Hacker News Reader

Comment by jszymborski | Hacker News Reader

Kagi is a bit of a HN darling so I know I'm putting my neck out here, but even after trying the trial, I'm not sure what the value prop is.

I would happily pay for a search engine that provided good results, but seeing as it's ostensibly Just Bing under the hood, I'm not sure what I get that I'm not getting from DDG or MetaGer + an ad blocker.

Kagi doesn't strike me as a privacy alternative given that all my searches are necessarily tied to a user id which is further tied to a payment method.

Finally, the metered usage feels far too much like an ancient cell phone plan. I really hate the idea of me paying for search only to have to switch to actual Bing when I hit some limit.

I like what Kagi is trying to do, but I don't see it as either sustainable or awfully useful unless such a service owns its index.

Feel free to let me know what I missed, I'm sure Kagi's paying customers are paying for a reason!

View on HN

I used to use DDG, but the search results were so bad that I had to !g all the time. With Kagi, the results (which are mixed from various services, not just Bing) are actively superior to Google (so much less spam!) almost all the time, and the only Google search service I still use regularly is Maps.

That said, while I've been happily paying $10/mo for the pro plan for a while, I'm not convinced the 1000 searches/mo will suffice. Guess I'll find out soon enough.

Update: I checked https://kagi.com/settings?p=consumption and to my surprise I'm averaging under 20 searches a day total, even though Kagi is my main search engine on both mobile and desktop. So looks like I'll be OK after all.

Sadly, for programming, I've had to use Google quite a few times lately.

Their raise feature is nice, but always finding stack overflow (because I raised it) on top of ACTUAL documentation for a specific topic kinda ruins it.

>Their raise feature is nice, but always finding stack overflow (because I raised it) on top of ACTUAL documentation for a specific topic kinda ruins it.

So you changed the results, and now you are complaining because the results are changed? I don't get it.

I think it makes sense. They boosted stack overflow because they think it's often relevant, but not necessarily always the best resource. But now it's always #1 when oftentimes there is a clearly better resource. They just wanted it to be #2 instead of #10.

Basically they are wishing the boost feature had room for more nuance.

There is a bit more room for nuance, you can pin domains, which places them above "raised" domains. I've pinned MDN and official documentation sources, raised StackOverflow and blocked SO clones. It works well for me but YMMV.

Thanks for providing the link.

I used 517 searches in February.

For March I look to be using 700 searches.

Seems like the pro plan will be sufficient for me.

Flip side, it's not uncommon for me to do 200+ searches a day. My lowest month was 900 searches but I went over 1800 frequently. Not worth $25/m for me personally at that point. I was sold on the idea of being grandfathered in at $10/m...

You can still be grandfathered into unlimited, for the next year anyway.

I just updated my monthly subscription to annual for just that purpose.

> I'm sure Kagi's paying customers are paying for a reason

No ads, blazing fast, and the search results are more than adequate for me (I don't find myself ever reaching for Google as I did with DDG). I'm not too bothered where the results come from when those three things come together. It's just the premium search experience I want.

> all my searches are necessarily tied to a user id

I know that theoretically Kagi could sell me up a river at some point in the future. However, they have every incentive not to - they'll see a huge exodus of users if they ever do that - so the alignment of incentives feels right here.

For the moment I feel reasonably confident that my search preferences aren't being sold to the highest bidder (or every goddamn bidder). If that ever changes, I'll reconsider.

Overall I feel more comfortable paying for my search, and I want to support this business model.

> the metered usage feels far too much like an ancient cell phone plan

I'm going to wait and see how this works out for me. If I find I'm exceeding the $10/mo plan by a lot I'll also have to reconsider. But because I'm partly here to support the business model of paid search, I'm willing to give Kagi some leeway to find a sustainable setup.

The alignment of incentives is not as tightly aligned as I'd like. Because Kagi has this data, they could be bought out by another company and release blog posts about how this merger is actually a harmonious union of privacy. A sinking ship with the livelihoods of many families on the line can easily convince leadership to sell. Having a distressed company bought out is not some special story, and things like this easily go on without many customers knowing. How many Tom Bihn customers know that the company has been bought by a VC?

On a mildly tangential note, the way their marketing language describes their new plans leaves a poor impression with me, saying that I get something for free if I pay for it. Did you know? If you buy a Twix for $2, the first bar is free.

Your concerns are valid indeed, and on top of them one must also think about what's the alternative. Because if you leave service A you'll necessarily use service B - I assume nobody leaves Kagi to give up searching. So the choice is between a possible bad outcome (Kagi) versus a known bad outcome (Google).

The difference in privacy I see between Kagi and Google is that Kagi requires me to have a signed-in account to search, and Kagi accurately associates more valuable information with a signed-in account.

I don’t think you can have both ad-free (e.g. paid) search and anonymous search at the same time.

You have to pick one. And you can’t really fault Kagi for providing the one when the other is already “adequately” covered by Google.

cryptocurrency?

Interesting point.

Maybe Kagi could offer a more privacy friendly "fill a wallet with crypto and we'll deduct a certain amount per search" option that doesnt require user billing data.

Im not a current user but would be more likely to test it if available.

Surely you can. You can use blind signatures, it's like issuing anonymous tokens in the style of David Chaum digital cash. Or you can just use existing cryptocurrency.

Supposedly they are not doing it because of advantages of associating your payment method and searches.

I don't know how Kagi internally operates but for billing purposes I imagine it would be enough to record that you searched. Not what you searched.

On your last point, you can check the usage statistics in your Kagi settings for the previous months. That way you can see if you would exceed the limits.

Apparently I search around 1k times per month, which means the new Professional plan wouldn’t fit my needs.

I decided to cancel my subscription, Kagi didn’t provide that much benefit over Google to me. I just needed this last push! I do think it’s a good idea, we definitely need more competition in this market.

I’m a paying customer.

I pay mainly as an experiment to figure out what my searches actually cost.

If those costs are covered by advertisers that I block, my free search experience is paid for by people who don’t block those ads. I don’t like that thought.

I may go back to DDG if the new payment structure is insufficient (I don’t know if I search more than 1.000 times a month). But it is somewhat eye-opening how much I take search for granted.

If you click into your kagi billing, there is a usage dashboard.

I thought I'd be fine with the 700 searches a month.. but my average is 1600.

$300/year for a search engine is probably more than I am willing to pay, especially for an individual subscription.

> Kagi doesn't strike me as a privacy alternative given that all my searches are necessarily tied to a user id which is further tied to a payment method.

I agree to some extent, but also consider: if you are not paying for search, then they are going to have to monetize somehow, and the most likely way they'll do that is through selling search history/data.

The only way to be respected as a customer is to actually be a customer, and for that you have to pay. Presumably Kagi could start accepting anonymous payments, but your searches would still be tied to an account.

> the most likely way they'll do that is through selling search history/data

No? Google, the largest search engine, doesn't monetize by selling search history/data. I would hazard a guess that not a single free search engine monetizes by selling search history/data.

Selling ads is extremely and literally different from selling search history/data.

> Google, the largest search engine, doesn't monetize by selling search history/data.

Google, the owner of the largest ad network in the world, does not make money from your searches? Are you trolling or completely misinformed?

Google indeed does not sell your search history to third parties. They make money by letting third parties target ads to you with it.

I think they are being pedantic. The point they are making is that Google does not sell your search data directly, rather it sells ads that are optimized thanks to your search data.

IMO it's a moot point.

Paying customer since last year. Kagi's results are okay or really good depending on what I'm searching. After around 1 month of experimenting, I blacklisted/boosted several domains on Kagi. After that, I never had the need to go to DDG or bing anymore.

That being said, I think Kagi being a search engine itself AND basing their statistics for pricing on Google and DDG is a horrible move...

For the past 4-5 months I've kept three browser profiles, which I use simultaneously to compartmentalize things across work/personal/hobbies and to try out different browsers and search engines. I use Kagi on two of them and Brave Search on the other (I gave up on DDG as it disappointed me too often; Brave seems slightly better than vanilla Bing as they combine it with their own index).

The difference in results quality is stark. On Kagi I have to !g maybe once every few days, and usually I don't find what I'm looking for on Google either, but on Brave it feels like I do it for a third of my searches, often for ones where I actually know exactly which result I want and am using the search engine as a shortcut. So whatever Kagi is doing, it's definitely not "Just Bing".

From their Discord, it is apparently some combination of Bing+Google+their own small index, which makes sense because for niche searches where Bing and Brave completely fail, Kagi's results seem to closely mirror Google's, with the benefit of its result boosting and filtering features. For more common searches it seems to match or surpass Google regularly.

I've just switched to the annual plan to lock in my current $10 unlimited searches rate, and will re-evaluate after a year. I can't justify $25 a month, but looking at my history (and guessing how much I'd need if I moved all of my Brave searches to Kagi) I may not actually need more than 1000 included + 200-500 pay-as-you-go searches per month, which is juuuust on the boundary of what I'd be willing to pay.

I was curious how much/if any unique index Kagi maintained because I also finally gave up on DDG as its repackaging of Bing became worse and worse to the point where there was no value add vs just going back to Google.

But now I’m even more hesitant on Kagi since I knew it used bing, whose results both direct and from ddg have been supremely disappointing, but it also pulls google which…I can just get direct from the horse’s mouth? So the value add seems to be whatever is in their small index. Hmm I just dunno.

I've found being able to change the weight of different origins in search is super powerful over time. I've filtered out most of the SEO spam and low-quality websites that plague my searches in other engines at this point, Kagi doesn't serve them to me. I've also pinned high quality resources like MDN so it always shows up at the top of the list if it has relevant content.

Kagi saves my time and gets me to better results faster. At least for now, my time is worth more than Kagi charges. As long as the math keeps working out that way, I'll continue paying for Kagi.

I pay because their results are as good as or better than Google, and because they aren't in the business of marketing to me or selling my personal information. That's it.

All this crap with AI and Kagi releasing a browser is a distraction from that.

Those are my exact feelings since testing Kagi. I did not found it appealing as a plain search engine, nor private.

But they have recently shown a tech preview of an "Universal Summarizer"[1] which I have found far superior to ChatGPT (cGPT is worse than some freely available pre-trained models in this regard).

I have asked Kagi and cGPT to summarize my own post[2] which Kagi did rather well (I don't have a screenshot), while cGPT spilled utter nonsense[3], mentioning hardware and software that isn't mentioned in the post and which - in some cases - I have never used.

While Kagi is unappealing to me at the moment, I will happily try it out the future again.

tl;dr: I wrote about using: FreeBSD, Linux, Fail2ban, blacklistd, Terraform, Ansible, Postfix, Dovecot, Kubernetes, Jenkins, ArgoCD, Git and nginx.

ChatGPT stated that I wrote about MacBook Pro, iPhone, AirPods, Kindle, Chrome, iTerm2, VSCode, Vimium, Adblock Plus, Dropbox, 1Password, Trello and Slack.

[1] https://labs.kagi.com/ai/sum

[2] https://wojteksychut.com/posts/work-tech-i-use-privately/

[3] https://wojteksychut.com/pic/chatgpt_post_summarization.png

> I have asked Kagi and cGPT to summarize my own post which Kagi did rather well (I don't have a screenshot), while cGPT spilled utter nonsense

> https://wojteksychut.com/pic/chatgpt_post_summarization.png

ChatGPT currently has no capability to access external sources, and its training data cuts off at the end of 2021, so there's no way for it to have ever succeeded at this task. It was essentially asked to summarize an article it has never seen, so it hallucinated the entire summary based off of the words in the URL as that is the only information the model was provided. For a more equivalent comparison you'd need to copy the text of the article into the chat input.

Of course that doesn't change the fact that it responded to your query with totally fabricated nonsense, which is a horrible failure mode that points to why it can be so hard to trust LLMs' responses, but I do believe the disclaimer popup that appears when you launch ChatGPT notes this particular limitation.

It spelled my last name correctly in 4th paragraph (although could have got it from domain name). This domain was registered in 2023 so it could not have data from 2021; but maybe has been fed from Bing later?

It currently states: "As an AI language model, I don't have direct access to real-world events that have occurred after my knowledge cutoff date of September 2021. However, I can access information and data from websites created after September 2021, as long as they are publicly available on the internet."

> it responded to your query with totally fabricated nonsense, which is a horrible failure mode

Agree. I would be fine with getting "I cannot summarize this post as I don't crawl external sources" but not this.

Reply on news.ycombinator.com