StumbleUpon was amazing, I loved it. I knew a load of very non-tech-savvy people that used it as well and remember having some really good conversations about the cool stuff they found with it....
StumbleUpon was amazing, I loved it. I knew a load of very non-tech-savvy people that used it as well and remember having some really good conversations about the cool stuff they found with it. This is a wonderful spiritual successor!
Same story here, god I miss StumbleUpon so much. Back then RSS was the way to go and I had so many webcomics I'd found via SU set up in my RSS reader sidebar in Firefox.
Same story here, god I miss StumbleUpon so much. Back then RSS was the way to go and I had so many webcomics I'd found via SU set up in my RSS reader sidebar in Firefox.
Same here, I loved StumbeUpon but eventually got jaded by all listicle blogs, blogspam (even without AI there was already a lot zombified content on the web back then) and lack of interaction with...
Same here, I loved StumbeUpon but eventually got jaded by all listicle blogs, blogspam (even without AI there was already a lot zombified content on the web back then) and lack of interaction with people on it. I sometimes tried to engage on comments under some blogs but you can guess how well that went.
So when I came across reddit it was really awesome to discover actual discourse and (somewhat) effective user curated content.
I still can't believe I pay for a search engine, but every time I try a search using anything else it all falls so flat. Especially if you do any sort of programming or code work.
I still can't believe I pay for a search engine, but every time I try a search using anything else it all falls so flat. Especially if you do any sort of programming or code work.
Hi, I`m quite curious, (if its ok for you to answer), how Kagi search works with copyright-grey material and adult stuff? Like torrents, different NSFW, etc. etc?
Hi, I`m quite curious, (if its ok for you to answer), how Kagi search works with copyright-grey material and adult stuff? Like torrents, different NSFW, etc. etc?
They obviously know who you are. Their assertion is that they don't record searches in a way that they can be linked back to the account that initiated them. Sort of necessarily, we only have...
They obviously know who you are. Their assertion is that they don't record searches in a way that they can be linked back to the account that initiated them. Sort of necessarily, we only have their word for that, but they don't really have an incentive to do otherwise and the risk to their business and reputation if they did and got caught would be existential.
Isn't their claim more "we promise that we don't track you/link searches to your account, but we understand that you have to trust on that and might not, so for extra security you can also pay...
Which is kind of strange why they would do that, if as they say they weren't linking searches to accounts anyway
Isn't their claim more "we promise that we don't track you/link searches to your account, but we understand that you have to trust on that and might not, so for extra security you can also pay anonymously so that it's harder for us to work out who you are"?
There are numerous ways to operate an application (such as a search engine) in a way that authenticates you as simply a paying or non-paying user, then generate a key that pretty much anonymously...
There are numerous ways to operate an application (such as a search engine) in a way that authenticates you as simply a paying or non-paying user, then generate a key that pretty much anonymously says, "Hey, I am a paying user, here's my key," without knowing who the person with the key would be specifically. I am overly simplifying the actual implementation, but the idea is still the same.
Also, would you prefer trusting a company that claims to not store this information ever, or a company that makes all of its money by literally storing as much of that information as possible?
You can simply make sure never to couple the two distinct pieces of information. You have to trust that they are doing it right, but it's easy. Years ago I wrote voting software and it was easy....
You can simply make sure never to couple the two distinct pieces of information. You have to trust that they are doing it right, but it's easy.
Years ago I wrote voting software and it was easy. When you login and vote, we recorded that you voted. That's it. The part that actually records the vote is separate.
Similarly, they can allow you to login, record that you searched so they can keep count, but not actually record your search. But yes, you have to trust that the information is kept separate.
Depending on how the key is generated and given out, it can easily be anonymized in a way that is not linked to an existing user. Once the user signs in, a token could be generated by a hash based...
Depending on how the key is generated and given out, it can easily be anonymized in a way that is not linked to an existing user. Once the user signs in, a token could be generated by a hash based on Date/time/username/etc. and then easily have been stored in database table that simply stores: TOKEN, PAY_STATUS, SEARCHES_LEFT, and EXPIRATION_DATE, without linking that token/key back to the user. When a user does a search, check if the provided token is in the table and not yet expired, and then provide results if so. Otherwise, require authentication again.
I'm not saying it is done this way in Kagi's case, I'm just saying requiring a sign in does not mean that once your authenticated you're not anonymized. Anonymous authentication has been a thing for a long time.
I agree, thought, it is better to be on the safe side and assume everything is linked to you. I am personally operating under the assumption that it would make less financial sense risking their privacy oriented focus by storing information that, if ever used or found out about, would completely destroy the driving force of its users.
EDIT: forgot to add the SEARCHES_LEFT column name to my example
This is the first time I'm hearing about Kagi, and I'm definitely interested, especially because Google and other engines have been feeling so broken lately. What makes it particularly good for...
This is the first time I'm hearing about Kagi, and I'm definitely interested, especially because Google and other engines have been feeling so broken lately. What makes it particularly good for programming/code work?
They feature a tool call "lenses" which limit results based on types of responses and sources. It's really insanely useful. I particular like limiting my responses to forums.
They feature a tool call "lenses" which limit results based on types of responses and sources. It's really insanely useful. I particular like limiting my responses to forums.
Probably don’t need to tell you this, but for the others here considering Kagi: In addition to the lenses feature, the fact that you can promote, demote, and even outright pin or block any domain...
Probably don’t need to tell you this, but for the others here considering Kagi: In addition to the lenses feature, the fact that you can promote, demote, and even outright pin or block any domain in your search results is a level of UX you’ll never find in Google Search.
Originally I was going to also, in great detail, list all of the things I like so much about Kagi that make it the best search engine period, plus by extension why it’s worth paying for. However...
Originally I was going to also, in great detail, list all of the things I like so much about Kagi that make it the best search engine period, plus by extension why it’s worth paying for.
However instead, I thought of bringing a different perspective this time.
As a sort of meta-level gauge of how good Kagi’s product is: Their marketing budget is zero.
The entirety of their customer base of more than ≈20,000 has either been convinced to pay from word-of-mouth advertising of existing, content customers, or found the service by first stumbling upon their adjacent offering, the Orion browser (which tangentially, is apparently as good as Kagi Search, but still in macOS only Beta)
It’s been a very long time since I’ve heard of a startup that didn’t need to push marketing at all. Maybe Tesla? I don’t know the background/specifics there.
My pipeline was the typical Google → DuckDuckGo → Kagi one, I’d been trying to de-monopolize in all sorts of areas for a few years at that point… but to be honest, DDG wasn’t the perfect replacement some people make it out to be. Kagi on the other hand? Kind of is. Of course due to the nature of being a paid service, some people can not or will not want to ever afford such a service, but for those who are in the position financially and do like the concept, it is a great way to spend your money.
I also like their approach to “Safe Search”, so if you have kids I’d highly recommend checking out the family plans.
Consider labeling off-topic but I just wanted to share this.
Edit: If you include the family users in the above stats page, it’s probably getting closer to over 25k users too.
Kagi search, at least for me, functions as good, if not better, than I remember Google back in the pre-2010 days. When I search something, and I know what I am looking for, it 90% of the time the...
Kagi search, at least for me, functions as good, if not better, than I remember Google back in the pre-2010 days. When I search something, and I know what I am looking for, it 90% of the time the first link, 9% the first 3, and the remaining 1% is on the first page. I have yet to search for something and not get exactly what I am looking for on the first page (although I only switched within the last month). If I don't know exactly what I am looking, like if I have a programming issue or if I am trying to figure out an error, the results are 50% the first link, 30% the first 3, 10% the first page, and 10% I need to simply reword my question (literally no different than the way I use Google).
I have wanted to move away from Google Search (I already am shifting away from all Google services, but gmail specifically is a task that is taking longer than I want) for years. I tried DuckDuckGo, Brave, Bing (not that it's any better than Google in a privacy sense) and even others like MetaGer. None of them really competed in terms of performance to Google. Then I tried Kagi, and have seen just how good search engines can be (even in this post AI generated content world).
I love it, as I can block, lower, or boost certain domains in my personal search results. I never get stupid Pinterest links flooding my search results anymore, Wikipedia is now one of my highest results most of the time, and the number of times I have come across actual human run blogs has increased tenfold.
Paying for a search engine is still weird to me, but I feel like I am finally getting a more refined product that the industry has let stagnate for at least a decade.
I'm using Kagi for almost half a year at this point, and day-to-day it doesn't feel like anything special, but then once in a while I stumble upon some weird SEO garbage website and remember how...
I'm using Kagi for almost half a year at this point, and day-to-day it doesn't feel like anything special, but then once in a while I stumble upon some weird SEO garbage website and remember how often I used to see those when using Google and DDG.
For example, I remember that "X vs Y" style searches always returned a bunch of annoying websites like https://slant.co that just list random pros and cons of random things and don't help at all but are clearly designed just to abuse SEO. I don't think I ever saw a website like that with Kagi.
Small web is great. I do wish there was a bit less of a focus on tech devops/management/investments stuff on there though. Like it’s not even that it’s tech stuff, it’s that it seems to be that...
Small web is great. I do wish there was a bit less of a focus on tech devops/management/investments stuff on there though. Like it’s not even that it’s tech stuff, it’s that it seems to be that particular flavour of professional business (and maybe startup?) tech over represented. Maybe this is just my bias against finding that interesting though. For me it’s at its best for me when it recovers a travelogue, or some random thoughts on a subject I’m not super familiar with, new perspective on something I am interested in, a hobby or a special interest type site.
edit: it occurs to me that this is also part of why i much prefer tildes to HN
I noticed that, too. I wonder if it's a source bias: setting up and running a "small" website (even if you go the Blogger or Wordpress route) is a disproportionately technical endeavor compared to...
I noticed that, too. I wonder if it's a source bias: setting up and running a "small" website (even if you go the Blogger or Wordpress route) is a disproportionately technical endeavor compared to posting on social media, and so the smallweb is inherently more technical than the Web as a whole.
That’s no doubt part of it, but these days there’s ways to run a blog or small site with little to no technical knowledge and it’s only getting easier. I suspect that a big factor is that...
That’s no doubt part of it, but these days there’s ways to run a blog or small site with little to no technical knowledge and it’s only getting easier.
I suspect that a big factor is that technically inclined people are considerably more likely to care about or see the value in having a small web presence. To the non-technical layman, there’s little to distinguish a page on the small web from a Facebook page or similar, especially if they’re one of the huge number of people who weren’t on the web prior to 2005-2010.
I actually don't mind business content, but even then I do think HN suffers somewhat from "cocktail party" syndrome where people discuss business but at a very high level, without really develing...
I actually don't mind business content, but even then I do think HN suffers somewhat from "cocktail party" syndrome where people discuss business but at a very high level, without really develing into any detail. High-level advice is semi-useful but not as useful as more concrete stuff imo.
I’ve been a user for over a year and a half actually surprised that I stayed. I was on the early adaptor plan where they considered going pay per search or limiting number of searches. I’ll admit,...
I’ve been a user for over a year and a half actually surprised that I stayed. I was on the early adaptor plan where they considered going pay per search or limiting number of searches. I’ll admit, it didn’t sit well with me but I understood the economics of it.
It made me on the fence to say the least keeping track of my search count. Thankfully the numbers worked out and they grew enough to offer plans without them.
their biggest dev and marketing team is their users. I’ll pop into the discord or forums a few times a week to read the backlog and the owner with the rest of the team is constantly engaging others
they have no VC funding, so you’re less likely to get screwed by outside influences vs the team themselves (not that it can’t happen)
there’s a mutual respect in the community. Ideas are received and discussed. Hard truths are spoken with decorum “We can’t do this at this time because we’d be broke in 2 months since we pay per api hit.”
adding new features (I ditched my OpenAI sub) but I’ll admit, they are in danger of scope creep. Thankfully, they’ve been made aware of and are receptive to.
Others have mentioned the features so I won’t repeat them but after almost 2 years, happening upon Google accidentally is painful. And for those of you worried that they’re linking all of your BDSM porn searches, what do you think Google is doing? Except with a shittier experience.
Anyways, give it a try. You may like it or may not want to pay. Everyone has stuff that’s important to them. =)
This is interesting, but I'm seeing exclusively blogs, and I've never been interested in blogs in general. Specific posts? Sure. A whole blog? Nah. I was hoping to find interesting, obscure...
This is interesting, but I'm seeing exclusively blogs, and I've never been interested in blogs in general. Specific posts? Sure. A whole blog? Nah. I was hoping to find interesting, obscure non-blog sites. To me, it feels like the title should be "i.e. personal blogs."
Amazing, instantly reminds me of StumbleUpon back in the day, which is how I eventually discovered Reddit.
StumbleUpon was amazing, I loved it. I knew a load of very non-tech-savvy people that used it as well and remember having some really good conversations about the cool stuff they found with it. This is a wonderful spiritual successor!
Same story here, god I miss StumbleUpon so much. Back then RSS was the way to go and I had so many webcomics I'd found via SU set up in my RSS reader sidebar in Firefox.
To scratch that StumbleUpon itch even further, may I suggest CloudHiker
Same here, I loved StumbeUpon but eventually got jaded by all listicle blogs, blogspam (even without AI there was already a lot zombified content on the web back then) and lack of interaction with people on it. I sometimes tried to engage on comments under some blogs but you can guess how well that went.
So when I came across reddit it was really awesome to discover actual discourse and (somewhat) effective user curated content.
I still can't believe I pay for a search engine, but every time I try a search using anything else it all falls so flat. Especially if you do any sort of programming or code work.
Hi, I`m quite curious, (if its ok for you to answer), how Kagi search works with copyright-grey material and adult stuff? Like torrents, different NSFW, etc. etc?
They obviously know who you are. Their assertion is that they don't record searches in a way that they can be linked back to the account that initiated them. Sort of necessarily, we only have their word for that, but they don't really have an incentive to do otherwise and the risk to their business and reputation if they did and got caught would be existential.
Isn't their claim more "we promise that we don't track you/link searches to your account, but we understand that you have to trust on that and might not, so for extra security you can also pay anonymously so that it's harder for us to work out who you are"?
There are numerous ways to operate an application (such as a search engine) in a way that authenticates you as simply a paying or non-paying user, then generate a key that pretty much anonymously says, "Hey, I am a paying user, here's my key," without knowing who the person with the key would be specifically. I am overly simplifying the actual implementation, but the idea is still the same.
Also, would you prefer trusting a company that claims to not store this information ever, or a company that makes all of its money by literally storing as much of that information as possible?
You can simply make sure never to couple the two distinct pieces of information. You have to trust that they are doing it right, but it's easy.
Years ago I wrote voting software and it was easy. When you login and vote, we recorded that you voted. That's it. The part that actually records the vote is separate.
Similarly, they can allow you to login, record that you searched so they can keep count, but not actually record your search. But yes, you have to trust that the information is kept separate.
Depending on how the key is generated and given out, it can easily be anonymized in a way that is not linked to an existing user. Once the user signs in, a token could be generated by a hash based on Date/time/username/etc. and then easily have been stored in database table that simply stores: TOKEN, PAY_STATUS, SEARCHES_LEFT, and EXPIRATION_DATE, without linking that token/key back to the user. When a user does a search, check if the provided token is in the table and not yet expired, and then provide results if so. Otherwise, require authentication again.
I'm not saying it is done this way in Kagi's case, I'm just saying requiring a sign in does not mean that once your authenticated you're not anonymized. Anonymous authentication has been a thing for a long time.
I agree, thought, it is better to be on the safe side and assume everything is linked to you. I am personally operating under the assumption that it would make less financial sense risking their privacy oriented focus by storing information that, if ever used or found out about, would completely destroy the driving force of its users.
EDIT: forgot to add the SEARCHES_LEFT column name to my example
Works just fine for NSFW content
This is the first time I'm hearing about Kagi, and I'm definitely interested, especially because Google and other engines have been feeling so broken lately. What makes it particularly good for programming/code work?
They feature a tool call "lenses" which limit results based on types of responses and sources. It's really insanely useful. I particular like limiting my responses to forums.
Probably don’t need to tell you this, but for the others here considering Kagi: In addition to the lenses feature, the fact that you can promote, demote, and even outright pin or block any domain in your search results is a level of UX you’ll never find in Google Search.
@gpl you will definitely notice the difference in how much easier it is to find real content that is not centered around affiliate links
Originally I was going to also, in great detail, list all of the things I like so much about Kagi that make it the best search engine period, plus by extension why it’s worth paying for.
However instead, I thought of bringing a different perspective this time.
As a sort of meta-level gauge of how good Kagi’s product is: Their marketing budget is zero.
The entirety of their customer base of more than ≈20,000 has either been convinced to pay from word-of-mouth advertising of existing, content customers, or found the service by first stumbling upon their adjacent offering, the Orion browser (which tangentially, is apparently as good as Kagi Search, but still in macOS only Beta)
It’s been a very long time since I’ve heard of a startup that didn’t need to push marketing at all. Maybe Tesla? I don’t know the background/specifics there.
My pipeline was the typical Google → DuckDuckGo → Kagi one, I’d been trying to de-monopolize in all sorts of areas for a few years at that point… but to be honest, DDG wasn’t the perfect replacement some people make it out to be. Kagi on the other hand? Kind of is. Of course due to the nature of being a paid service, some people can not or will not want to ever afford such a service, but for those who are in the position financially and do like the concept, it is a great way to spend your money.
I also like their approach to “Safe Search”, so if you have kids I’d highly recommend checking out the family plans.
Consider labeling off-topic but I just wanted to share this.
Edit: If you include the family users in the above stats page, it’s probably getting closer to over 25k users too.
Kagi search, at least for me, functions as good, if not better, than I remember Google back in the pre-2010 days. When I search something, and I know what I am looking for, it 90% of the time the first link, 9% the first 3, and the remaining 1% is on the first page. I have yet to search for something and not get exactly what I am looking for on the first page (although I only switched within the last month). If I don't know exactly what I am looking, like if I have a programming issue or if I am trying to figure out an error, the results are 50% the first link, 30% the first 3, 10% the first page, and 10% I need to simply reword my question (literally no different than the way I use Google).
I have wanted to move away from Google Search (I already am shifting away from all Google services, but gmail specifically is a task that is taking longer than I want) for years. I tried DuckDuckGo, Brave, Bing (not that it's any better than Google in a privacy sense) and even others like MetaGer. None of them really competed in terms of performance to Google. Then I tried Kagi, and have seen just how good search engines can be (even in this post AI generated content world).
I love it, as I can block, lower, or boost certain domains in my personal search results. I never get stupid Pinterest links flooding my search results anymore, Wikipedia is now one of my highest results most of the time, and the number of times I have come across actual human run blogs has increased tenfold.
Paying for a search engine is still weird to me, but I feel like I am finally getting a more refined product that the industry has let stagnate for at least a decade.
I'm also a Kagi user. I have no problems with DDG's results (no idea what Goog does nowadays), but I like the principle of paying for stuff.
I'm using Kagi for almost half a year at this point, and day-to-day it doesn't feel like anything special, but then once in a while I stumble upon some weird SEO garbage website and remember how often I used to see those when using Google and DDG.
For example, I remember that "X vs Y" style searches always returned a bunch of annoying websites like https://slant.co that just list random pros and cons of random things and don't help at all but are clearly designed just to abuse SEO. I don't think I ever saw a website like that with Kagi.
Nice! I just signed up for Kagi yesterday and have been liking the more powerful search tools. I'll have to check these out.
Small web is great. I do wish there was a bit less of a focus on tech devops/management/investments stuff on there though. Like it’s not even that it’s tech stuff, it’s that it seems to be that particular flavour of professional business (and maybe startup?) tech over represented. Maybe this is just my bias against finding that interesting though. For me it’s at its best for me when it recovers a travelogue, or some random thoughts on a subject I’m not super familiar with, new perspective on something I am interested in, a hobby or a special interest type site.
edit: it occurs to me that this is also part of why i much prefer tildes to HN
I noticed that, too. I wonder if it's a source bias: setting up and running a "small" website (even if you go the Blogger or Wordpress route) is a disproportionately technical endeavor compared to posting on social media, and so the smallweb is inherently more technical than the Web as a whole.
That’s no doubt part of it, but these days there’s ways to run a blog or small site with little to no technical knowledge and it’s only getting easier.
I suspect that a big factor is that technically inclined people are considerably more likely to care about or see the value in having a small web presence. To the non-technical layman, there’s little to distinguish a page on the small web from a Facebook page or similar, especially if they’re one of the huge number of people who weren’t on the web prior to 2005-2010.
I actually don't mind business content, but even then I do think HN suffers somewhat from "cocktail party" syndrome where people discuss business but at a very high level, without really develing into any detail. High-level advice is semi-useful but not as useful as more concrete stuff imo.
I’ve been a user for over a year and a half actually surprised that I stayed. I was on the early adaptor plan where they considered going pay per search or limiting number of searches. I’ll admit, it didn’t sit well with me but I understood the economics of it.
It made me on the fence to say the least keeping track of my search count. Thankfully the numbers worked out and they grew enough to offer plans without them.
Others have mentioned the features so I won’t repeat them but after almost 2 years, happening upon Google accidentally is painful. And for those of you worried that they’re linking all of your BDSM porn searches, what do you think Google is doing? Except with a shittier experience.
Anyways, give it a try. You may like it or may not want to pay. Everyone has stuff that’s important to them. =)
This is interesting, but I'm seeing exclusively blogs, and I've never been interested in blogs in general. Specific posts? Sure. A whole blog? Nah. I was hoping to find interesting, obscure non-blog sites. To me, it feels like the title should be "i.e. personal blogs."