in the second example, the answer is the same as in the first ... anyway the system should figure it out.

Do you still have any of the reports open? I might have some debug info. This appears to be an error with consecutive months with 31 days.

Jan 2024 - Nov 2023 example (both cases show "2 months 1 day"):

https://www.wolframalpha.com/input?i=2024-01-31+to+2023-11-2...

https://www.wolframalpha.com/input?i=2024-01-30+to+2023-11-2...

Aug 2024 - Jun 2024 example (same bug):

https://www.wolframalpha.com/input?i=2024-08-31+to+2024-06-2...

https://www.wolframalpha.com/input?i=2024-08-30+to+2024-06-2...

The same thing doesn't happen from Jul 2024 - May 2024 (results vary by 1 day):

https://www.wolframalpha.com/input?i=2024-07-31+to+2024-05-2...

https://www.wolframalpha.com/input?i=2024-07-30+to+2024-05-2...

I’ve been a paying Kagi customer for a few months now. It has reconciled me with search. Result quality is great, and the tools like fastgpt and the summarizer are precious.

I was a bit reluctant but I don’t regret doing it.

Really happy to see Wolfram added to it!

It feels like a glass of water on a hot day, when you didn't even realize you were thirsty. Suddenly you can find things again.

Recently I've been adding ? to a number of queries, to play with its knowledge graph and answering capabilities. It's been remarkably useful at surfacing information.

The feature to block sites is very very helpful. It is so much easier now to find small suppliers, get rid of all the blogspam. I can even block legit merchant sites that just aren't good and that I don't want to see clogging up my results.

Can you expand on the “adding ?” - I’m unclear what you mean..?

Adding a question mark to your query will open an AI-generated quick answer with references.

Oo, thank you, I didn't know that!

this has probably been the first and only new features added to a product in years, that's actually been useful to me personally.

I think it was Scott Galloway who said that advertisement is a tax on the stupid and poor. It’s been true for news for a long time. It’s becoming, with Kagi, true for search.

I like Kagi, and this integration seems useful, but I'm wary of search engines showing "widgets". That is often a slippery slope that incentivizes them to add more features to keep me on their site, instead of leading me towards what I'm looking for.

A search engine should show web results based on my query. That's it. Some summaries and highlights are useful, but show them in the sidebar, and make them optional.

If I need a calculator, I have plenty to choose from, including the Wolphram Alpha site.

If I need an answer to a question, LLMs do a good job at that.

Please don't make the common mistake of making your search engine "useful". My average session on your site should last seconds, which is the time it takes me to see the results, and click on the most relevant one. If you achieve that, you're doing a great job.

Respectfully, I disagree. If I can get the answer without having to click through to another website, I'm a happy camper.

you have option to turn it off from settings

I'm concerned regardless.

If I am paying for a product and a 90% majority of its features I don't care about and only care about 10% of it working really well I'm going to feel like I'm overpaying since I only use and care about 10% of the product, but pay for 100% of it.

And if I were a paying customer I'd keep getting more and more weary if I see the focus of the product keeps being this 90% I don't care about.

TIL that Wolfram Research has 1000 employees. what do they... do? what is Wolfram's revenue? i'm completely out of the loop. This is the size of Automattic - which makes Wordpress. is Wolfram on that scale? idk.

Mathematica is a commercial product: https://www.wolfram.com/mathematica/pricing/

Who uses Mathematica, in this age of python/Julia etc? The Grad student world seems to be captured by Matlab

Every university on the planet uses Mathematica, and many schools too.

Also Maple. Many non CS folks use Matlab over Python.

I work at a mid-size private university in the US. We hold a license for both Matlab and Mathematica. Both are used in our labs and for independent study and coursework.

I believe that NIH also licenses Matlab for use on BioWulf, but I never see more than a license or two being used at any given time. Pretty much everyone I work with uses R or Python.

it would be nice in my field...a lot of computational neuroscience projects seem to default to MATLAB, so i have to keep my license active to use them, although I feel like Python is making inroads.

Also, if you don't need to run Mathematica offline, their cloud version of it is completely free (with some storage limits that you're unlikely to exceed). It's actually a pretty cool product, but the pricing model is a mess.

If you are doing anything serious then your problem will be the free cloud cpu limits. This is why most peoplr would prefer offline version. Also many universities and institutions will have policies against doing this work on the cloud (data policies... etc)

You know how Python is described as "batteries included" because it has a lot of things in its standard library that in most other languages would require downloading and installing third party libraries?

Mathematica is "nuclear power plant" included. It includes not only a very wide range of functions but also a huge amount of data.

For example several years ago they started adding data on biological organisms. They now have data on over a million species of animals, with hundreds of properties for most of them. You could for example with one simple command get a scatter plot showing bird weight vs bird lifespan. Or for mammals lifespan vs number of teeth. It's just a ridiculous amount of data and it is all easily accessible for computing and graphing.

Same for astronomy. It has a ton of data about the solar system. It's got physical properties of numerous bodies, and orbital data too, and it has functions for various astronomical computations. So if you wanted to see how the distance between Mars and Earth varies over time it is a simply command to get a plot of that over say the last 50 years.

Want a list of all geological formations where T. rex have been found? It can make that list for you, because it has a bunch of geological formation data and it has a bunch of dinosaur data.

They've got data on countries, and cities, and geography, and demographics, and economics, and more.

Other systems can beat Mathematica in specific areas, but nothing else comes near its breadth, and in those specific areas where something else beats it it is still usually decent in those areas.

Mathematica is still unparalleled in computer algebra. Yes, you can do computer algebra with Matlab or sympy, but I know a few theoreticians who use Mathematica to derive complicated equations.

I'm gonna get a lot of hate for this but... people who work on scientific computing or engineering and value their time choose Matlab and Mathematica over Python. Source: I've used Matlab and Python extensively, Mathematica a little (never Julia so I don't know)

I gave Python a good shot but gave up. A beautiful language but what a waste of time trying to get any simulation going.

> This is the size of Automattic

Automattic has almost 2000 employees.[0] I believe that number even automatically updates as folks come and go, but I could be wrong.

[0] https://automattic.com/about/

The headquarters are in Champaign, IL which is a small town with a much lower cost of living than Silicon Valley - so salaries are 20-40% lower than Automattic or other comparable mid-sized firm. The main revenue driver is Mathematica (now called the Wolfram Language) which has several lucrative and long-standing academic licensing deals in place. I don't think the other endeavors (Wolfram Alpha, Wolfram cloud, book publishing) have been nearly as profitable.

(I worked at Wolfram Research around a decade ago, when it was ~500 employees)

I interned there and got an offer over a decade ago; notnsure if better now, but starting pay was 50% silicon valley.

Wolfram research was lucky in getting A 6 m usd relief package donated by Us US law abiding tax paying citizens. Smart cookie how he got this donation.

How do new search engines like Perplexity and Kagi avoid getting rate limited or hitting captchas as they build and maintain their indexes?

Sites have to make exceptions for Google, but likely wouldn’t care enough to allow other search engines in.

I’m fairly sure Kagi reuses Google and Bing search results, then de-craps and applies your customized preferences to them. I’m not sure it even runs its own spider.

Kagi sends “anonymized API calls to traditional search indexes like Google, Yandex, Mojeek and Brave, specialized search engines like Marginalia, and sources of vertical information like Wolfram Alpha, Apple, Wikipedia, Open Meteo, Yelp, TripAdvisor and other APIs,” and also maintains its own “web index (internal name - Teclis) and news index (internal name - TinyGem)” [1].

[1] https://help.kagi.com/kagi/search-details/search-sources.htm...

Umm, So Google and MS can leave Kagi high and dry if it eats into their revenue?

they can, on the other hand, kagi launched recently something that would improve the crawling process

Thanks for correcting me on this.

My take (as a subscriber) is still that the majority of information I'm seeing is assimilated and distilled from the more mainstream sources, but I'm possibly underestimating how much comes from their own indexes.

I will admit that for news I tend to use the !gn bang to jump straight to Google News as it tends to be a more complete feed than Kagi's news tab, so wasn't thinking about TinyGem. But I'd also forgotten completely about Kagi's "Small Web" initiative for indexing non-commercial content that's not interlinked enough for other engines to prioritize it. I believe that's what Teclis (and possibly TinyGem as well) concentrates on.

IIRC, the few times I tried routing searches into that index I didn't find much, so I set it aside, but that's probably because I'm usually searching for more widely-consumed content.

This does not seem to be correct. I asked Kagi "does kagi build its own index". From the first result (on help.kagi.com):

> Kagi Search includes anonymized requests to traditional search indexes including Brave, as well our own non-commercial index (Teclis), news index (TinyGem), and an AI for instant answers. Teclis and TinyGem are a result of our crawl through millions of domains, focusing primarily on non-commercial, high-quality content.

[Edit] I guess it is possible that Google or Bing are hiding under "traditional search indexes including Brave".

I believe Kagi just pays to use Google's and Microsoft's indexes.

Google sells their index??

(Microsoft famously sells theirs to DuckDuckGo, right?)

Bing index has always been licensable , Yahoo was using it too, and they have other customers .

I don’t think Google has a publicly known licensing deal ,

They sell API access to Google searching.

Microsoft sells Bing to a ton of search engines, it has been the go to for everyone who wanted to run a search engine.

Ecosia have also been relying on Bing, but recently announced that they've now added Google as well.

I don’t think that’s correct.

Kagi's internal index seems to focus on sites that wouldn't necessarily block spiders.

https://help.kagi.com/kagi/search-details/search-sources.htm...

I've crawled small web sites (even hosted on major providers) to archive them and never hit a wall.

Kagi has tried a few different AI solutions, and now this integration with a different search engine. I think what I want is just a redoubled focus on improving search results (not quick answers) while maintaining or improving confidentiality.

If they announce that my monthly payments are going up, and I suspect it's because they're paying for partnerships with services I don't view as core to the product value that brought me in, I would feel cheated. And if they add those integrations and don't charge more, I'd have to wonder about that too: are they trading something for the integrations, or are they charging me too much?

Neither, we are passing all the savings (we get better search API deals as we grow, AI is getting cheaper...) into making the product better, at no additional cost for the user, to ensure we are still around next year.

I think that in 5 years nobody will want to use a search engine which doesn’t have AI capabilities. The stuff which turns you off may be critical to Kagi’s survival.

A prescient and pragmatic reply. I too, felt a bit miffed that they were jumping onto this hype bandwagon.

However, as an early user, I've continued to support them and I think you've articulated really well why they're doing what they are doing -- and what made me not drop my subscription when I saw them integrate "GenAI" into Kagi.

Don't forget that the standard plan of Kagi recently got infinitely cheaper, it used to be limited to 1000 searches a month and switched to pay-as-you-go afterwards, but now it has unlimited searches.

I believe cutting out the whole GPT craze would be best. OpenAI bills are always anything but negligible and I think most users would agree lowering the subscription cost or using it for things like Wolfram integration may be best for the core product.

I use the language model features of Kagi many times a day. Specifically, I pay for Ultimate, which gets me access to several different language model providers. I find it to be a great supplement for things that are hard to search for.

> ...We are even thinking of providing a consumeable API for users to add their own widgets to Kagi search results triggered on demand (and shareable with others via an open-source repository[1]).

Neat!

1: https://help.kagi.com/kagi/support-and-community/open-source...

I‘d love to see an IMDB widget.

That would be pretty great. Even better - a movie widget that integrates IMDB and JustWatch. JustWatch tells you where you can actually stream/buy/rent a movie… I find it very helpful for dealing with ever-changing streaming catalogs.

https://apis.justwatch.com/docs/content_partner/

This is probably similar to the Wolfram|Alpha + Siri integration (which AFAIK no longer exists):

- Kagi will rely on Wolfram's okay-ish stack to immediately add features to their engine

- Kagi will, over time, 1-by-1 remove the reliance on the (generally slow and expensive) Wolfram API with built-in features

- Once the meat of the features have been rewritten internally by Kagi in more performant and self-controlled ways, remove the dependence on the slow and expensive Wolfram API

Not saying that's a bad thing, but it's quite likely this is how it plays out. We've seen it before.

I can see them doing that for some things, but kind of doubt it for the more mathy things.

There are things that move the needle (unit conversions) and things that don't matter as much ("how many tea cups of water would fit inside the moon"). It's the "things that matter" that get implemented and make the rest of it not worth it.

if that was the plan, they wouldn't have Stephen Wolfram on their board

It is expensive I'm sure - and they attested to that in their Discord - but it doesn't seem slow.

I pay for Kagi’s duo plan, $14/month or $151.20/year [1], a discount which implies a 10% cost of capital.

At that discount rate, a lifetime membership at $1,680 would be rational for Kagi ($600 individual). And at that price, I’d pay! Hell, do $3k flat (5% cost of capital; ~$1k individual) and throw in an annual in-person event and I’d call it a great deal. For Kagi, moreover, it would be permanent capital.

(I would also love for Kagi to launch a K-12 education product, possibly marketed at first to PTAs.)

[1] https://kagi.com/settings?p=billing_plan&plan=family&period=...

I don't think this would ultimately be healthy for Kagi. The continuing pressure to maintain the value of the subscription since people can cancel at anytime IMHO drives their success.

There's also a paradoxical effect where paying for something can make you value it more, whereas if you paid for it once and now it's "free" you'll value it less.

Once off lifetime memberships suck. They seem good but once they've got my money (and spent it in that year's budget or whatever), I'm basically just a free user with extra obligations from the company's perspective.

I strongly believe part of the reason Plex has gone down the shitter is because they sold too many lifetime memberships and expenditure has outstripped recurring revenue.

The steve jobs example has a typo, "place of dearth"

(https://kagifeedback.org/assets/files/2024-03-05/1709652268-...)

Morbidly, it's still true.

Hey! This is off-topic, but I'm not sure where else to post it.

Does Kagi hang indefinitely for text searches for y'all? I experience this behavior about 1/4 searches. No error; just no page load. I am about to cancel my sub and switch back to Google.

This does happen to me sometimes! Not 1/4 though, more like a few times a day. At first I thought it was my internet connection but it happens often enough and is usually reproducible for about 15 to 20 seconds, so does seem to have something to do with the query.

Works perfectly for me on various browsers and operating systems, and for my friends as well.

Hopefully the kagi team will see this though and chime in, or you should report the bug

That sounds like a good bug report for https://kagifeedback.org

Make sure to include example searches when this happens.

Try getting on contact with them. They've been helpful when I had a problem. I've born tried kagifeedback.com and their e-mail support (two separate questions), and in both cases I received help promptly.

I get sent to the login screen maybe once per day. Very annoying if you searched from the address bar because your query isn't preserved anywhere. Not actually logged out, resending the query works.

@the__alchemist emailed you

I love using WA as a calculator since it's so good at handling units.

Silly example: I just tried "average distance to moon/Eiffel Tower height" in Kagi and got "= about 1.167 million" (powered by WA).

Good to know! This will help in figuring out when reading the news and they describe everything in tennis courts , school buses and football fields. How do you convert from tennis court to school bus?

You forgot Olympic swimming pool as both a length unit and a volume unit.

I once worked out that an Olympic swimming pool would hold about 500,000 human being's worth of blood. The topic was Communism.

It can't handle non-numeric quantities like "average distance to moon" but Google's calculator has handled calculations with units very well for 15 years at least. It definitely puts Apple's Spotlight calculations to shame.

I find that W|A works fine until it fails to parse my query and then I have no idea how to fix it, because the syntax is unclear.

There should be some doc on the English syntax. I wonder if WA lets you fall back to the proper Mathematica representation. Something like the below but in reverse as WA typically does the free form by default I think.

https://wolfram.com/mathematica/new-in-8/free-form-linguisti...

You can input Mathematica (ahem, Wolfram Language). I suppose yet another use for LLMs: better translation of natural language input into structured Wolfram calculations. Unsurprisingly, they're already advertising "Wolfram GPT" and have added LLMSynthesize[] etc. to Mathematica.

I assume the translation is a bit more rigorous than running through an LLM.

Which is the problem: It is using natural language processing techniques that were state of the art in 2009, but have been completely eclipsed in the past couple of years. The "rigor" tends to be at odds with the fluidity of user input: typos, search query-ese, ambiguity, etc.

The challenging problem that Wolfram|Alpha tries to solve is conversion of natural language queries to structured ones. Although I doubt Wolfram's parser has been completely static for a decade, the most recent generation of language models are vastly better at translating natural language queries to structured ones. See also: how terrible Siri is at everything.

Ugh I can’t wait for Apple to roll out Siri II, re-baselined with contemporary LLM tech. So much of Siri’s architecture is exactly what I want… it’s just too dumb.

I just tried some time math and it got it wrong

3:38 - 0:10 = 8 hours and 32 minutes

I'm guessing it was trying to convert it to times of day instead of 3 minutes 38 seconds.

Interestingly if I click the quick answer button it gives 3:48 as a result.

Google - doesn't even try.

WA directly also had to nudge it to use it as a unit instead of a time of day.

Still some work to do here it seems!

> got it wrong

But your query was ambiguous as written.

8 hours and 32 minutes is how long it takes to go from 3:38 pm to 10 after midnight? Perhaps the am/pm thing is also underspecified, so it chooses the shorter time delta?

That makes sense, ideally it should come back saying it's unsure but here's the best guess. I like the way WA explains it directly. The problem with these quick answers is you still can't blindly trust them unless they show their work.

Exactly!

This is pretty damn cool. It might finally push me over the edge to give the subscription a go.

Just do it. Kagi has been my favorite $10/mo spent before this collab already.

I pay for it for my entire company. Life's too short to make people click through to page 9 of Google to get past the ads, the spam, the malware, the made up AI bullshit etc.

Just the other day I was searching for something on iOS and the results were really suspicious, they mostly didn't match the query or didn't make sense at all. My brain was going on autopilot and it took me a few tries to realize my Kagi extension was disabled and I was using Google.

I also spun up a Kagi account for work. The current Teams product is fine, but hasn’t been a priority for the team and lacks features available in the Ultimate plan.

I’ve wound up moving folks over to individual Ultimate accounts. They’re working on a revised one payer/many users product. Once that’s available I’ll consolidate everyone there and experiment with making it the default search experience in our managed browsers.

The cheaper 300 searches/month subscription is a great way to try it out.

I‘ve been using it for two years as my primary search engine, and they’ve been nailing it. Kagi‘s worth every cent. No ads, no need for DDG g-bang, because Kagi search results are above par, excellent support via email/feedback, „quick answer“ feature gives gpt reply, and now Wolfram.

Happy Kagi user here.

Random aside: this blog is the first time I've seen a monospace font used for prose that doesn't look absolutely terrible; in fact, it's very readable for me. Inspecting with Firefox, I see the font is Menlo--and looking it up, it even seems like it's a web-safe font. Pretty interesting.

AFAIK that's the default monospace font for apple devices as well.

A lot of people may not realise that Apple's Spotlight has "instant knowledge" built in, too. The "norway gdp" example given in this post also works perfectly, telling me the GPD of Norway without having to actually do the search.

One thing I miss from Google is relatively seamless integration with maps. Sometimes the address queries just fail for me on Kagi. And even when they don’t, Google’s maps are still far more useful IMO.

!gm is one of my most frequently used bangs on kagi

Why do you use bangs and not keywords in the address bar? It's so much faster, no need to go to kagi first...

I have 10-20 keywords set up in Firefox for example. It works in chrome too.

But maybe I'm missing some use case for kagi bangs...

The real killer for me has been the fact that the keywords or bangs aren’t tied to a single platform/browser. I set them up once and they are available on my iPhone & iPad & 4 different browsers on the mac. Add this the fact that if you set Kagi as the search engine, you can perform all of these from the address bar already.

Aaaah, now it's clicking. This is indeed really cool.

I run into this sometimes, but there's nothing stopping me from !gmaps until I'm happier with kagi map results.

Tangentially related: does anyone have a recommendation for adding a Kagi search widget to Android home screen? I have read some related documentation [1] but it's not super clear if what I am looking for is supported.

[1]: https://help.kagi.com/kagi/getting-started/setting-default.h...

Its easy to save a shortcut homescreen widget with Firefox Android. But, that's just a link to kagi.com, not a type to search widget.

If you're using Firefox and have Kagi as your default search engine, then just adding the Firefox search widget seems to do the trick.

Set kagi as the default search engine in edge on android (you need to install edge) and add the edge search widget to your home screen. I also switched to the MS launcher so I could get rid of the Google search bar.

kagi is the new apple: shut up and take my money

Yeah I’m very seriously regretting passing on the private investment opportunity that early adopters were given.

I love the idea of Wolfram throwing his brainpower and associated ego behind this problem.

Wow! This might just convince me to use Kagi as my primary search engine.

Crystal lang is their secret sauce!

I just tried all those demo queries in Google and it had similar output for all of them, except "MangoldtLambda[11]". Looks like that's a function specific to Wolfram.

Bing couldn't answer "MangoldtLambda[11]" either, and it also couldn't answer "differentiate x^2 with respect to x", and didn't show a graph for "norway gdp".

I feel Kagi can be good just because it is not relevant for sites. Once Kagi gets to a size where it is profitable to be in its first page results, it will be the same cat and mouse game that Google plays with the SEOs white/grey/black hats…

Does Kagi use Google to host their blog or am I misinterpreting the log entry of my pihole?

🞬 blog.kagi.com (blocked ghs.googlehosted.com)

Yes, Kagi is using GCP (+ AWS + some baremetal)

Is there a plan to eventually switch over to self-hosting?

It seems like it might be bad to use a competitor for cloud services, although Kagi is small enough right now to not be on Google's radar...

I bought it as I needed it for an exam that needed me to do a lot of research and it combined with bing gpt really was a big help. I probably won’t use at as a main driver as I only have the 300 search a month plan and I get a feeling like I wasted a search when I searcg something stupid like x documentation.

> I get a feeling like I wasted a search when I search something stupid like x documentation.

I have unlimited searches with Kagi, but I have still gotten into a habit of obsessively taking notes. I personally use Org Roam, but I think anything with hyperlinking and allows you to take notes non-hierarchically is important. This kind of information is worth searching the web once, taking note of, and then using your notes later to access.

While yes, realistically this is just a lonesome longwinded approach to bookmarks, I have personally never found myself going back to browser bookmarks. I actually refer to my notes all the time though.

I'm just happy if I can do some basic W|A without it being so slow. Things like dates or solving simple equations with unit conversion work great but is so slow on W|A. The actual wolfram language is fast, I just find their website feels slow (it's probably only 30 seconds)

This is a wonderful and fitting collab!

I’m really curious if they still maintain their privacy promises with wolfram results

Wolfram isn't really in the ad business, as far as I know. They probably retain queries blindly for metric tracking, but my (informed) guess would be that it's very separate from and PII.

Not to mention that presumably all of the requests will just come from Kagi itself, with no indication of which user asked for the information.

Really glad to see so much Kagi love here. It’s such a great search experience

Pity it can't subtract time.

> 16:28 - 14:50 = 22 hours and 22 minutes

although, Google never could not too.

In WolframAlpha (no Kago subscription, but I assume it works just the same), to achieve your desired result, you merely have to use:

> 14:50 - 16:28

> 14:50 to 16:28

It makes sense when you think of it.

I’ve been using this for years, and it works wonders for datetimes too.

To me it seems to compute number of hours in the interval between those times.

I don't use kagi. On WA, this search results in the answer with a prompt with the engine's assumptions, and includes a link to change the assumption to "math" from "word."

These prompts don't show on kagi? Seems like including them would be an avenue for improvement.

Yes, one vexing thing is the inability of search engines to convert time. 5209s to hours and minutes? Oh, I still need to tell you how to do in steps. Why is this so?

> inability of search engines to convert time. 5209s to hours and minutes?

Kagi just resolved “5209 s in h” quite perfectly.

Yes that's the news here, it now does all the math for you. Other search engines don't know how to do it -or- they only do the first step. Your example doesn't do what I asked. I want input to be converted to hours, minutes, and it should ideally give me both outputs: additional remaining seconds and the decimal. Wolfram Alpha's output has what I want, but it also adds what I don't:

Input interpretation

convert 5209 seconds to hours, minutes

     Results

     1 hour 26.82 minutes <==OK

     1.447 hours <==Not asked for

     Additional conversions

     1 hour 26 minutes 49 seconds <==Acceptable interpretation

I'm getting the correct result: "= 1 hour and 38 minutes"

This is the amount of time between 16:28 today and 14:50 tomorrow. You got a correct answer. Just not the answer you were thinking of.

Anyone using their own searx-ng, the metasearch engine? Have you tried Kagi? How is it comparatively when it comes to quality of results given that with searx I can customize the engines and such.

I was relying on my own self hosted seaexNG before moving to kagi. I still have and use the instance from time to time. Using kagi is better and made my life easier. I stopped worrying about rate limited, bad UI and actually managing and following updates. It also did not solve the problem of ranking. Garbage results still showed. Kagi solved all of these problems and more and I didn't have to spend couple of hours per month to follow updates and manage my instance.

We have both available as a result of independent & simultaneous innovation/20%/side project things. I’m biased because I own the Kagi side, but have thought about this a bit.

SearX has some advantages. It’s free, except for the compute and labor costs. It runs on our infrastructure, so SSO is easy and we have complete visibility on what’s leaving our perimeter.

Ignoring nuanced search quality evaluation, SearX main disadvantages are (1) it’s only available on our network, so if you set it as the default search that feature breaks if you’re not at the office or on VPN and (2) it’s slow. Users get a blank page for a meaningful moment before results show up.

Kagi is snappy AF and adds solid AI features. Their Universal Summarizer (which isn’t actually LLM powered) works well and the fastGPT is the new standard for open ended information retrieval IMHO. (I have a Perplexity subscription I don’t bother to use)

What I want from Kagi is (1) a single payer team plan that gets all the features, (2) more information & contractual guarantees about privacy. I argue that Google is a terrible default, but the brand is trusted while Kagi is still mostly unknown.

Now, the real benefit of SearX would be integrated custom search providers for in-house data sources. We haven’t done that, and I’m not confident we have the resources to do it well. This is also something I’d love to get from Kagi, but it’s not well-aligned with their current roadmap. I’m content to use them to search external data and pursue other solutions for internal stuff.

I kept getting rate limited or blocked with selfhosted Searx. Kagi is much better since it just works every time.

Every day I'm getting closer to buying a kagi subscription.

Just do it.

i love kagi results, its a great search engine, and wolfram is quickly becoming the backend of the internet - which is fantastic

fantastic news, really glad to hear about this

Too bad they do not offer a self-hosted option where you subscribe to updates to the index pushed to your host.

What a mistake by Vladimir Prelovac to get Stephen numbram involved. I was considering continuing My subscription to KAgi but Finding out that kagi becomes A wolfram toy - no thank u. I pass. Wolfram alpha works only half the time Correctly for anything practically useful. And half of the time it works, it’s pain To make it work. Uggghhh wolfman.

Someone did not like a dose of truth.

I absolutely love Kagi. The summarizer feature in itself is incredibly useful. Wonderful engine.

Somehow the search function will turn into a cellular automaton in the near future.

Kagi is well worth the money. Just converted my monthly sub into an annual. It just flat out works better than any other search engine I've come across. Also, the ability to just filter/weight sites that it returns is incredible. I'm not sure how Google is this far behind at search, but here we are.

This is trite, but the simplest answer is that Kagi's product is search and its customers are users, and Google's product is ad impressions and its customers are advertisers.

How is Kagi compared to You.com?

I’m, Doggo clearly has a boner….

Why am I starting to see a lot of promotional statements within HN? Are there any bot or some army of manual labor around?

I think mostly a lot of cool stuff is being launched lately. But if I were a tech company, I'd definitely have a HN Plan ;)

Kagi just happens to be pretty popular here, it's how I found out about it.

I have tried Kagi and I don't find anything magical. And I see a lot of testaments saying how good Kagi is, and most of them aren't in any specific comparison.

For me just the ability to filter/downrank certain domains and the ability to rewrite urls (universally, without having a separate plugin setup on mobile devices) is a very specific example of what I like.

Makes it easy to ensure that I'm always on old reddit and wikipedia with the better (older :p) layout. Also allowed me to clean up a lot of the results that I knew were from low quality domains, such as getting rid of pinterest since results from it almost never led to anything useful.

I also enjoy the lenses, I hated how the Google results for recipes tended to only make already frustrating recipe sites even more frustrating to dig through. The small web lens is also very nice for when I'm already searching for a niche topic.

Besides that, I also reflect the vague description of the search results seeming to be better.

And of course on top of that, there's the lack of tracking and putting my money where my mouth is regarding preferring to pay for a good privacy preserving service.

What exactly are you looking for in a testimonial? Do you want a side-by-side comparison of results or something?

My experience has been that Kagi's customizations are the key selling point, allowing you to block and bump domains to meet your needs:

* I have Wikipedia pinned, so if there's a Wikipedia article it's always the top result, whereas Google has lately started downweighting Wikipedia much of the time (as a simple example see COVID-19).

* I have Pinterest blocked entirely, so I never see results from them. As an example, try "living room refurbishment ideas".

* I have MDN upweighted, so it tends to rank higher than random people's blog posts for web dev queries.

If you need specific domains, you can always search within that domain. With Chrome, you can even setup shortcut to search certain places like Wiki. Most features I can do within Google or Chrome, and many things are still better within Google search and sometimes being paired with Chrome browser.

> If you need specific domains, you can always search within that domain. With Chrome, you can even setup shortcut to search certain places like Wiki.

That's not the same thing as a pinned domain, though: just because I want Wikipedia or MDN to always surface if present doesn't mean that I'm not interested in what comes further down the page. Doing a domain filter requires me to do two searches to see the same content I'd get from a single Kagi search.

> Most features I can do within Google or Chrome, and many things are still better within Google search and sometimes being paired with Chrome browser.

If Google is satisfactory for you, that's great! I'm just sharing my own perspective, everyone's search habits are different.

>Kagi just happens to be pretty popular here

_Pretty_ popular?

Outside of Rust, Kagi might have the most fanatic following of any topic on HN. Which, while I'm sure Kagi is very happy with, is really rather annoying as a someone who doesn't care for it. Every conversation search-adjacent is just comments of "I pay for Kagi and it was the best dollars I've ever spent".

Yeah that's fair, I can see how the gushing might be annoying to someone who isn't interested.

I tend to just ignore or hide the posts I find very annoying, but I understand that isn't necessarily how everyone is.

Is every other search tool's complete failure to come close to Kagi in being useful, and marginalia in being interesting, a particular pain point for you for some reason?

>Is every other search tool's complete failure to come close to Kagi in being useful

This is exactly the fanaticism I was referring to.

When I say this, I have some already voted down (you know who), and some voted up. If you notice, you will see Kagi keeps appearing on HN very often lately, and often there're people talk very good thing about Kagi. This clearly supports my suspicion here.

I find that people often don't fully know how to search and blame Google for not having the feature. Google even implements things in the browser that people can take advantage.

If people aren't smart/skilled enough to get good results with Google, yet get good results with Kagi, then Kagi is clearly a better product.

People are just hyped about a new search engine dramatically improving their online search experience, we're not bots (or paid). Just check out some user histories.

Kagi founder here. Also a long time HN user and as such I too share your concern that we get too much exposure here. I do not know what to say but that I am eternally grateful to HN, as it is what enabled Kagi's existance. We have no marketing budget and majority of our users come from HN, so I can confirm that the comments you see here are authentic.

Hi, I appreciate your response here. However, having no marketing budget and comments being authentic don't clear those Kagi related comments from being associated with Kagi.

I would interpret "authentic" here to mean that they are coming from real and authentic HN users who happens to be kagi users. The intersection is probably high enough.

A lot of HNers would have very positive views and comments about any particular thing, which does not mean they are associated with them.

I have multiple comments on HN, Reddit, and Lemmy recommending Kagi with effusive praise and apologizing for sounding like a paid ad.

I use Kagi as my search engine, and GNU/Linux for my operating system. I think both people who use each of these like recommending them to people for the same reason: like like using it, and the greatest danger to not being able to use it is if no one else does, as that would cause it to cease. People recommend it because they genuinely think other people would enjoy it, but also selfishly because they want to continue to use it themselves.

I tried Kagi and it was meh. I may try it again if they actually integrated unit conversions in search results.

Kagi does unit and currency conversion.

It doesn't work that way in the actual search results:

https://kagi.com/search?q=iphone+case+%2450..%2460

All my results are in dollars. And they're not in range.

Huh? How do you expect it to work? Do other search engines support this?

How do you expect “iphone case $50..$60” to be interpreted as not in dollars?

Show in USD, CAD, EUR, GBP, BTC.. etc.

I’m just confused why that would be expected behaviour unprompted, and particularly after you specified “$.”

I don't expect it to do any of these conversions, however it would be a noticeable improvement in terms of search.

There was an old article explaining how Google interpreted "Polish" from "polish" in its search results. Of course, kagi, Bing, Google, etc. can no longer find this article..

Try Kagi’s “Quick Answer” feature. You could probably specify a multi-currency lens.

Did you try quick answers? It works fine for your query. You just need to specify the currency. https://imgbly.com/ib/SfcLVHFMmV

Did kagi recently (past 12 months) add the unlimited plan? Last time I checked even my human search usage was way beyond their limits (Google regularly accuses me of being a bot!), and my primary interest was in 100Xing that volume with autonomous agents.

Now it says unlimited, but I'm not sure "you and a swarm of a thousand digital clones" counts as fair use...

yes https://kagi.com/settings?p=billing_plan

To rephrase my question, was the unlimited plan not available about a year ago? Otherwise I don't think I would have disqualified kagi at the time. I recall the volume being too low for my needs, even without hooking it up to any automated systems (which for me at least, was the whole point of looking into paid search engines).

Iirc it was available but more expensive as part of the Ultimate $25/mo plan.

They made the $10/mo plan unlimited in September 2023

was it not available about a year ago? I'm not sure. Maybe the wayback machine can tell you. But I have a gut feeling that there is a slight misalignment between your use case and their intended use of their service :)

> I'm not sure "you and a swarm of a thousand digital clones" counts as fair use

It obviously does not. You’re a commercial client.

> You’re a commercial client

No, I'm broke and have no income. But about a year ago I wrote some 20 line Python programs that use GPT (feed it text from web searches) and was surprised to find they worked significantly better than Bing's AI search...

I used it at a small scale with Python's (unofficial?) DuckDuckGo API, but I was looking for a way to scale up (ideally for free or very cheap). At the time, Bing's API was the best option I found, and it was pretty expensive.

I think it would be really cool if I could have an autonomous agent doing research on my current interests in the background 24/7, but it seems my best option is a very low volume of (free) web searches to seed a small web scraper, plus possibly a custom-built search engine (see also https://news.ycombinator.com/item?id=38703943 )

You say you're broke witbout income, but from reading your comments you have at least some skills to pay the bills. And I bet that if you put your back into it you can do it without misusing the only good search we've got. Or anything else for that matter. Keep it up!

Thanks. And I'm sure you can succeed without implying your prospects are crooks.

Edit: from your tone I assumed you worked for Kagi, but now I'm not sure

Don`t work for kagi, and definitely not implying you are a crook. I believe your use case is not in good faith, and I also believe that you know that, but given your position it should be ok. And that is where we disagree, because I also believe you can do better. We all can. And sometimes that is something that needs to be said. I really wish you the best, and please do not take my words as an insult. It`s just the observation that my perspective grants me.

There was actually an outage due to this (their first major outage!) a few weeks ago. Someone was scraping via their API a little too hard, which had unexpected effects on upstream systems. So 100x human search volume is likely to be an issue unless you discuss it with the devs first.

The page with the post-mortem for that outage now gives a 404 error, but was archived by the Wayback Machine: https://web.archive.org/web/20240121000935/https://status.ka...

(It was discussed on HN at the time: https://news.ycombinator.com/item?id=39019119)

We switched to a differnt monitoring system and the URL changed to https://status.kagi.com/clrnl9zwl97290beoine8zlvzx

>regularly accuses me of being a bot!

>swarm of autonomous agents.

So....Google is right?

I haven't launched the swarm yet. Google is reacting to (as I said), my "human" search usage.

(Also possibly Brave browser's anti-tracking, which their captcha system hates and actively punishes.)

Apparently the average person does 3-4 Googles per day.. I checked my search history and I'm somewhere in the 100-200 range.

I've sometimes had a similar issue with searching at www.startpage.com. It became increasingly frequent during work-from-home, blocking me with a captcha or a false message titled "Just Checking…" (implying they're inspecting something, but actually requiring that you fill out a freeform text box explaining yourself). After a few dozen times, I started appending to the textbox an appeal for a paid version.

When I switched to eu.startpage.com, it stopped treating me like a bot. But eventually it started misbehaving in other ways: more than a page of ads (more than two, on mobile!), inserting the ads at the last moment so that your eyes and sometimes your click or tap would land on an ad. Occasionally, I'd win by searching for something so esoteric that no ads displayed. But later they started inserting results specific to my realworld location for searches that returned few or no results.

I reluctantly removed the adblock exemption for StartPage, which causes them to now insert a plea to support them by enabling ads. I contacted their support, who refused to believe that the had pages of ads, and they had no interest a paid version.

Then I found Kagi, and tried it every once in a while until it matured. I still sometimes goto StartPage's EU site on other computers, but on my own I mainly use Kagi.

They added the unlimited plan in September last year. Announcement: https://blog.kagi.com/unlimited-searches-for-10

Reply on news.ycombinator.com