Which privacy-focused search engine are you using?

schmurnan@lemmy.world to Technology@lemmy.world – 142 points –

TL;DR - which privacy-focused search engine do people recommend, preferably one that can also easily be used as a default option in Safari?

I ditched Google in about 2016ish I would guess, and since then have used DDG as my default search engine.

As someone entrenched in the Apple ecosystem, it’s always seemed like a sound choice, as it’s one of the search engines built in to Safari on both iOS and macOS.

After spending a bit more time recently playing around with and updating my Docker containers, I started hosting a Whoogle container, which seemed to work pretty well, but I don’t see many out there talking about it, so not sure how good it actually is. I then tried a SearXNG container, but either had it misconfigured or just wasn’t getting many search results back.

At the moment I’m trying out Startpage, but I know there are potential privacy concerns since they were part-bought in 2019 by a US ad-tech company.

I’m also playing around with different browsers at the moment, flicking between Safari, Firefox and Brave. At which point I stumbled across Brave Search, which seems pretty promising.

So, which search engines do you all recommend?

UPDATE: Probably should’ve done a poll! But latest (if I’ve captured everything correctly) is:

  • DuckDuckGo - 10
  • Qwant / SearXNG / Kagi / Brave - 4
  • Startpage / Ecosia - 2
  • Google - 1

As to my other questions around browsers:

  • Majority seem to use Firefox
  • Some mentions of Brave
  • One mention of Arc
91

I started paying for Kagi a few months ago and I'm loving it. Search results and tools are great. People balk at paying for a search engine, but at least this way I know I'm not the product.

I’m seeing a lot of love for Kagi as well. Not sure I’m quite ready to have to pay for search results, but I fully appreciate people that do. I have lots of subscriptions already and am trying to reduce them (i.e. self-hosting Vaultwarden as a replacement for 1Password).

I often recommend people using the "100 searches" trial of kagi to see if it might be worth it for them. I tried it and for me the price is worth it because of the custom filters and weights for results. It's really nice to search for something and all results on the first page are relevant content (no ads, seo-bait pages or stuff like Pinterest).

People don't even know what that feels like I think. Google has messed them up.

I remember the days when a Google search would usually get me what I wanted in the first result. And if not, at least it would be on the first page.

Now it's a mess of sponsored links, videos, and other irrelevant suggestions.

Kagi has been great. I'd like to see more searches on their $5 plan as 300 searches a month doesn't feel adequate. It has been great being able to promote, demote, and block sites from searching and I have found my results to be more helpful and relevant than Google alternatives I tried. I don't think I have used Google search at all since signing up. Highly recommend.

I've been loving my kagi trial. I used my 100 searches. I'll wait to see how long I can suffer before I buy it.

The weird thing for me is that by paying to use, you will need to be uniquely identified, and that opens doors for losing privacy in several ways. How is that addressed by kagi?

They address this on their website and go in further detail on their privacy policy. You'd have to read through that to decide if they do enough to earn your trust. But in my opinion not having any advertising removes most of their incentive to try to collect user data. I suppose there could be a temptation to collect the data to resell, but since a large portion of their income relies on the reputation of being a search alternative that has a focus on privacy I feel the risk to their reputation would be greater than whatever revenue that would generate.

Of course there are all kinds of companies that flew too close to the sun and sold out user privacy for a Coke and a smile. I've decided to go with Kagi and have been very happy with them so far. If they ever sell out I'll cash out, but they seem to be the best option for me right now.

DuckDuckGo from the browser, because 90% of the time I can get where I want with the appropriate ! bang from the address bar.

With the appropriate ! bang from the address bar

What does this mean? I want to like DuckDuckGo, but it’s kinda messy.

They are basically shortcuts. For example, I can type "!w ibuprofen" into DuckDuckGo (or the address bar because I have it set as my default search engine) and be brought immediately to the wikipedia page for Ibuprofen. There's also !yt for youtube search, !so for stack overflow search, and many more.

Just using duckduckgo. I'm not happy with my search results as they heavily prioritize clickbait CEO blogs instead of showing official documentation / sources.

Duckduckgo results are pretty bad in my experience, brave search and startpage are way better.

I’m using Ecosia. Planting trees FTW!

I run a pihole to block ads network-wide. I tried doing a general search for a bit of info on it and didn’t find much, but I guess my question is are the ads they run more like sponsored results, or like actual advertisements?

Using qwant because it's developed and hosted in France. Better than supporting a US company as a European.

I started using qwant last year for the same reason and I haven't looked back. Search results are excellent. I sometimes switch to Startpage as a proxy for google.

I’ve been using DuckDuckGo for years now and recommend it all the time.

I get decent, reliable results so haven’t shopped around.

This. Clean, reliable and functional. Although I’ve never wrapped my head around how they make their $$

They place ads in the search results.

If you disable your ad blocker, the first search result will almost always an Amazon ad. I test searched for "Headphones" and it also shows a row of JBL headphones at the top ala google shopping, which is an ad as well.

Really, they make their money the same way as google, they just don't track your searches and site visits.

@Viper_NZ

@schmurnan

Been a duckie for years now but I find myself going back to google for things like maps and shopping (that nearby search is a godsend) have you found any privacy mindfull alternatives?

Unfortunately not. If anyone knows of something I’d be keen.

On Android you can install Organic Maps, but the downside is that you'll need to download the maps for each area you want to have a map of (which, takes storage)

OsmAnd+ seems to be a good alternative recommended often, but I just haven't used it.

On PC, OpenStreetMaps is your friend

On Android you can install Organic Maps, but the downside is that you'll need to download the maps for each area you want to have a map of (which, takes storage)

OsmAnd+ seems to be a good alternative recommended often, but I just haven't used it.

On PC, OpenStreetMaps is your friend

Do you find DDG to be slower than Bing or Google? A few years ago I thought there was a noticeable delay with DDG, but not with Google.

Not really, they’re both nice and snappy.

Default search in Firefox: SearXNG (List of Instances) (solves 60-80%)

if not the solution, I then search for "dd [term]" which goes to duckduckgo. Solves mostly the rest.

If not, and I am really desperate, I try: "dd !g [term]" so it goes to ddg, redirects to google and then I am reminded how bad a first page result can be. Only ads, sponsored entries and only big company names. Good luck finding anything from a forum or a small blog on google today. All the search words are bundled up in company results that has nothing to do with the topic.

You don’t use an adblocker?

Sure. But how is that changing the quality of google results? Are you trolling?

I meant because you mentioned there’s nothing but ads which would be gone with an adblocker. Did you mean the actual results are essentially ads or literally the ads and Sponsored post at the top?

Sponsored results. I do not see ads, but I can see from the thing that I was searching for, and the difference between the that abd the first 4 results, that this can not be a real search result but a highly customized and profiled result. The results are just straight garbage.

ddg.gg for the win.

Started using them when I confronted the Google filter bubble for the first time.

But what totally sold me? DDG provided me link for a product I had searched for in vain on Google for years at that point, using the exact same query.

I used to think Google search was clean until I used ddg. For some reason I prefer how ddg display searches.

Please reveal the product. I am consumed by curiosity.

Steel leveling jacks, manual, center screw variety, used extensively for leveling mobile homes back in the day.

First time I used DDG to look for them, BOOM, half way down the page a manufacturer in Enid, OK. G? Literally for years kept serving me up pneumatic, expensive, industrial gear or services.

That is definitely not what I was expecting

That is definitely not what I was expecting

Please reveal what you were expecting. I am consumed by curiosity.

Your post may have just convinced me to try ddg. My problem with Google is that increasingly it does not give me what I am looking for and instead gives me what it thinks I should want. From your description it sounds like ddg is what Google used to be.

Pretty much. They deliver a couple of sponsored entries per search, but even those have been occasionally productive. No harm in trying for a while, you can always change if it’s not working. GL&HF!

My own searxng instance, listed at https://searx.space

Perhaps I need to go back to figuring out SearXNG. Although I did read that there’s a slight privacy compromise to use SearXNG over SearX.

Perhaps that doesn’t apply if it’s self-hosted, though.

Using self hosted SearXNG as well. If you have docker running already it's a snap to setup.

Yeah I’ve got Docker up and running with lots of containers, but wasn’t getting any results back from SearXNG for some reason, kept getting timeouts. I’ve got everything running behind Traefik so might just be a bit of a config issue.

I probably wouldn't expose it externally to your network, but that's just me. Unless that's what you're doing, there aren't really any ports to expose with Traefik.

No, I’m using local DNS in Pi-hole with a domain I own but nothing is externally accessible. And even if it were, I’ve got Authelia on all my containers (except, Vaultwarden because I can’t get it to work).

I've had a pretty similar journey to yours and I'm currently using Qwant, although the only reason I'm using them is because they're based in Europe and haven't had any scandals that I could find. If you're really concerned about privacy I've heard good things about Kagi.

They have a partnership with Microsoft. Not saying it’s a scandal but that was my reason to stop using it

It's so difficult to compete with free though. The monthly fee for kagi is a bit steep, which is not for everyone (>5 €/month), even if that is the 'true' price of searching the web. It does have some very interesting features though. A colleague of mine is always praising it, saying that the search results are better than anything else he has tried.

Personally, I use duck duck go as my primary search engine but I only use it to search for things. But since chatgpt4, I use perplexity.ai for actual answers. Browser wise, I use duck duck go where I can or Arc on my Mac with all the usual privacy plugins.

What are your thoughts on Arc? I tried it a couple of months ago but couldn’t really get used to the layout, etc.

Sure it’s as good a browser as any, I just wasn’t feeling it.

For me personally, it fits how work with the web. I really like the full screen view and the tab navigation between spaces and bookmarks. Because it’s so different, There is a learning curve but once passed it, I genuinely enjoy the experience it offers. I don’t care much for the other customisation features but I can see why they’d appeal to users.

Currently using Startpage and my only criticism of it is that it is noticeably slower than the other 2 that I have used DDG and Brave

I would recommend DDG for most and Startpage only for the people who don't want to use Google search anymore but want similar if not identical search results

I did change the default search engine of some of my relatives to Startpage without them knowing:)

I like the fact that brave search has an AI, that's why I use it. I might self-host a foss search engine though

I recently started using Duck Duck Go instead of Google.

I still had this thought that Google was the best technology as I went from the early web (remember... altavista.digital.com, yahoo, ...) and I remember that Google was really a game changer when it started to become popular.

I tested setting DDG as default search engine in my desktop and mobile browsers, thinking that when I don't have expected results I would go to Google... I never had to switch to Google because I was wrong, DDG is as good as Google while being better from privacy point of view.

For the browser I use Vivaldi on both android and desktop.

Does anyone have experience/thoughts about Mojeek?

I just recently learned about it and haven't really explored it, but curious.

Thanks!

1 more...

I'm using an Ad-, Tracking- etc blocker in all my devices, so I'm not too worried about using Google or Bing when I do.

But I'm hosting my own instance of SearchXNG and that's often simply the most powerful and flexible search engine.

Also self-host SearchXNG. Its definitely the way to go. I like you can choose which search engines to pull from

I replied to another comment on here saying that I'd tried this once before, via a Docker container, but just wasn't getting any results back (kept getting timeouts from all the search engines).

I've just revisited it, and still get the timeouts. Reckon you're able to help me troubleshoot it?

Below are the logs from Portainer:

 File "/usr/local/searxng/searx/network/__init__.py", line 165, in get
    return request('get', url, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/searxng/searx/network/__init__.py", line 98, in request
    raise httpx.TimeoutException('Timeout', request=None) from e
httpx.TimeoutException: Timeout
2023-08-06 09:58:13,651 ERROR:searx.engines.soundcloud: Fail to initialize
Traceback (most recent call last):
  File "/usr/local/searxng/searx/network/__init__.py", line 96, in request
    return future.result(timeout)
           ^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/concurrent/futures/_base.py", line 458, in result
    raise TimeoutError()
TimeoutError
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
  File "/usr/local/searxng/searx/search/processors/abstract.py", line 75, in initialize
    self.engine.init(get_engine_from_settings(self.engine_name))
  File "/usr/local/searxng/searx/engines/soundcloud.py", line 69, in init
    guest_client_id = get_client_id()
                      ^^^^^^^^^^^^^^^
  File "/usr/local/searxng/searx/engines/soundcloud.py", line 45, in get_client_id
    response = http_get("https://soundcloud.com")
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/searxng/searx/network/__init__.py", line 165, in get
    return request('get', url, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/searxng/searx/network/__init__.py", line 98, in request
    raise httpx.TimeoutException('Timeout', request=None) from e
httpx.TimeoutException: Timeout
2023-08-06 09:58:13,654 ERROR:searx.engines.soundcloud: Fail to initialize
Traceback (most recent call last):
  File "/usr/local/searxng/searx/network/__init__.py", line 96, in request
    return future.result(timeout)
           ^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/concurrent/futures/_base.py", line 458, in result
    raise TimeoutError()
TimeoutError
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
  File "/usr/local/searxng/searx/search/processors/abstract.py", line 75, in initialize
    self.engine.init(get_engine_from_settings(self.engine_name))
  File "/usr/local/searxng/searx/engines/soundcloud.py", line 69, in init
    guest_client_id = get_client_id()
                      ^^^^^^^^^^^^^^^
  File "/usr/local/searxng/searx/engines/soundcloud.py", line 45, in get_client_id
    response = http_get("https://soundcloud.com")
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/searxng/searx/network/__init__.py", line 165, in get
    return request('get', url, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/searxng/searx/network/__init__.py", line 98, in request
    raise httpx.TimeoutException('Timeout', request=None) from e
httpx.TimeoutException: Timeout
2023-08-06 10:02:05,024 ERROR:searx.engines.wikidata: engine timeout
2023-08-06 10:02:05,024 ERROR:searx.engines.duckduckgo: engine timeout
2023-08-06 10:02:05,024 ERROR:searx.engines.google: engine timeout
2023-08-06 10:02:05,024 ERROR:searx.engines.qwant: engine timeout
2023-08-06 10:02:05,024 ERROR:searx.engines.startpage: engine timeout
2023-08-06 10:02:05,024 ERROR:searx.engines.wikibooks: engine timeout
2023-08-06 10:02:05,024 ERROR:searx.engines.wikiquote: engine timeout
2023-08-06 10:02:05,024 ERROR:searx.engines.wikisource: engine timeout
2023-08-06 10:02:05,025 ERROR:searx.engines.wikipecies: engine timeout
2023-08-06 10:02:05,025 ERROR:searx.engines.wikiversity: engine timeout
2023-08-06 10:02:05,025 ERROR:searx.engines.wikivoyage: engine timeout
2023-08-06 10:02:05,025 ERROR:searx.engines.brave: engine timeout
2023-08-06 10:02:05,481 WARNING:searx.engines.wikidata: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,481 ERROR:searx.engines.wikidata: HTTP requests timeout (search duration : 6.457878380082548 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,482 WARNING:searx.engines.wikisource: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,484 ERROR:searx.engines.wikisource: HTTP requests timeout (search duration : 6.460748491808772 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,485 WARNING:searx.engines.brave: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,485 ERROR:searx.engines.brave: HTTP requests timeout (search duration : 6.461546086706221 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,487 WARNING:searx.engines.google: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,487 ERROR:searx.engines.google: HTTP requests timeout (search duration : 6.463769535068423 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,489 WARNING:searx.engines.wikiversity: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,489 ERROR:searx.engines.wikiversity: HTTP requests timeout (search duration : 6.466003180015832 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,490 WARNING:searx.engines.wikivoyage: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,490 ERROR:searx.engines.wikivoyage: HTTP requests timeout (search duration : 6.466597221791744 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,490 WARNING:searx.engines.qwant: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,490 ERROR:searx.engines.qwant: HTTP requests timeout (search duration : 6.4669976509176195 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,491 WARNING:searx.engines.wikibooks: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,491 ERROR:searx.engines.wikibooks: HTTP requests timeout (search duration : 6.4674198678694665 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,491 WARNING:searx.engines.wikiquote: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,492 WARNING:searx.engines.wikipecies: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,492 ERROR:searx.engines.wikiquote: HTTP requests timeout (search duration : 6.468321242835373 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,492 ERROR:searx.engines.wikipecies: HTTP requests timeout (search duration : 6.468797960784286 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,496 WARNING:searx.engines.duckduckgo: ErrorContext('searx/engines/duckduckgo.py', 98, 'res = get(query_url, headers=headers)', 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,497 ERROR:searx.engines.duckduckgo: HTTP requests timeout (search duration : 6.47349306801334 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:02:05,511 WARNING:searx.engines.startpage: ErrorContext('searx/engines/startpage.py', 214, 'resp = get(get_sc_url, headers=headers)', 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:02:05,511 ERROR:searx.engines.startpage: HTTP requests timeout (search duration : 6.487425099126995 s, timeout: 6.0 s) : TimeoutException
2023-08-06 10:04:27,475 ERROR:searx.engines.duckduckgo: engine timeout
2023-08-06 10:04:27,770 WARNING:searx.engines.duckduckgo: ErrorContext('searx/search/processors/online.py', 118, "response = req(params['url'], **request_args)", 'httpx.TimeoutException', None, (None, None, None)) False
2023-08-06 10:04:27,771 ERROR:searx.engines.duckduckgo: HTTP requests timeout (search duration : 3.2968566291965544 s, timeout: 3.0 s) : TimeoutException
2023-08-06 10:04:50,094 ERROR:searx.engines.duckduckgo: engine timeout
2023-08-06 10:04:50,187 WARNING:searx.engines.duckduckgo: ErrorContext('searx/engines/duckduckgo.py', 98, 'res = get(query_url, headers=headers)', 'httpx.ConnectTimeout', None, (None, None, 'duckduckgo.com')) False
2023-08-06 10:04:50,187 ERROR:searx.engines.duckduckgo: HTTP requests timeout (search duration : 3.0933595369569957 s, timeout: 3.0 s) : ConnectTimeout

The above is a simple search for "best privacy focused search engines 2023", followed by the same search again but using the ddg! bang in front of it.

I can post my docker-compose if it helps?

First thing that comes to mind is are you running it on Host Network? That's a requirement

I am using startpage.com since they include Google results and DDG just has bad results sometimes

For search engine, I go with SearXNG, and for web browser, Mull. Or hardened Firefox, on PC.

Do you use a self-hosted SearXNG, or one of the other hosted instances?

Self-hosted under a VPN would be the way to go if you want to be 100% sure.

The problem with public instances is that you can't really know what the owner does with the data. There are safe ones, and malicious ones. I'd just look for one that has a good reputation.

Yeah I have a self-hosted one but I’m struggling to get results. I posted under another comment on this thread, was just gonna ask for some support troubleshooting.

Completely agree though, self hosting over public instances all day long.

I tried ecosia and startpage, the results through both are just 95% garbage. Duck Duck Go seems to be okay.

I use Vivaldi as my browser, includes ad block and various useful, if bulky, features.

DDG cause the bangs are nice, but I find myself searching generally very little these days. I usually just use a bang to search a site I know I will find what I want on, if no bang then I will just navigate to that site. Search results have been shit for over a decade.

I use Firefox as main browser, but I discovered for my use case google provides the best results without needing to setup every workstation e.g. 2 home PCs, 1 mobile, 2 for work. And that I need to use all main 3 browsers. Also google provide good service with functions to quickly make currency conversions, simple math ekvations etc which even Bing is far behind.

"AI" services will change this but for now it's too slow.

But in general for me, I have given up that fact to try stay private many years back, it's all a dream just like living off grid, 99% of would not survive 4 days.

But the information can be scrambled, ie shift user accounts, services, software etc. It would also provide better competition due to the userbase is moving around... But most of us are too lazy or afraid to lose history, backups, photos etc. Just see how many that can't just delete an old reddit account due to the time spent to reach an level you aren't ready to leave. To lead to famous qoute I follow online

"Don't let yourself get attached to anything you are not willing to walk out on in 30 seconds flat if you feel the heat around the corner"

Its all about supporting the services you like and are trying to be an counterweight to the other common commercial services... Meaning we need to found/pay for good services, privacy is a luxary looking on the whole user base.

Companies, I based on an idea, but exist to make someone money and if it's tracks it will make many people money and in the end majority will lead the company to earn money and leave the base idea behind.

I use presearch. Pretty happy with it. Never paid out my rewards, they're just growing

Been using Qwant for a few years. Good enough for 90% of searches imo. For whatever's left, I'll use DDG, Google in incognito, or Bing.

privacy focused ... called qwant

Hmm.