I used it for a long time (and 2 years as my main search engine) for 3 reasons...
- it is french
- it is working in Europa/USA/China (I studied in Paris/San Francisco/Tianjin)
- my best friend used to works their.
We had some good talks about Qwant, but it will not be relevant for my user review.
So here we are:
1. They are not Open-Source (as opposite to Searx).
Only the plugins are.
Can we trust them when they say to respect our privacy ?
2. They are not TOR friendly.
Full of captchaS !!!
3. They have no onion site.
DDG have one.
4. They run some kind of weird analytics.
Each time we click on a search, the JS code trigger a fetch to `https://api.qwant.com/api/action/url` and include:
- our current language
- the query we searched
- the link we clicked
- etc...
They were backed by Bing before, is it still the case or are they running their own stats engine? I do not know.
If we trigger ourself some false fetches, can we show "twitter" as first result when someone searches for "facebook"?
5. Their lite version is not lite (=/= duckduckgo.com/html/).
In sum up, in a customer point of view, Qwant is just a frenchy Google hosted mainly in Europa and allowed in China. Nothing new here, I recommend to stick to DDG for the moment.
Edit: fix minor typo + add proof for QwantLite analytics
"If you use one of my open-source repository, you can get in trouble, mouhahaha"
It feels like a dick move to me.
But then, I realize that the RIAA bans can now be used as an anti RIAA partnership against GitHub and others, which is... pretty cool.
So here we are:
1. They are not Open-Source (as opposite to Searx).
Only the plugins are. Can we trust them when they say to respect our privacy ?
2. They are not TOR friendly.
Full of captchaS !!!
3. They have no onion site.
DDG have one.
4. They run some kind of weird analytics.
Each time we click on a search, the JS code trigger a fetch to `https://api.qwant.com/api/action/url` and include: - our current language - the query we searched - the link we clicked - etc... They were backed by Bing before, is it still the case or are they running their own stats engine? I do not know. If we trigger ourself some false fetches, can we show "twitter" as first result when someone searches for "facebook"?
5. Their lite version is not lite (=/= duckduckgo.com/html/).
They redirect all our clicks to be able to run their analytics without the JS's fetch API (proof: `https://lite.qwant.com/?l=fr&q=hacker+news&t=web` the first link does not offer `https://news.ycombinator.com` but `https://lite.qwant.com/redirect/yFOdE8r1P1LTSLsA9IIBNKaZmDF1...`). Possible attack with `https://lite.qwant.com/redirect/yFOdE8r1P1LTSLsA9IIBNKaZmDF1...` (see the fake "&query") ? Also, They cannot store the config: go to settings, do whatever you want, do your search, switch tab (go to "news" for example) => your settings are reset to default. You may fix it by adding your settings to the link "&l=fr...".
6. Their front-side is broken.
If we visit their site without user-agent, we have an exception in their JS which crash the page (blank page). And their is more.
7. They do not care.
I emailed them maybe 5-6 bugs, they never replied nether fixed them.
8. Their API is sometime weird (just because not documented ?).
I ran a custom front-end without Qwant's analytics and the minimum working request is: `https://api.qwant.com/api/search/web?q=hackernews&count=10&o...`. What is "&uiv=4"? Why can't it be null or 0? What is "&t="? Is it really needed? Why is it not needed everywhere? => `https://api.qwant.com/api/suggest?q=duckduck`.
In sum up, in a customer point of view, Qwant is just a frenchy Google hosted mainly in Europa and allowed in China. Nothing new here, I recommend to stick to DDG for the moment.
Edit: fix minor typo + add proof for QwantLite analytics