Qwant, a European search engine that respects your privacy | Hacker News Reader

Qwant, a European search engine that respects your privacy

about.qwant.com

214 pointskome121 comments

Privacy loving European here.

I want to like it, but it gives off a bad vibe.

First of all it's too loaded and it doesn't seem to work at all without Javascript enabled. Give me a text field, I don't care about the news.

Second, no easy way to add it in my Firefox search engine list. This is really something obvious they should fix, come on.

Third, they want you to install an extension that will indeed set qwant as default search engine, but also seems to try to shove down your throat some sort of bookmark and "board" (their own thing, pinterest like?) management which requires login to their service.

http://img.nux.ro/q9M-Menu_292.png

IMHO it doesn't really live up to the expectations and it seems more like a social media experiment attempt.

If we forego the whole privacy thing which I think they are only using for marketing here, then the search results don't seem too bad (in my 2 minutes of testing) and the boards thing is interesting, but nah.

> First of all it's too loaded and it doesn't seem to work at all without Javascript enabled. Give me a text field, I don't care about the news.

I genuinely don't understand this hostile statement. Why do you complain that a website isn't functional in pure html in 2017? do you want the web to look like it's 1998?

Why is it a problem to have page functional without Javascript? It's the number one tools for spying on users, so especially if you're a privacy focused search engine, you should except people to have Javascript disabled, until they choose to trust you.

There's no real need to javascript on search engines or news sites. There might be some additional features, like polls that will require Javascript, but the base functionality should work.

I'm still a firm believer in progressive enhancement, even if it's no longer trendy. Modern Javascript framework however seems to be firmly in the "Javascript or die" camp.

Because a search engine must run on the widest array of web browsers possible. For being a European search engine and based in France to boot, they should very well be aware of platforms which have minimal or no JavaScript support in their web browsers, like for example the classic AmigaOS. If someone with a platform like that tries to use their search engine to perhaps find a JavaScript enabled client, they couldn’t. Lots of TV sets nowadays have built-in web browsers with no or incomplete JavaScript support, as another example.

However the worst offense is that JavaScript is utterly unnecessary in a web search application, so that would be introducing an artificial dependency, one of the worst crimes in software development. Software should be designed with minimal dependencies it needs to do the job and the extras should be the users’ choice. Developers who made this choice for users historically lost their user base as soon as a competing application which had less dependencies showed up; there is a lesson to be learned from that.

> For being a European search engine and based in France to boot, they should very well be aware of platforms which have minimal or no JavaScript support in their web browsers,

I don't understand this statement. Is there less js-enabled browser in Europe, or in France?

Commodore Amiga and ATARI ST were big in France (and still are, if the demoscene contributions are anything to go by); In Europe, Amiga is still a thing. ATARI ST is still a thing. Neither of those have complete JavaScript support, if any. That’s where “the French connection” comes from in this context.

You are conflating two different things. It is indeed completely possible to have nice looking websites without javascript. Heavy use of javascript is often a huge waste of resources anyway.

The question is only; how much is too much. And considering your competition is Google, that answer is: not a lot.

Not parent, but Javascript enables a lot of nasty business and often slows pages down on my not-so-fast work PC, so I disable it. If a website has a good reason to need it (e.g. apps like Google Docs), it gets whitelisted. Typically, one doesn't need a script for displaying text and filling forms.

Let's reverse the question: the task "map this text input to a list of links" could be done in 1998 with no scripts, why is it dependent on JS in 2017?

> I genuinely don't understand this hostile statement. Why do you complain that a website isn't functional in pure html in 2017?

Because sites that rely on javascript typically have terrible performance and broken UI, and add insult to injury by not doing anything that actually needed javascript in the first place.

The https://lite.qwant.com/ seems to work without Javascript.

>First of all it's too loaded and it doesn't seem to work at all without Javascript enabled. with noscript turned on, quant.com redirects me to lite.quant.com

>Second, no easy way to add it in my Firefox search engine list. This is really something obvious they should fix, come on. I see a green plus in the corner of the magnifying glass icon in the search box and when you click it, at the bottom of the search engines I have installed I see "Add quant."

Thanks, "lite" should be the default really. Much better.

I don't see that green plus you are talking about, I stared and stared.

Here's a screencap on my end: https://imgur.com/9ZgNh4Y

Thanks, I was blind, but now I see.

Agree. That's the reason why Google has won the minds and hearts during the search engine wars: no bloat.

Right now I'm reading the frontpage and getting distracted from news with political opinions to which I don't even agree. Basically getting in the way of engineering.

> Second, no easy way to add it in my Firefox search engine list. This is really something obvious they should fix, come on.

That's weird, on my side I have the "Add Qwant" option show up in my firefox search field. I'm on Firefox 56.0.2 Win7.

I see others already mentioned the lite version. For a quickly loading page with minimal extras that just let's you search you can try https://ixquick.com.

It integrates nicely into Firefox. I've found results to be pretty good and relevant when I use it. Found it before DDG and now I use both depending on what machine I'm on.

Anyway, they claim they are "the world's most private search engine".

Adding to that: they advertise the app for android. The app wants access to contacts, identity, all available networks (wifi scans) and a bunch of other stuff, where I ask myself why it is needed. I also want(ed) to like it, but you are exactly right about that bad vibe. We finally need a search engine that just costs a monthly fee or gets paid via coinhive or the like....

I think it is great news. Europe is so much behind the US in the IT-Sector. Even russia has a successful search-engine.

The default view is a bit overloaded in my opinion...

Lite version: https://lite.qwant.com/

A plus point is that qwant also supports hashbangs like duckduckgo.

looks promising, qwanting 'java concurrency' gives a nice set of relevant results

I hope that one day when Qwant is big, this is cited as the first use of 'qwanting'

Europe is so much behind the US in the IT-Sector.

Isn’t that a sad truth, only made worse by the irony that a lot of us went to work in the States. A lot of Europe is still conceptually stuck in the pre-UNIX, lone desktop Windows/MS-DOS, early ’90’s era, while the rest of the world runs them by with macOS PC’s and iPads. Lots of European UNIX talent left for the States right at the cusp of the dot-com era and helped build the Internet because that’s where all the action was. We never really recovered from that.

I am based in Copenhagen and can see some innovation at the application level but there is a lot of catch-up as well, mostly focused at the nordics internal market.

For lower levels of the stack (e.g.: IaaS), there is no way to be competitive against SV without subsidy b/c fwcit there is no access to capital nor enough talents to this problem size.

The prospective is not getting any better at the immigration level, the greencard schema has been repelled and companies are having even harder time to attract talent.

"A lot of Europe is still conceptually stuck in the pre-UNIX, lone desktop Windows/MS-DOS, early ’90’s era" So Europe is stuck in the 90's which somehow predates the 70's ??

"while the rest of the world runs them by with macOS PC’s and iPads" Buying istuff is somehow progressive?

It's not the buying hardware thing that's important:

> conceptually stuck in the pre-UNIX, lone desktop Windows/MS-DOS

It's about understanding that quality matters, that UX is a thing, that design is more than just a theme, that machines are networked to varying degrees and frequency, with varying performance profiles, form factors and interaction models, and that this field is different from fordist industrial endeavours and you'd better read the Mythical Man Month if you want to have any chance at a viable let alone competitive RTT instead of treating your coding engineers as crappy brick layers.

You betcha: that stuff runs UNIX, doesn’t disrupt one’s concentration with modal windows and runs circles around Windows desktops performance-wise. For most people an iPad and web applications are apparently enough, which means they don’t need or want a Windows desktop PC any more.

The early ‘90’s computing paradigm with a lone desktop PC user and application isn’t cutting it any more. At the very least, one good thing came out of change.

"that stuff runs UNIX" osx and ios have an very old UNIX legacy but as they currently stand are extremely restrictive walled gardens completey controlled by a single giant avaricious corporation. If you were true to your ideas you'd be running Linux which is open, open source and btw is most popular in Europe, eg. being the OS used by the public sector in Germany. "doesn’t disrupt one’s concentration with modal windows" What? please give an example. Linux's UI can be configured however u want. "and runs circles around Windows desktops performance-wise" That is just nonsensical. To talk about performance you need to be comparing 2 specific things. Which things are you talking about? As an example, heres two things I'm comparing; For fun once at work we compared benchmarks between two new laptops each costing about €1400. My colleague had a macbook pro, I had a msi gs60. My laptop scored about 3 times as much in cpu performance and 5 times as much in a unigine valley graphics becnchmark. "The early ‘90’s computing paradigm with a lone desktop PC user and application isn’t cutting it any more" Sure things are going that way, but if you think most users are ready to do everything via web-apps now then u are living in a very rarefied bubble. photoshop, autocad, 3dsmax, ableton, an endless list of specialised, high performance demand, software for professionals, theres a long way to go before that kind of thing can run in a browser. Also how is iOS os OSX somehow more web-app friendly? Youre much more restricted on them on what browser you can run and how u can configure it than on linux or even windows. html5 canvas doesnt even work properly on iOS.

If you were true to your ideas you'd be running Linux

I would rather drop dead. I’m a UNIX guy, Solaris guy to be precise and that means illumos and SmartOS in particular. I value stability, formal specifications, the ability to introspect the system, a system which can self heal and is paranoid about correctness of operation and data. A system which is always engineered to be backwards compatible. Exactly everything that SmartOS is and GNU/Linux isn’t! illumos and SmartOS are using CDDL which is by and large less restrictive than the GPL, so GNU/Linux stands for everything I fight against. There are no words in any of the languages I speak which can express my hate of GNU and GNU/Linux.

As for macOS and iOS, I don’t care about open because I neither intend nor want to tinker with it: I want to pay, bring home, turn on and use. As long as all the hardware works, the thing is responsive and I have xterm and ssh to log into my SmartOS datacenter, I’m good to go. Not going to waste my life with GNU/Linux, I do that too much at work aleady. Simply not going to happen.

What I mean by responsiveness is that my 2013 MacBook Air is usable instantaneously while my barely a year old work laptop with Windows 7 needs half an hour to get to the point where it is able to launch an application. Then I launch it, but since it has other apps starting in the background as I’m working they keep popping up their windows and taking the focus away, interrupting my thought and workflow. A macOS app will never appropriate the focus, nor will subwindows be modal, forcing me to close them first if I want to get at the data in the main window. Windows is just crap.

European here.

First of all, this is okay. But not good.

I've tried several "slang" searches that google gets just fine. I've tried these because ddg had/has problems with them for a long time.

E.g. If I google for "poe shavs" (a slang term for an item with an abbreviation of the game name), I expect https://pathofexile.gamepedia.com/Shavronne%27s_Wrappings

from what I've tried this seems to work even without a search history on google. I've tried this across different cookieless browsers and IP's. Google just seems to get it.

This seems to return (almost) the exact same shit as bing on Qwant. So I do not see how this is an improvement over bing/ddg/whatever.

Man I'd love to get rid of google in my personal life, but... for my searches nobody else seems to perform. Google just seems to know what I want to search, even if I don't have a search history on that machine.

Edit: For searches in german it seems to perform similarily shitty to bing. Do they get search results from bing?

I feel like Qwant works much better in french than english.

I think qwant is a company from france.

The thing is, Google improves its results THANKS to its privacy invasion.

To understand "poe" as path of exile, it helps to know that you at least once searched for path of exile. After it has seen a few hundreds users do that, it will make the connection.

To know that "poe shavs" refers to the game and not Edgar Alan Poe, it probably had to provide a link to a relevant author's name and to the game's page and watch which links people would more likely click.

There is a chicken and egg here. People want a search engine that can read their minds without invading their privacy, that's a hard problem.

If users search for "poe" and can't find anything, some of them will search for "path of exile" next. You'd probably only need to look at request k and request k+1 to figure it out eventually.

Also just seems like "bat cave" (animal) vs "bat cage" (sport). poe + shavs forms its own context.

Google is really good at that. I can't use DDG for programming because DDG thinks I want to learn more about Elm trees.

This privacy invasion is a two edged sword too. Google is getting very good at searching for stuff you use daily, especially if it's gaming or some other media in fashion.

However, it's becoming useless for infrequent searches. Try googling any random error message and notice 90% of the results don't even contain it.

Can you give an example of a random error message where the results do not match?

Try to compile some software, if you get a compile error paste the generic portion into Google and watch it become completely useless.

Thanks, but can you please give a specific example?

One, I do not share this experience, and think the quality issue is overstated (you are more likely to remember the time that a query failed, than when it succeeded).

Two, if you can produce a POC, Google can use this to improve the search results.

As is, the issue brought up by original poster is too vague and unspecified to be of any use.

Dude, you're treating it like a bug. It's not. It's intentional behaviour that started 2-3 years ago. Instead of giving you just the formerly relevant results, they give you all kinds of crap that they think you are interested in.

You can still revert to the original behaviour by selecting 'Verbatim' from the search tools menu, but you cannot make it default. And in time I'm sure some marketing head will remove even that option.

Edit: and if you never experience it, my guess is you're working with some technology that's "in fashion".

Maybe I never experienced it, because if I want to force exact match results I simply place the error message within double quotes. But often you do not need exact match searches to get good results for a random error message. If it happens to you a lot, it should not be a problem to post a single example query that returns useless results.

I do agree that Google is focusing more and more on the common internet user, and not the early tech adopters. This forces us to use tricks like the double quotes, while keeping the search engine user-friendly for the vast majority of ad-clicking internetters.

The behavior you are referring to is called https://en.m.wikipedia.org/wiki/Query_expansion and while this improves the search results for many people with imprecise or misspelled queries, if you trained yourself to search with exact matches, you'll need some time to adapt (or get in the habbit of adding double quotes).

Maybe I never experienced it, because if I want to force exact match results I simply place the error message within double quotes.

Come on!!! This is not my first rodeo. I’ve been on the InterNet (yes, with a capital N) since 1993 and I use double quote searching all the time, except that in cases like these one gets zero results back.

It is not relevant that you were on the internet before Google even existed.

Apparently you are the only person in the universe that regularly gets never-before-seen error messages. I feel for you, but I hope you stay away from any issue tracker I am involved in, because no matter your vast experience, the quality of your error reporting is downright poor.

If exact match search is unable to find your error message, then you wouldn't have found it on 2005 Google either. You wouldn't find it on any other search engine. You are (poorly and vaguely) describing a problem that must have always been there and blaming it on an unrelated recent UI change.

All the while unable/unwilling to give a single concrete example, just noisy ranting. For all I know you are, despite your experience, banging your head against the keyboard, until you get 0 results. The burden is on you to show that there is a teapot orbiting the Sun. Good luck!

Very interesting. We tell you that google's invasion of privacy is polluting search results, and you keep telling us there are ways around it.

Why should we have to work around it?

It is not relevant that you were on the internet before Google even existed.

That’s where you’re wrong: I’ve used the double quotation marks search technique since before Google and I’ve known about it and used it since Google’s debut.

I already told you I’m not compiling anything and can’t give you a reproducible test case right now but you chose to disregard that; I’ve also told you what to do to reproduce the problem yourself (“attempt to compile GCC, get an error, search with Google with and without double quotation marks”), but you don’t want to do that because it’s a lot of work, I know, but that’s your problem and here’s why:

The burden is on you to show that there is a teapot orbiting the Sun.

that is why, since you’re wrong again: I’ve switched to “DuckDuckGo” as my primary search engine and rarely use Google any more since the results are nothing but advertising-soiled false positives; I don’t care whether you do something about it or not. Now, you might wisen up and take my feedback about exact or partial error searching earnestly or not. You wanted feedback, you got it; your move on what to do about it. Good luck.

> First of all, this is okay. But not good.

They are going through aggressive growth phase right now and could need a hand fix many things. As you mentioned (the NLP problem):

https://about.qwant.com/job/ml-engineer-specialized-natural-...

Unfortunately (according to Wikipedia) like DuckDuckGo, also Qwant is mainly a meta search engine, meaning the are using Bing & Yandex and have only a small scale own web crawler. https://en.m.wikipedia.org/wiki/Qwant#Criticism

Sponsored by European Union, France and a Germany and dubbed as Google clone a few years ago is "Exalead", a real large scale web search engine. Unfortunately it got bought by Daussault (Catia 3D CAD), but it's still online and works:

https://www.exalead.com/search/

I'm not seeing any advantage over duckduckgo.com.

I want a search engine that loads quickly and gives good results, DDG does that while protecting my privacy.

Do you remember DDG five years ago? Everyone needs to start somewhere.

One potential advantage is it's run out of Europe instead of the US, which is significant for certain people.

Exactly. I still use !g with DDG if I can't find something; but 1/3 of the time DDG works just fine.

Back in the day you'd go to Lycos, Hotbot, Yahoo, et. al. and you'd get different results! There was a difference with search engines that would leads you to all kinds of interesting stuff. Today it's all monolithic. It's going to be either indexed by Google or on some mega walled garden like FB/Twitter/Tumblr/etc.

We need more alternative search engines. I personally stopped using the verb "Googled" and try to say "Searched for" instead. The world needs to be more than just Google.

I'd love if DDG would provide a European version. A subdomain that redirects to data only hosted in the EU. Not sure how hard that would be to accomplish but I'm sure it'd be a good argument for many to start using it.

some competition, that's an advantage ;)

Sadly, the search engine reports many duplicate results. For example, I searched: "arid land reforestation", and results were (https://screenshots.firefox.com/wl6JuVa5sbWNgy2o/www.qwant.c...):

* 1st: http://haringey.store/reforestation/in/reforestation_in_arid...

* 2nd: http://haringey.store/reforestation/in/reforestation_in_arid...

* 4th: http://maestron.store/reforestation/in/reforestation_in_arid...

* 5th: http://pipcoins.store/reforestation/in/reforestation_in_arid...

etc, etc. All of which were "bad" websites. It's a shame because some good results (science direct, …) were inside the lot. The content isn't even a pdf.

You should implement a "unique" webpage result fingerprint in order to avoid showing duplicates (well, I suppose it's easier said than done..).

60 requests, Loaded in 6.49s. Cripes, if you don't have some marketing department shoving tracking scripts all over the place against your will what's your excuse?

I got a question from our marketing department why it takes a while for our website to load just the other day. My answer was exactly what you mentioned - forcing us to include tracking scripts from google analytics, tag manager, oracle maximizer, segment, facebook, doubleclick and a handful of other monstrosities that do god only knows what on startup.

Prediction: no change will take place because someone else will say 'without that data, we're flying blind.'

Definitely a problem. I had 10.24s on first load, which is about 9 seconds longer after google/duckduckgo becomes usable.

Because DDG and Google search don't show dozens of media files on its search page. They really need to default to a simple search page, and then they can give users the option to use the current richer theme. This product reminds me of the one developed by Wolframalpha a couple years ago with a fancy interface.

You can try the lite version that should be the default version IMHO: https://lite.qwant.com

I actually like its theme stylistically more than I like DDG.

A couple of points of feedback though:

- When I go through the lite version (which should almost be the default), if I image search, clicking on the results takes me to the webpage that the image is on, not the image itself. If i wanted that I wouldn't be doing an image search.

- There's an "install qwant" button in the top right hand of the screen. I'm not clear what i'd be installing if i clicked on that, or why i should need or want to install anything to use a search engine?

- I've gotten two separate front-pages when visiting www.qwant.com on desktop. One is a semi-minimalist page, a bit more complicated than the lite version, but quite nice and clean. If this was the default i'd be happy. The second was a version more reminiscent of the old "yahoo" type setup, with news and stories and random stuff splattered everywhere below the search bar. Subjectively, the second one needs to die. However, i'm not sure why its no longer turning up for me (even though I'm glad it isn't). I certainly haven't saved my account, nor am i consciously using a remembered URL.

- I got a captcha type thing popping up saying there'd been a lot of activity from my location. I'm pretty sure I'm not currently behind my VPN, so that seemed weird for me to be seeing such...

- On mobile, that just took WAY too long to open. Almost 10 seconds. As others have mentioned, you need to eliminate half the crap and make the rest more responsive. I don't go to search engine to view tabloid stories (or if i did, i'd search for them), i'm there to search.

- Its not entirely clear why it seems that the links on the lite page are redirects, but the links on the main page appear to be direct? Is that deliberate? A bug? Obviously one would prefer them all to be direct.

- Don't know what's going on, it appeared to me subjectively that the main page was picking up successfully that I was from Australia. However, during one of my searches using the lite page, the options up the top of "web, news, social", etc looked like they were in another language (looked dutch-ish to me).

I'll try to use it but those news on the search page... I'm there to search, not to read news.

There is an addon for Firefox to use it as a search engine:

https://addons.mozilla.org/it/firefox/addon/qwantcom-for-fir...

I believe you can choose what you'd like to see.

It was possible to close and turn off the news panels and it is possible to see only web results (by clicking on the left panel). This choice is unfortunately not remembered.

The downside of being "cookie free" - instant forgetting of your preferences.

Not really - it seems that you can set your settings and get url that remembers them.

"If you haven't created a Qwant account to save your settings, you can use the following link as a homepage or drag it into your bookmarks: https://www.qwant.com/?l=en&h=0&hc=1&a=1&s=1&b=1&i=1&r=US&sr...

Well that's nice and all, but how is a typical user supposed to find and understand that? (Also drag it into your bookmarks? I don't know a single user who will understand that that's a thing.)

It's OK, reasonably fast, and I do like the interface, but search results as a list is so 1990s. Give me the result as a DAG. Give me sort options. Let me pin a result and then iterate in relation to it.

Show me something different, not the same thing as everyone else with a promise about my privacy and some different graphical sugar. To beat the established competitors you don't need to be incrementally better, you need to be qualitatively different.

I do prefer it DDG, at least for desktop use, because it's actually using the whole of my screen for a change. But wow me with something. I haven't had that in a while.

I don't want privacy to be a parameter that companies compete on, I want to government to enforce something sane that all the companies will then have to follow. Imagine if there were '50% less lead than our competitors' labels on food products.

I never understand log in on a search engine. I'm supplying keywords to search for, please don't try to profile and put me in a bubble.

After like 2 searches I got:

A high amount of connections have been detected from your location and you have been blocked.Please, validate the anti-robot below to be allowed access to the website.

I got the same notification

> https://about.qwant.com/job/ninja/

What a disappointment, I was expecting some job requirements regarding the handling of shuriken, nunchakus, sword fighting, art of camouflage, ...

Ah, and something like the graduation note of Shaolin Temple, in case of attendance.

How does qwant pays it bills? How do you do business? I didn't see any advertising, and there are no paid subscriptions, is the project funded by the EU?

This is an important question, I was wondering the same for very long with DDG. The answer seems to be ads, like everybody else:

https://help.qwant.com/help/overview/how-does-qwant-make-mon...

They say: "When you use Qwant, no personal information whatsoever is neither captured or transmitted to advertisers.", but then they also say: "We believe we now have designed a simple and efficient offering by working with the Microsoft Bing ad network."

Is it just me, or are these two statements oxymorons?

They probably say "give me some ads for keyword 'react js'" and don't pass on your IP or other identifying info.

Still a problem if you search for your own name, but I don't think you can do much about that. Or maybe MS provides a bloom table / index of all keywords they currenty have ads for, so you can check "offline" before sending it to them?

> is the project funded by the EU?

Yes. The EIB "loaned" them 25 million Euro:

http://www.eib.org/infocentre/press/releases/all/2015/2015-2...

I can't view the page either with emacs-w3m nor with w3m itself.

So that's one thing that DDG has over Qwant for me: it works with my browsers.

I like w3m and am a DDG user, but you have to admit you are using quite an esoteric browser:-)

Basic web site functionality should work in any browser.

If browsers can't keep up with basic web site functionality, they risk becoming irrelevant.

Have you tried lite.qwant.com? It's pretty usable for me in w3m, although still not as nice as DDG.

Thanks. lite.qwant.com seems to work.

Now I have to find a way to make it not give me indirect links like:

https://lite.qwant.com/redirect/IDSLFJKsdf99wejweiDFISDJFL92...

But give me direct links instead, like:

https://en.wikipedia.org/

On DDG this can be done by appending &kd=-1 to the URL.

Filtering by page language would be useful. Half of my results for a search query of a scientific term were the Wikipedia pages in other languages.

I'm really liking it so far. I was able to find some obscure stuff that google failed me on with image search, using rather ridiculous queries. A few random searches also returned more relevant results on the first page. Will definitely keep using it for a while to see if this keeps up.

So where is the bloody product? The silly little link on the about site is easy to miss.

Quite like it on first use. I've been meaning to ditch Google (can't be arsed with "do do no evil" anymore) for a while now and this looks like a good start.

I'm now close to convinced but the domain is .com - fair enough for world use but this sells itself as an EU based affair. Be who you are (says the MD of a UK company with a .net email address!)

I genuinely hope this works as advertised but I am having a bloody hard time finding out what is behind the scenes from the about website.

I will stick with it because their advertised basic premise strikes a note with me.

I simply fired up Firefox and searched (probably via G) for "quanta" and got a pretty severe Google notice.

Now I know what I am going to do. Bye Google - I was a customer for around 20 years.

I'm off.

What exactly do you mean by "a pretty severe Google notice"?

Nice, but I want an open source one with open datasets so that I know for sure what they are storing and eventually allow others to develop alternative engines. Nevertheless some competition in the search engine world is never bad.

you should have a look at searx instead, is also from the eu, and is completely free software, so free, that you as a user can demand the source code of the instance running due to the agpl license.

I like searx myself. To bad I can't get the search suggestions to work in the address bar. Also, I wish there was an app to run it off of your phone. It would be nice to not have to use a public instance.

many privacy search engines like duckduckgo really just query bing or google (sometimes through a proxy)...

any idea if qwant does this?

https://help.qwant.com/help/overview/how-does-qwant-index-th...

Although they just show Bing results for me so don't know how much of their own index is actually used.

It seems to depend on your location.

> In March 2017, news articles revealed Qwant displays mainly search results from Bing, except in France and Germany, despite several commitment to be exclusively "made in France".

https://en.wikipedia.org/wiki/Qwant#Criticism

Yeah, for the queries I tested with it returned the exact same results as Yahoo (which in turn gets its results from Bing).

The auto-completions are also very limited, no long tail suggestions at all.

They also seem to use Elasticsearch from looking at their job postings, but mostly the focus for hiring seems to be on front-end and scripting.

Too busy design wise at the moment and lacking back end details but the results seem decent at first glance.

No one can deny competition in search is desperately needed. Without that Google inevitably becomes more emboldened and seeks to further normalize and expand their creepy behavior.

An approach for a new engine should be both privacy which is a strong message that will resonate and also the technical details so people don't just assume its consuming another backend. That's what will make it interesting.

That's nice but it needs better support for indexing JavaScript single page apps. Right now only Google seems to be doing it properly.

That sounds like a feature, punishing poorly made "apps" that require javascript.

Not sure why you're getting downvoted, I'd use the shit out of a search engine that was optimized for sites that don't use javascript for basic functionality.

I might try it again sometime in the future, when it doesn't take more than 5 seconds to get a result to a query.

Not sure how it "social" search works, but it gives me a bunch russian bots, when I search "ukraine"

https://www.qwant.com/?q=%D1%83%D0%BA%D1%80%D0%B0%D0%B8%D0%B...

How is the privacy assurance of this vs DDG? What are the promises and does either proide more than promises?

Idk, not to disparage services like these, I'm sure they can be valuable to some degree from some threat actors, But I think on a personal level, one would think that the user should take a more adversarial role to protect their privacy rather to rely on Bottled Privacy™.

“Why Don’t We Have Both?” is a meme for a reason :P

Well at least for me, I'll wait for a non JS endpoint before trying some of this…

We track usage share of search engines. We do not have Qwant detection at this point, but what I can see:

- Out of 7972 sites, 561 have received at least one referral from Qwant.

We will need to explore this further to get usable numbers.

Looks promising, but my initial searching for Dart lang didn't return any results in News and Social, and unrelated images in the Images search.

The main thing I like about this service is the looks. It's mostly sleek and refreshingly minimal with a colorful and expressive logo to boot, a la Google.

DDG on the other hand... while I love the idea and would love to use it more often, not only gives poor results for most of my searches but the heavy use of red literally gives me a physical anxiety that makes it basically unusable. I could style the page myself but they really need to consider losing the red in favor of green or purple or gold because I can't be the only one who experiences this.

DGG has about as much red on it as Hacker News: is it such a heavy use of color? At least on desktop in Safari, the site is pretty 'white', and I'm surprised you get literal physical anxiety.

Interesting! Last time I used the service, about 3-4 months ago, the search results page was awash in red. Now it seems that the style has been much improved to the point of usability. I don't get anxiety like before. FWIW I was/am using Firefox on both Fedora and Windows.

How long has DDG had its current look for you?

I think I switched to DDG perhaps 5 months ago, although I'm not sure. Happy ducking, hah!

Image results is using either too big or too low resolution images that are irritating to look at.

Could be solved by making images smaller or increasing resolution.

I just did 3 searches on there and 2 times I got a server error. I will try Qwant now as default, but it already feels bad.

Been using qwant since it came out and I never looked back: great UI and search results

Won't even load the search input field without javascript...

Will they censor results like EU has been forcing Google to do?

Based in EU, I can't see how they wouldn't.

I don't want censored search.

If you mean the "Right to be Forgotten", you should know that there is a clear difference between censorship and this ruling.

http://ec.europa.eu/justice/data-protection/files/factsheets...

You mean, follow the law? Yes. Shocking, isn't it?

So people seem to want a privacy-protecting search engine that provides censored results based on the whims of governments. Ok, that is logical. Did a single person in the EU willingly vote for a censored web?

Does anyone actually want the government of France Germany or Turkey deciding what websites you are allowed to visit?

Why not put all the servers in a location that actually respects freedom and not censor anything AND not track us? That would be something I would be interested in. The whole concept of geo-restricted anything is a ridiculous antithesis to what the internet is supposed to be.

Turkey is not part of the EU.

Do they run their own crawling server or do they "repack" search request to google like DDG, Startpage or ixquick?

If they do it from scratch, kudos.

Reply on news.ycombinator.com