I remember when the search engine space was full of cool competitors. AltaVista, Yahoo, GoTo, Dogpile, Northern Light, Ask Jeeves, Lycos, Excite, WebCrawler.
Now we have Google whose value comes with their incessant privacy abuse, and Bing/Bing front ends (DDG). Cool and exciting are definitely no longer the emotions I associate with web search.
I've never used their normal search engine, but their image search blows Google out of the water.
I highly encourage anyone who frequently uses Google images to try out Yandex.
What's really interesting is the differences in the results -- it seems like Yandex is actually filtering for high quality art. I've got a project on the back burner to scrape google and Yandex for the same queries and then do different sorts of analysis on them and see if anything interesting pops up.
Going by eye, most photo results on Yandex are higher resolution, and more saturated. It also doesn't pull images from video thumbnails like Google does.
seconded. it's stellar, even better than what google used to be. feed it a cropped frame from a japanese cartoon and it returns links to illegitimate websites containing the full episode. google returns a nonsensical ML suggestion and no results
My favorite part about Yandex is that there's no DMCA. The amount of DMCA notices against search engines like Google and Bing is absolutely astonishing. Thousands of them are filed every day and a massive amount of webpages are delisted because of these complaints. You can look up some of them in the LumenDatabase [1].
Yandex might be good in Russian (I don't know) but it's barely mediocre in English and really bad at other less common languages.
The only good thing about it was that it used to give unfiltered results for some queries that google had shadowbanned (such as for piratebay or scihub). However as far as I can tell thats no longer always the case.
The search engine space is full of cool competitors today. Google, Bing, Yandex, DuckDuckGo, Kagi, Neeva, Brave, Startpage, Alexandria, Right DAO, Wiby, Marginalia, Teclis.
Certainly, a few of those are Bing/Google frontends, but many more are independent or weighted mixes of commercial and noncommercial indexes. I really like Kagi, Wiby, Marginalia (Search), and Teclis to name a few.
I kinda wish there was more of a push to investigate what search can be than just copying what Google is doing. In terms of development, I feel the "genre" has stagnated after very few honest stabs at the problem.
I do think it's a lot harder to beat one of the largest tech companies in the world at their own game than carving out some niche for yourself not just in what is indexed but what you are doing (and then maybe taking over the world once that foothold has been firmly established).
Some honest stabs in the past have been crushed, which has contributed to stagnation. You also don't have to "beat" them, to make a difference and contribution. Not everyone aspires to take over the world. Cooperation can help us; happy to explore with anyone.
Yeah I largely agree. I don't think the aim should be to replace Google. A large part of the problem with our information access right now is that we have a google, when we could probably benefit from more disparate services for different needs.
Right now Google is tackling website discovery, fact retrieval, shopping, "what's the address for that site I sometimes visit", and so on, through the same search bar, mostly by guessing what you want based on your search terms. Like a swiss army knife, which is capable of many things but not great at any particular thing, I think this is in the end incredibly limiting. If two people search for "coffee", one of them wanting to find a place to buy it, and the next looking for information about the beverage; there can't be one best result. You get this weird dance where the search engine attempts to read your mind about what you want, and you're attempting to read its mind about which terms it expects to show you what you want. Very frustrating.
You can probably take any of these niches and make a superior product, something that isn't constrained to the search bar format.
Right now I'm really sort of down a rabbit hole of finding new formats for discovering and exploring websites. Previously I've built <https://search.marginalia.nu/explore/random>, which led to the logical next step of <https://explore.marginalia.nu/>. They're just frivolous toys, but the more I dabble in this space the more I think there are viable alternatives to search bars and social bookmarking sites like Reddit and HN that deserve serious exploration.
I’ve seen you post links to the mojeek blog on HN a few times now.
I have never used mojeek for the simple fact that there’s no link to the app from your blog.
Having to guess (or google) what the actual url for the app is is a bit too much friction for me. Especially when I do most of my HN browsing on mobile.
Kagi is really shaping up to be a great search experience. The fact that it’s going to be a paid service is a killer feature. No ads is incredibly refreshing.
You sound like the type to also disable cookies and use an adblocker and not use Chrome, but there are many ways Google can collect your credit cards or associate your purchase history. From Google Analytics to remembering the card in Chrome and syncing Chrome browser history to having a Gmail record of your purchases or even adding a credit card to your phone to use with Google Pay on an Android device… or while we’re at it, using Android. Lots of spying there unless you run Blokada Proxy or equivalent. Not to mention, if cookies are disabled, but you click on a Google ad, Google will likely help track the purchase with the third-party advertiser’s site using some alternative identifier they could assign outside of a cookie. After all, if you control the URL, you can inject session tokens. Not saying they do, but they could, and they probably make the data available in Google Analytics somehow, even with cookies disabled.
I realized the other day that thanks to Microsoft 365 and Google Workspace, these big companies know a ton about you just because your employer uses them. For example you might not be signed in using that web browser, but your work account just leaked your IP address or device metadata to Google, so they know it’s you anyway. It’s almost impossible to stay completely private online, these days, and still participate in the economy or society.
I would, though, draw a line and say that just because it’s possible to trace doesn’t mean it should be allowed… but we’ve all seen how the most effective countermeasures isn’t legislation but instead clients that can outsmart the trackers - like adblockers, privacy VPN proxies, DNS black holes, and Apple’s iCloud Private Relay (or similar VPNs) which would anonymize your IP address, partially defeating the point made earlier. (I say partially because it’s not like every packet is going to get a new IP address, that would break TCP connections… they could associate your work profile with your anonymous traffic using the same relay IP, if both traffic sources hit the same Google server from the same device, say…)
And of course we have historically shown that almost any collection of data can reveal a lot about a person, from Netflix watch history to search terms. I’m ignoring the obvious - that searching Google gives it information about you with every search.
- Google cannot legally protect payment information in any 1st, 2nd, or 3rd world country without explicit permission. There isn't a single country they operate in that will let them 'scrape' payment information in the manner which you claim.
- Any decent adblocker blocks GA cookies. GA does NOT collect payment info. Period. If you actually bothered to read that legally binding language or if you sniffed what is passed to GA you would know that. Please head back to your conspiracy subreddit and try again.
- Many folks like myself have completely left the Google ecosystem. Our adblockers block Google. We don't use google.com or Youtube. When you hit a popular Google property and notice an uptick in ads that is why. I use duck duck go as a search engine, and I personally dislike watching videos, however if you are someone who likes videos, I hear TikTok is all the rage these days.
I need to be clear here. I am not saying that it is impossible to track you, but rather, it isn't realistic or scalable across billions of users. For starters, Google would have to figure out how to beat Firefox containers or uBlock origin, and that is the simplest of problems. Next, they have to figure out how to bypass your local hosts file for a long period of time, otherwise they will get piholed...
If you use Chrome and opt out of sharing data with Google, they can still pair the ID from origin trials with your IP address and have a very good idea of who you are - or what installation of Chrome you’re using, which is basically the same thing: https://www.google.com/chrome/privacy/whitepaper.html#variat...
If you anonymize your IP, and turn off tracking, then yes, not much to say there until they start fingerprinting the fonts you have installed or something…
I don't understand why Kagi is appealing, tbh. It just gives you Google and Bing results. What type of censorship does Kagi really apply? It wasn't until just recently that DDG admitted to censorship. Not sure if I can trust Kagi's promises of privacy, etc since they don't open source it...the same problem I have with DDG.
It's worth paying for (maybe not at the rates they're planning to charge sadly) just for the fact that you can click a button and never see W3Schools results in your searches ever again. Or any other spammy domain that comes up top on every other search engine.
There are Chrome extensions to remove the ones you don't want so not sure why I'd pay a premium to remove them. w3schools is the first result when I search for "php loop" so Kagi seems to be full of the same spam you are trying to escape:
I'm not sure if you're deliberately missing the point or not, but a) A Chrome extension to remove all the spam means an empty front page of search results on Google which isn't very useful, and b) I literally just told you that there's a button you can click in Kagi to never see those again - pointing out that they appear in the search results for someone who hasn't clicked that button isn't quite the win that you seem to think it is.
We're talking about Kagi, not Google. I edited my answer yesterday shortly after posting so that might be the confusion. Why would I pay a premium for a search engine that returns a full page of spam?
There's a Chrome extension that removes spam results from Kagi? Are you sure?
And again, I'm not sure whether you're just pretending to not understand what I'm saying, but I've already explained (twice) that Kagi lets you choose which spammy results you don't want to see, so "Why would I pay a premium for a search engine that returns a full page of spam?" appears to be answering someone else's post entirely. The whole point of what I'm saying is that you don't get a page full of spam, which you do get on every single one of the big, free search engines. I think I've run out of different ways to explain the same thing now so I'm going to bow out here.
I've been trying brave as of recent. It feels like browsing the internet of the 90s; back when you search engines returned small no-name websites and blogs. Google gives me the same list of 50 or so "preferred" websites full of click bait run by media companies no matter what I search.
Somewhat related, but I remember Ask Jeeves as being my first exposure to search engines. I had just gotten internet on my home computer and one of my parents friends came over and excitedly told me about “this website that will give you the answer to anything you ask”. I didn’t even know what to search for (I think I ended up searching “tallest woman in the world”), the idea of being able to quickly look something up and quickly get references was pretty foreign to me and now it’s hard to imagine life without the ability to do so.
Edit: whenever I see the Jenkins logo, I always think of the Ask Jeeves logo from back then
Now we have Google whose value comes with their incessant privacy abuse, and Bing/Bing front ends (DDG). Cool and exciting are definitely no longer the emotions I associate with web search.