Live data from Hacker News

Unlimited Kagi searches for $10 per month

blog.kagi.com

501–510 of 983 posts

Re: Unlimited Kagi searches for $10 per month

#501

Earlier quoted context omitted.

Not disagreeing with your main point but some anecdata on local searches: for me, Google likes to show me the other, larger city by the same name several hundred miles away. If I use Google Maps it likes to teleport as well. Sometimes even serving results for unrelated businesses, like one time I was searching for a photo developer and it recommended a B rate fast food chain. Kagi has done a much better job of respec…

By “local”, I’m mainly concerned with the country rather than city level. I’m not US based and most search engines have that bias. For example, searches about news, public holidays, or certain statistics should default to the user’s home country (however that is determined, which is admittedly a whole other can of worms that even Google doesn’t always handle well).

Ah! Miss on local, apologies

Re: Unlimited Kagi searches for $10 per month

#502
Been using Kagi for 2 months now and am pretty happy. Have been using the $10 plan even though I didn't cross the threshold as I think it's a good cause.

My two main complaints:

- It's not very good with conversions. Say searching '120mph in kmh' doesn't work. Or the little widget with conversions that do work like '120cm in m' doesn't allow to change the input and change values on the fly.

- The image search is super slow for whatever reason. It takes about 5 seconds before images show when doing a search.

Re: Unlimited Kagi searches for $10 per month

#503
post #37

Earlier quoted context omitted.

I think I belong to the generation that grew up using just Google so I cannot comment on quantification. The biggest difference for me is that on Kagi, the first results are always relevant instead of clutter/ads that you see on Google.

I've been around to compare them all (and used to spend countless hours doing search engine research on Fravia's "web search lores" website). Google represented a huge step up in search result quality generally. But in recent years, the quality has really slid – even while tuning results using more advanced Google features. I don't think Google cares much about search result quality these days, except insofar as they…

One of the big differences is that at some point they seemed to basically delist forums, de-rank most blogs, and totally forgot that personal web pages still exist. All those things are still out there, but you are highly unlikely to find them via Google now. Maybe if you do an exact phrase search for the page's title, otherwise, they're buried.

Another is that they tried to get too helpful and it backfired. There used to be a bunch of search operators that you could use to be specific and most of them don't work anymore. Because that was too complicated for most people. So now instead Google just ignores what you searched for and tries to guess what you actually meant, and ends up showing a lot of irrelevant results, some of which don't match what you searched for at all.

The third is that nowadays almost always, Google tries to spin it into a commercial request. The top results are usually for e-commerce sites or something selling products or services, no matter what you search for. It always assumes you're trying to buy something, not trying to find information.

I switched to duckduckgo years ago, and it generally gives much better results, but even it seems to be slipping a bit now. Still way better than the modernized Google, but for how long?

Re: Unlimited Kagi searches for $10 per month

#504
post #451

Earlier quoted context omitted.

Google's gotta make money off Pinterest somehow, based on the most blocked domains on Kagi: https://kagi.com/stats Pinterest is a scourge on the modern web, worse than any other.

Might these stats be biased by the demographics that use Kagi? I don't use Pinterest, and never really encounter it on a daily basis, but some of my friends really like the site. Tbf, they might use stackoverflow more, but seeing Hacker News on the top 5 doesn't seem to reflect the average web user's usage...

I wonder if it is also biased in the sense that it's only certain people who customise these things. And that it's only the most common ones. You would never see my favourite car forum listed here, but bumping it in search results is where I see value with the feature.

Re: Unlimited Kagi searches for $10 per month

#505
post #19

Earlier quoted context omitted.

Can the search result quality be quantified somehow? I remember when Google was new, and I used to use Altavista and Hotbot and those older search engines, the difference in results were like night and day, and Google's results were "wow!" Does Kagi give a similar impression as Google did then?

In my experience it's not that there's better quality results found, but rather low quality results are skipped. There's a lot filtered by default + it's easy to click "block this domain" when you run into yet another stackoverflow copy. It means that when you're searching for code related things, you often get small relevant blogs in position 3+ rather than SEO spam. For example searching for "current time on JavaSc…

> “searching … on Google, I get SO, MDM and basically a lot of SEO spam sites”

I get the same in Kagi clicking your link above.

Both the 1st and 3rd result is SO. 2nd result is MDM.

I’m confused, so what’s different between paid Kagi and free Google search then?

(Note: I’m not hating on Kagi, I’m just genuinely wanting to understand)

Re: Unlimited Kagi searches for $10 per month

#506

Earlier quoted context omitted.

In my experience it's not that there's better quality results found, but rather low quality results are skipped. There's a lot filtered by default + it's easy to click "block this domain" when you run into yet another stackoverflow copy. It means that when you're searching for code related things, you often get small relevant blogs in position 3+ rather than SEO spam. For example searching for "current time on JavaSc…

> it's easy to click "block this domain" Friendly reminder that google had this feature a decade ago then removed it. Hopefully someone in the C-Suite got a few back pats for that decision. https://searchengineland.com/google-brings-back-blocking-sit...

The uBlacklist browser plugin adds support for blocking sites in Google searches.

https://iorate.github.io/ublacklist/docs

Re: Unlimited Kagi searches for $10 per month

#507

Been using Kagi for 2 months now and am pretty happy. Have been using the $10 plan even though I didn't cross the threshold as I think it's a good cause. My two main complaints: - It's not very good with conversions. Say searching '120mph in kmh' doesn't work. Or the little widget with conversions that do work like '120cm in m' doesn't allow to change the input and change values on the fly. - The image search is supe…

> searching '120mph in kmh' doesn't work

120 mph in km/h does work, though. I don't think "kmh" is a normal abbreviation. This is the first time I see it.

Re: Unlimited Kagi searches for $10 per month

#508
post #19

Earlier quoted context omitted.

Can the search result quality be quantified somehow? I remember when Google was new, and I used to use Altavista and Hotbot and those older search engines, the difference in results were like night and day, and Google's results were "wow!" Does Kagi give a similar impression as Google did then?

Maybe I'm simplifying, and I haven't used it. But it looks like you are basically paying for an anonymized (maybe, since you need an account) proxy to google. https://help.kagi.com/kagi/search-details/search-sources.htm... . Getting 10 year old flickr images is great, I'm sure. But probably the most relevant search results still come from google.

> But it looks like you are basically paying for an anonymized (maybe, since you need an account) proxy to google.

Well the whole point is to pay for the product instead of being the product. I am very happy with Kagi, but I mostly pay to show that there is a market for that business model.

Re: Unlimited Kagi searches for $10 per month

#509

For mass market I feel 'free' will always win. With that in mind I wonder if Kagi would be worth having ads for a free version but be super clear what the ads are based on. E.g purely based on 1) search KW and 2) IP location of the search. Nothing more and nothing additional stored. I wonder if privacy minded people would be happy with that as it's very limited data in today's ad world, a massive improvement on anyth…

I don't understand why Kagi needs to be mass market. Why not just be the search engine for people willing to pay for better search? Does Ferrari feel the need to enter the economical midsize sedan market?

Advertising is a lot of work, and introduces a lot of tension into the business model. Spotify premium showed me "promoted content" last week (but they promise it's not an ad). They also accept payola for placement in discover weekly playlists, which means their features are useless to me. There's no reason to encourage Kagi to go down that route. They have a good business model that they and their customers are happy with.

Re: Unlimited Kagi searches for $10 per month

#510
post #396

Earlier quoted context omitted.

I trust you will not, but what about after you exit? Anyway, you or someone else will probably start a new search engine, and we will start again. It's best to enjoy it until that happens.

I believe the founder, freediver. In the case of Radio Shack, they too made a promise never to sell or give the data to other entities. The CEO even fought for that in bankruptcy court. The judge deemed that the user data was worth a significant sum, and the judge screwed everyone over for the debtors. Once you capture the data, it's a toxic but valuable asset. And there's always someone willing to go to any end and…

> I believe the founder, freediver.

Oh, don't get me wrong. I believe them, too. My point was if they ever sell the company, the buyer might change the privacy statement and start collecting/selling information from that point on. I didn't mean the new owners would necessarily have access to the data collected previously.

Post reply on HN