I’ve seen a lot of discourse over which browsers we use and I myself have made the switch from brave to firefox. I still use brave as my search engine though, so… which do yall recommend? Is brave’s engine necessarily bad to use? I personally like its ui/theme.
I have been using duckduckgo. I personally think it’s pretty good. So much so, that I think I have used it for 10 years now.
Same here, but it has also gone bad IMO.
Guess all the fake sites really got em all this time.
Kagi! Worth every penny of the subscription. The emphasis on privacy is a big deal for me but the killer feature is the ability to customize results. I have sites I personally like/trust towards the top and have an ever growing blacklist of sites that don’t get shown at all. No more pinterest, spruce, or other seo spam sites!
I tried for years to breakup with Google search, but always kept coming back to it for one reason or another. I started using Kagi a few months ago and have not even thought about Google since then. I really can’t recommend it enough, especially now that the $10/month plan is unlimited searches.
Good to hear that they updated their model
But doesn’t that naturally just recreate an even more personalised and hence affirming echo chamber the like of the algorithms on Google or Bing or so usually do?
Granted, plenty Spam sites to exclude, but is it easy to keep it to just those?
This is an interesting observation, not really something I have considered. The key difference here is that you are the one in control of those customizations. Whether the customizations are useful or harmful is entirely up to the user, Kagi just gives you the option.
For me at least, the majority of my searches I just want the correct answer to a question or a link to a specific resource I’m looking for. I don’t really use it as a content discovery engine. Being able to prioritize sites that I have found through experience to have reliable results and exclude sites that are uninformative or irritating is valuable.
Just the ability to replace items URLs in the result is such a great feature. I get rid of all Google / bing AMP stuff, reroute all Reddit answers to old.Reddit etc.
Browser: Firefox because I can download its source code, use it, inspect, modify and share. All of these 4 freedoms make Firefox free, as in freedom.
Brave is non-free (closed source and not contributing to software freedom).I use Qwant (unfortunately, it only works in the EU) and site-local search (Wikipedia, ArchWiki, etc.)
All web search engines are crap, honestly. Maybe Kagi makes better, idk.
EDIT: Brave is free, apparently.
Could you please elaborate why Brave isn’t open source? The source code is on github and there’s even an Brave-Git package in the AUR.
Ok, it seems I’ve been wrong. Apologies.
Kagi is great
Your question seems to be confusing between browser and search engine. These are two separate pieces of software.
But to answer both:
- Browser: Firefox. Google has demonstrated clearly that they cannot be trusted as the sole owner of the web which is what is about to happen as Chromium (which Brave is based on) fully takes over. Mozilla (makers of Firefox) is the last holdout. If you care, this is case in point about how Google having a monopoly on browsers will kill the free web: https://en.wikipedia.org/wiki/Web_Environment_Integrity.
- Search engine: Another +1 for Kagi. It has completely replaced Google for me.
I personally use Ecosia, it’s got a pretty decent privacy policy (esp compared to google/microsoft) and it uses Bing’s index so the results are pretty good. The main selling point is they use 100% of their profits for planting trees. The quality of the results is 99% of the time fine but if I ever get something I can’t find (or if I’m doing an image search) you can just add #g to your search to search google.
It’s overall a pretty good search engine by itself, but the fact it plants trees pushes it leagues ahead of the others for me
Do you run an adblocker?
edit: because if you do, then your searches don’t contribute towards the tree planting
No I block trackers but I do still get ads on Ecosia
ever click on them?
Yep! (If it’s relevant to my search)
They don’t get ad money otherwise (I assume that’s what you were getting at right?)
I was, yeah
SearXNG. It’ll search all the other engines for you, in a privacy preserving way.
Caveat that it is only privacy preserving if you trust whoever is hosting it.
The solution to that is to host it yourself. A VPS with 2GB RAM would be would be more than sufficient for SearxNG, and you can often find those for around $15/year (see GreenCloudVPS budget KVM line, RackNerd sales, other hosts on LowEndTalk). Or, you could run it on a Raspberry Pi, especially now that Raspberry Pis are more accessible and not out of stock everywhere.
Be careful though. Self hosting is addictive. You start with one service on a low-end VPS or Raspberry Pi, then you outgrow the server, expand, and eventually end up in !selfhosted@lemmy.world with a full-size server rack in your house.
I suppose I don’t fully understand the model, but if you host it yourself, then wouldn’t that be significantly worse for privacy because you’re essentially forwarding your searches to multiple search engines instead of one?
Again, maybe I’m missing something, but SearXNG just seems like a bunch of privacy memes mashed together without actually considering the threat model involved very deeply. Ultimately all of your data needs to be forwarded to a search provider in the end, the only way you are gaining any benefit would be if you had a sufficient pool of other users to obfuscate who is querying what, with a host who you are able to trust with your data.
In terms of privacy, most people mean things like tracking cookies, where Google can track you across all your devices. With something like SearXNG, requests still go to Google, etc. but they don’t know who you are. There aren’t any tracking cookies for them to know who you are, and if you host it on a VPS, the IP the search comes from isn’t yours so there’s no way to correlate anything by IP address.
That’s not how cookies work, cookies are stored on, and controlled by, your client. Unless your client is sharing the cookie information across devices, google wouldn’t be able to track you across devices using cookies. Routing all of your searches to one machine allows google to build a richer search profile against you, rather than one scattered across multiple different IPs, device fingerprints, etc.
I assume you mean across the web? That too is more of a different issue, as it is unrelated to the use of google search itself, rather it is due to the existence of tracking services embedded into the websites you visit.
At the end of the day, your VPS is still having a search profile build against it, in a similar manner to just using your personal device. The main difference is that you are paying to have a specialized computer that serves as a single purpose google searching device. Perhaps it is more challenging for them to link that device directly to you specifically, but I’d honestly bet that it is achievable.
That’s not how cookies work, cookies are stored on, and controlled by, your client. Unless your client is sharing the cookie information across devices, google wouldn’t be able to track you across devices using cookies.
Sorry, I didn’t mean to imply the same cookie is shared between clients. Google place cookies when you log in, and they know all the cookies associated with your account. You could always do searches when you’re logged out, across all your devices, which would prevent them from tracking you cross-device. It’s pretty rare for people to not be logged into a Google service though, especially on mobile.
Routing all of your searches to one machine allows google to build a richer search profile against you, rather than one scattered across multiple different IPs, device fingerprints, etc.
At the end of the day, your VPS is still having a search profile build against it, in a similar manner to just using your personal device.
You can get a bunch of people to use it though (friends and family). I also doubt they build profiles just based on IP, since it’s not uncommon to share IPs given the IPv4 shortage. There’s also CGNAT where hundreds of thousands of people share a sml number of public IPs.
It’s pretty rare for people to not be logged into a Google service though, especially on mobile.
If that is the case then this conversation is somewhat moot isn’t it? But I also don’t really think it is all that rare for iOS users.
You can get a bunch of people to use it though (friends and family).
That’s a good point, definitely see a benefit then.
I also doubt they build profiles just based on IP, since it’s not uncommon to share IPs given the IPv4 shortage. There’s also CGNAT where hundreds of thousands of people share a sml number of public IPs.
Certainly, and because of that you don’t really need a proxy.
There are definitely some benefits to such a setup, I just don’t think it really is superior to a search provider that is built around not logging and selling your searches. At least not to the degree it gets recommended in these types of posts.
I’ve been using DuckDuckGo for a while now, and while it works great for searches in English, I’ve had some problems with my native language (Finnish). However, with the use of bangs, this downside can be mitigated.
It was an great moment when I learned that bangs exist. I only use two or three but it’s still amazing.
Search engine or browser? For browsers I’ll use Firefox, but if I’m logging into anything I’ll usually use Chrome or Safari. I’ll also use Tor browser sometimes.
On the search engine side, I’ll generally use DuckDuckGo but I’m trying out Kagi to see if it’s worth paying for.
I’m another Kagi fan - after customising it a little it’s just so good, and I haven’t even played with features like lenses.
I really like the custom bang searches (e.g. I could make
!ks gravitysearch on simple Wikipedia), especially on mobile since Firefox Android doesn’t support the normal browser quicksearches (where you set a keyword for each search).Askjeeves
They’re just repackaging AltaVista results.
Excite!
My mum used Excite for a long time, well into the late 2000s and early 2010s. It was sort of frozen in time, at least for Australian users. Search still worked well enough, but the news articles on the homepage hadn’t been updated in over 5 years.
Nice reference.
Any love for qwant?
I’ve been using Whoogle for a while now.
4get.ca, it’s a metasearch engine, you can use it to get results from different search engines without trackers and anonymously
deleted by creator
