Ask HN: Alternatives to DuckDuckGo and Firefox?
55 comments
I just got into the beta for https://kagi.com/ which at this time seems at least as good if not better than DDG
I've been using it for a few months now and I like it.
However, I don't want to pay more than $100/year for a Google/Bing frontend. Startpage works quite well for that. If they made their own indexer I would go all-in, and that would justify such a high price-tag.
However, I don't want to pay more than $100/year for a Google/Bing frontend. Startpage works quite well for that. If they made their own indexer I would go all-in, and that would justify such a high price-tag.
They also make a (as they claim) zero telemetry browser called Orion (https://browser.kagi.com). It is super fast, but there's not enough trust built there yet for me to use it as my main browser. I just use it as a dedicated HN browser for now. It has sidebar tabs built in which is _the_ killer feature for me. I think Edge did this best, but I refuse to use a MS browser.
I've been using Kagi full time for a few weeks now and it has been a very pleasant experience. Search results seem consistently better than DDG and my !g use is significantly down. And it is really really fast. YMMV.
I'm curious if this company will survive long term. I fully plan to pay a subscription once they introduce that, but the way they describe it in their FAQ makes it sound way too niche for my taste.
They should make one unified (low) price for Kagi with no search caps. And then offer a "Premium" plan with some extra features (NOT the number of searches done). Segregating into entry level and "unlimited" (horrible name btw) plans sounds like a recipe for disaster considering all other search engines are unlimited by default AND free by default.
Maybe techy people will value things like no telemetry and no ads but I question if there's enough of us that value these things AND are willing to pay for this AND choose to become their customer. Good luck to them. :)
I've been using Kagi full time for a few weeks now and it has been a very pleasant experience. Search results seem consistently better than DDG and my !g use is significantly down. And it is really really fast. YMMV.
I'm curious if this company will survive long term. I fully plan to pay a subscription once they introduce that, but the way they describe it in their FAQ makes it sound way too niche for my taste.
They should make one unified (low) price for Kagi with no search caps. And then offer a "Premium" plan with some extra features (NOT the number of searches done). Segregating into entry level and "unlimited" (horrible name btw) plans sounds like a recipe for disaster considering all other search engines are unlimited by default AND free by default.
Maybe techy people will value things like no telemetry and no ads but I question if there's enough of us that value these things AND are willing to pay for this AND choose to become their customer. Good luck to them. :)
> They also make a (as they claim) zero telemetry browser called Orion
Would be easier to trust them if they didn't want an email ID to download their browser. Seeing that they also want to create a search engine, I suspect that your search and browsing history could be linked to it to profile you.
Would be easier to trust them if they didn't want an email ID to download their browser. Seeing that they also want to create a search engine, I suspect that your search and browsing history could be linked to it to profile you.
That might be because both the browser and search are in beta. Though, since search will be a paid service they will necessarily need to have some info on the user.
Then zero telemetry offers no privacy advantage when they are collecting and collating even more personal data of the user!
> I think Edge did this best, but I refuse to use a MS browser.
Edge is a good browser, but an absolute privacy disaster. Hardware based session IDs, chatty start pages that push you to Bing, the typical search suggestions on by default. Microsoft runs their own sync backend but does not allow for end to end encrypting all data types. Among the ones not e2ee'd? Browsing history.
Edge is a good browser, but an absolute privacy disaster. Hardware based session IDs, chatty start pages that push you to Bing, the typical search suggestions on by default. Microsoft runs their own sync backend but does not allow for end to end encrypting all data types. Among the ones not e2ee'd? Browsing history.
I have also used kagi for a few weeks. The results are much much better than Google.
And I am using it constantly.
And I am using it constantly.
Interesting - thank you for the recommendation, just signed up as well. Found it a little cute they seem to be intentionally(?) riffing off of dogpile's old branding.
Kagi +1
I love blocking domains without uBlacklist.
I love blocking domains without uBlacklist.
+1 on Kagi for the nice UX.
Also, shameless plug: you can also remove whole domains from search results with uBlockOrigin and https://letsblock.it/filters/search-results
Our template currently supports Bing, DuckDuckGo, Google, Kagi, Searx and Startpage. It also includes github/stackverflow presets taken from the uBlock-Origin-dev-filter project.
Also, shameless plug: you can also remove whole domains from search results with uBlockOrigin and https://letsblock.it/filters/search-results
Our template currently supports Bing, DuckDuckGo, Google, Kagi, Searx and Startpage. It also includes github/stackverflow presets taken from the uBlock-Origin-dev-filter project.
Blocking domains for search results?
I understand why I need a sign up to use Search, so they at least can have a subscription model - but that seems like a barrier for entry and also the necessity to instill a lot of faith into a product that hasn't garnered trust (yet, rightfully so)
I understand why I need a sign up to use Search, so they at least can have a subscription model - but that seems like a barrier for entry and also the necessity to instill a lot of faith into a product that hasn't garnered trust (yet, rightfully so)
I often wanted to ban pinterest since I cannot view the source without account anyway. So there would be sensible applications.
Brave Search. Brendan Eich recently said Brave doesn't censor.
Here are some of his tweets on the matter:
- https://twitter.com/BrendanEich/status/1501978997860683776?s...
- https://twitter.com/BrendanEich/status/1502113286417838082?s...
Here are some of his tweets on the matter:
- https://twitter.com/BrendanEich/status/1501978997860683776?s...
- https://twitter.com/BrendanEich/status/1502113286417838082?s...
Brave is probably not that privacy friendly as their marketing says.
- https://blackgnu.net/brave-is-shit.html
- https://blackgnu.net/brave-is-shit.html
Let's go through some of this thing:
> But Brave is more private than Firefox by default > No, not at all. People who claim this have fallen for Brave’s marketing strategy which consists on telling lies and flawed arguments.
The first thing to note here: Firefox defaults to Google as the default search engine and has search suggestions turned on, so whatever you write in the address bar gets sent to Google. Doing worse than that is really, really fucking hard.
> Their adblocker is just a fork of uBlock Origin
How, when uBlock Origin is a browser extension and Brave's adblocker is written in Rust and integrated into the browser.
> Another problem with their built-in adblocker is that it’s better for extensions to be separated from the core of the browser, since they don’t follow each other’s update cycles. This means that you need to update the entire browser to fix a bug in the adblocker. Stupid, isn’t it?
Yeah, Manifest v3 would like a word. uBlock will be fucked, Brave Shields won't care.
> Another reason to avoid using Brave is that uBlock Origin works best on Firefox and there isn’t anything that Brave can do about it.
These are limitations of the extension system in Chromium. Brave Shields isn't an extension, and actually does do things like CNAME uncloaking, which extensions can do on Firefox and can't do on Chromium.
What was that about the blocker being integrated into the browser again?
> Rewards is their shitty program that will replace ads displayed on websites with their own.
The adblocking system and the ads from Brave Rewards are entirely separate.
You only need to identify to cash in the BAT. Yeah, some people are looking to make pennies from their cut of being an ad viewer (spoiler: single ad viewers arem't very valuable). The whole schtick of the system is to tip content creators, where a pileup of BAT can be a real sum of money and some compensation for adblocking tracking ads.
> Brave has been caught inserting affiliate codes > In June 2020, a twitter user (@cryptonator1337) discovered that Brave was automatically injecting referral codes into URLs for cryptocurrency exchange sites. > > So if you typed “binance.us” into the URL bar and pressed enter, Brave would take you to “binance.us/?ref=35089877”. > > There was a huge scandal when this was noted. Later, Brave disabled this in the code, in a “sorry we got caught” style.
Yeah. You know this Firefox Suggest thing Mozilla recently released? Same deal. Type incomplete URL, it suggests an ad from a local catalog. Suggesting an ad when typing a full URL was a bug, was fixed.
And you know what else they did? They turned the whole feature off by default.
I have to give the hit piece credit for not insinuating they replaced urls within the page, as many others do.
> But Brave is more private than Firefox by default > No, not at all. People who claim this have fallen for Brave’s marketing strategy which consists on telling lies and flawed arguments.
The first thing to note here: Firefox defaults to Google as the default search engine and has search suggestions turned on, so whatever you write in the address bar gets sent to Google. Doing worse than that is really, really fucking hard.
> Their adblocker is just a fork of uBlock Origin
How, when uBlock Origin is a browser extension and Brave's adblocker is written in Rust and integrated into the browser.
> Another problem with their built-in adblocker is that it’s better for extensions to be separated from the core of the browser, since they don’t follow each other’s update cycles. This means that you need to update the entire browser to fix a bug in the adblocker. Stupid, isn’t it?
Yeah, Manifest v3 would like a word. uBlock will be fucked, Brave Shields won't care.
> Another reason to avoid using Brave is that uBlock Origin works best on Firefox and there isn’t anything that Brave can do about it.
These are limitations of the extension system in Chromium. Brave Shields isn't an extension, and actually does do things like CNAME uncloaking, which extensions can do on Firefox and can't do on Chromium.
What was that about the blocker being integrated into the browser again?
> Rewards is their shitty program that will replace ads displayed on websites with their own.
The adblocking system and the ads from Brave Rewards are entirely separate.
You only need to identify to cash in the BAT. Yeah, some people are looking to make pennies from their cut of being an ad viewer (spoiler: single ad viewers arem't very valuable). The whole schtick of the system is to tip content creators, where a pileup of BAT can be a real sum of money and some compensation for adblocking tracking ads.
> Brave has been caught inserting affiliate codes > In June 2020, a twitter user (@cryptonator1337) discovered that Brave was automatically injecting referral codes into URLs for cryptocurrency exchange sites. > > So if you typed “binance.us” into the URL bar and pressed enter, Brave would take you to “binance.us/?ref=35089877”. > > There was a huge scandal when this was noted. Later, Brave disabled this in the code, in a “sorry we got caught” style.
Yeah. You know this Firefox Suggest thing Mozilla recently released? Same deal. Type incomplete URL, it suggests an ad from a local catalog. Suggesting an ad when typing a full URL was a bug, was fixed.
And you know what else they did? They turned the whole feature off by default.
I have to give the hit piece credit for not insinuating they replaced urls within the page, as many others do.
Okeano [0] is a free speech absolutist and privacy friendly search engine. [1]
[0] https://okeano.com
[1] https://twitter.com/OkeanoSearch/status/1502077779243847686
[0] https://okeano.com
[1] https://twitter.com/OkeanoSearch/status/1502077779243847686
Librewolf is a Firefox fork stripped out of all telemetry with additional privacy oriented default settings.
Vivaldi browser - https://vivaldi.com/
I switched to it from chrome and never looked back. Tons of good features, really customizable, and (what seems like) a pretty sound business model (no crypto!)
I switched to it from chrome and never looked back. Tons of good features, really customizable, and (what seems like) a pretty sound business model (no crypto!)
One of my colleagues uses Vivaldi, and he is constantly finding the weirdest bugs that I can only ever reproduce in Vivaldi, and that seem to break with every spec I can find.
* In all other browsers, clicking a "select file" element and then cancelling out of the dialog does not fire a `change` event on that element - except in Vivaldi, where you get a `change` event with no files in it. * In all other browsers, setting the `Content-Length` header is enough for the browser to be able to give the user a file size estimate. Not in Vivaldi, which seems to use some other heuristic that I never managed to figure out - currently on our systems, the download dialog shows the file size as "unknown" and I've left it at that. * Vivaldi has a feature where, if you close the window and then reopen it, it opens the page you were previously on. It's a nice feature, and I like it. Unfortunately, for the instant that the page is being reopened, any `matchMedia()` listeners will produce weird results. (Note that `window.screen.width` will still report a reasonable value, just one that contradicts the result of the `matchMedia()` query.) This is by far the weirdest and most niche bug I've found.
Obviously every browser has its own quirks - although modern browsers tend to have far fewer of them, so I'm much less tolerant of an evergreen browser that doesn't match spec these days - but what makes it far worse in Vivaldi's case is that, if you do find one of these bugs, the only way to report it is an online form with almost no feedback at all - there's no public list of bugs and not even a way to view the status on your own bug. Apparently they email you if something changes, but I've not experienced that yet.
In all fairness, from having tried out Vivaldi to reproduce these weird bugs, I'm quite impressed with how the browser feels, but the non-compliance with standards and the inscrutability of the bug report system have turned me off completely.
* In all other browsers, clicking a "select file" element and then cancelling out of the dialog does not fire a `change` event on that element - except in Vivaldi, where you get a `change` event with no files in it. * In all other browsers, setting the `Content-Length` header is enough for the browser to be able to give the user a file size estimate. Not in Vivaldi, which seems to use some other heuristic that I never managed to figure out - currently on our systems, the download dialog shows the file size as "unknown" and I've left it at that. * Vivaldi has a feature where, if you close the window and then reopen it, it opens the page you were previously on. It's a nice feature, and I like it. Unfortunately, for the instant that the page is being reopened, any `matchMedia()` listeners will produce weird results. (Note that `window.screen.width` will still report a reasonable value, just one that contradicts the result of the `matchMedia()` query.) This is by far the weirdest and most niche bug I've found.
Obviously every browser has its own quirks - although modern browsers tend to have far fewer of them, so I'm much less tolerant of an evergreen browser that doesn't match spec these days - but what makes it far worse in Vivaldi's case is that, if you do find one of these bugs, the only way to report it is an online form with almost no feedback at all - there's no public list of bugs and not even a way to view the status on your own bug. Apparently they email you if something changes, but I've not experienced that yet.
In all fairness, from having tried out Vivaldi to reproduce these weird bugs, I'm quite impressed with how the browser feels, but the non-compliance with standards and the inscrutability of the bug report system have turned me off completely.
Firefox has bad management after they fired people with wrong opinions, but it is still better than the Chrome monoculture. This will especially apparent if they migrate to manifest v3 completely around next year.
DDG is far easier replaced. Brave for example.
DDG is far easier replaced. Brave for example.
https://brave.com
Browser and search engine.
Browser and search engine.
The browser with built-in ads lol?
Reminder: https://neilzone.co.uk/2021/11/brave-browser-less-privacy-re...
i.e. the browser redirecting you to affiliate links without permission? https://www.zdnet.com/article/privacy-browser-brave-busted-f...
i.e. the browser redirecting you to affiliate links without permission? https://www.zdnet.com/article/privacy-browser-brave-busted-f...
"Sorry only because we got caught, and only because it looks like enough people care".
Every browser out there makes its money from ads, more or less. Either that, or they're a pure charity project forking a browser that makes their money from ads.
well, after closing a page that advertised some disney crap on firefox...
You know those features can be easily disabled right? I never got any ad on firefox.
i actually had all of the settings off, it was that fake patch notes one :/
> Brave Search doesn’t use secret methods or algorithms to bias or censor results
I'm sold.
Then I open and type "rt". The autocomplete tells me this is a "state controlled Russia network", while for BBC it says "public service broadcaster". So much for "no bias" (technically all they said is the bias is not secret, but still). Not sure where they get these snippets from. Looks similar to wikipedia but it's not verbatim.
I'm sold.
Then I open and type "rt". The autocomplete tells me this is a "state controlled Russia network", while for BBC it says "public service broadcaster". So much for "no bias" (technically all they said is the bias is not secret, but still). Not sure where they get these snippets from. Looks similar to wikipedia but it's not verbatim.
> BBC it says "public service broadcaster"
I'm not the biggest fan of the BBC, but it certainly is not state controlled like RT is.
I'm not the biggest fan of the BBC, but it certainly is not state controlled like RT is.
BBC is just way way better at it, way more subtle, having been doing this for 100 years. When push comes to shove they toe the line just like Russian or Chinese media. RT just doesn't have the same funding or talent to make stories more sophisticated. Obviously, the gov must hide any links to BBC to maintain the appearance of free press. But I'm sure pressure is exerted one way or the other, for example:
> From the late 1930s until the end of the Cold War, MI5 had an officer at the BBC vetting editorial applicants.
Looking at how UK and EU now outright banned Russian TV and media, I think it would go unnoticed if they started removing or just not hiring certain inconvenient journalists. Certain thoughts are just forbidden these days; it's career suicide to even doubt the gov line, just like during covid when it was practically forbidden to question gov measures or vaccines.
> From the late 1930s until the end of the Cold War, MI5 had an officer at the BBC vetting editorial applicants.
Looking at how UK and EU now outright banned Russian TV and media, I think it would go unnoticed if they started removing or just not hiring certain inconvenient journalists. Certain thoughts are just forbidden these days; it's career suicide to even doubt the gov line, just like during covid when it was practically forbidden to question gov measures or vaccines.
The BBC sometimes expresses a fairly conservative world-view (I mean conservative in the traditional sense, not populist like Boris Johnson) which sometimes happens to broadly align with the government's perspective. I think the BBC often seeks the outsider perspective, but seldom centres it.
The current UK government is not as enthusiastic about the BBC as they might be; and the BBC's senior management treats its relationship with the government as something to be managed.
The BBC is not a monolith. Even if the government was able to unduly influence some people inside the BBC, that influence wouldn't extend to much of the BBC's output, because frankly the BBC is not very good at operating as a single joined-up entity.
The current UK government is not as enthusiastic about the BBC as they might be; and the BBC's senior management treats its relationship with the government as something to be managed.
The BBC is not a monolith. Even if the government was able to unduly influence some people inside the BBC, that influence wouldn't extend to much of the BBC's output, because frankly the BBC is not very good at operating as a single joined-up entity.
If anything the Tory governments have more control over it because they actively threaten its existence.
That's an absurd comparison. Tory may threaten BBC. There are still articles which stop just short of "here's how Boris lied" (https://www.bbc.com/news/60679290) On the other hand RT publishes what Kremlin wants it publish. There is no opposing view allowed. They'd simply not exist if they wanted to be independent.
I'm not playing in your RT vs BBC sandbox.
They're literally legally required to do that by EU sanctions, and Eich has been clear that they comply with the law, but no more.
A lot of other search engines could also be doing only that, but are not communicating the scope and whether they're doing more than what they're compelled to. The chief concern is with search companies editorializing out of their own volition.
A lot of other search engines could also be doing only that, but are not communicating the scope and whether they're doing more than what they're compelled to. The chief concern is with search companies editorializing out of their own volition.
> So much for "no bias"
Welcome to reality.
Welcome to reality.
They have to go with the flow or they money is cut.
I’ve been using Brave and quite like it. Every now and then I find myself missing some features from Chrome, but nothing I miss enough to remember what it is. The idea of earning crypto for surfing the web is also interesting, but it’s mostly a gimmick to me and I ignore it.
Searx is interesting for replacing DDG.
Interesting, haven't heard of it. Would be curious as to what made you pick it, if you don't mind.
Their tagline says it all:
“a privacy-respecting, hackable metasearch engine”
“a privacy-respecting, hackable metasearch engine”
+1 for Searx - love the idea of meta-search
mojeek is truly a totally different index.
I've been using https://metager.org for a month. Looks pretty neat, no ads so far.
Startpage has been my search engine for a while. I’m surprised no one has mentioned it here.
neeva.com is a subscription search engine. I think privacy counts as a premium service now.
Although, I wouldn’t expect anyone to not work with the police, even if you pay.
Which is why providers should not persist things about their users.
This is the only acceptable direction now.
This is the only acceptable direction now.
Why should you care whether or not the service will work with the police? Seems specific
Police may be enforcing laws that you consider immoral. There's lots of current laws I would consider immoral. (requests like "Confirm whether this person searched for abortion clinics")
On the other hand if any business exists as a legal, registered entity, of course they will comply with local LE requests. Not sure who expects otherwise.
On the other hand if any business exists as a legal, registered entity, of course they will comply with local LE requests. Not sure who expects otherwise.
neeva is not respecting GDPR.
They have checked checkbox newsletter by default. Moreover, why they need a name for registration?
They have checked checkbox newsletter by default. Moreover, why they need a name for registration?
Are there alternatives that some of you would recommend in this sense?