Some Surprising Things About Duckduckgo
Key topics
The debate rages on about DuckDuckGo's claims of not censoring search results, with some commenters calling foul and others defending the company's stance. At the center of the controversy is the company's assertion that it doesn't censor results, with critics arguing that this is "technically-bullshit" since they rely on potentially censored sources. CEO Gabriel Weinberg weighs in, clarifying that his company doesn't censor results itself and actively monitors for anomalies, inviting users to share specific examples of suspicious results. As the discussion unfolds, it becomes clear that the real issue lies in the nuances of what "not censoring" really means.
Snapshot generated from the HN discussion
Discussion Activity
Very active discussionFirst comment
22m
Peak period
57
0-6h
Avg / period
15.2
Based on 137 loaded comments
Key moments
- 01Story posted
Dec 13, 2025 at 4:57 PM EST
26 days ago
Step 01 - 02First comment
Dec 13, 2025 at 5:19 PM EST
22m after posting
Step 02 - 03Peak activity
57 comments in 0-6h
Hottest window of the conversation
Step 03 - 04Latest activity
Dec 16, 2025 at 8:38 AM EST
23 days ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.
Please don't post like this on HN. It breaks multiple guidelines. https://news.ycombinator.com/newsguidelines.html
Maybe you meant to type "fluff piece"? That could be a matter of opinion.
Either that is completely bullshit, or it’s technically-bullshit.
1. They don’t have to censor because their sources censor for them. “Oh we’re just an aggregator of censored results” doesn’t mean “this is an uncensored search engine” like the claim would have you believe.
2. Proof of this is evident in by comparing Russian yandex.com, my now go to for anything related to hacking, pirated anything, topics of censorship or controversial discussion, even “legit” but rarer information like how to train or use X or Y AI model, etc. The domains that appear on yandex remind me a time gone by. Like image search before Pinterest, unreliable but not sterilized.
3. I use DDG everyday. In the last year or so, I have found myself going to Google, Bing, Brave, Yandex, SearX, and other more than ever. The quality of DDG has for me, unquestionably slipped. I have a strong distaste for Google, and have used them this year more than ever.
They are not uncensored, although maybe they allow that burden to be done for them to keep their nose high in their air on the topic.
However, I fear it may be a moot point as I find myself looking elsewhere often now.
Why do you feel that the actions of a search provider(s) should be reflected so negatively and angrily on the aggregator?
If you see results missing, I'd be happy to look into them. My email is in my profile.
I mean… that’s exactly my issue. That’s just another way to say “we present censored results”.
To be fair… my real issue is the last year or so the results have been noticeably sub-par for me.
I knew there would be no point discussing a subjective matter like that. So I brought up statement that I found misleading.
If the source has not put something on the web... how is anyone supposed to get past that? Or am I misunderstanding your statement?
All search engines got so much worse in the last years - it is so sad. We lost some of our knowledge that way.
This already started before AI, but AI further reduces the quality now.
As someone using Duckduckgo a lot, I feel the opposite sometimes regarding google :>
I still use google occasionally for its google search image option but yea thats about it
While it seems DDG is on the same path of AI / chat centric search UX, at least they allow me to turn off all that stuff. But... search has gotten so bad in general, DDG is having the same results issue I had on Google. I don't see DDG as a player in the Ai space so I think my usage will only decrease as search result quality continues to decrease.
I am hopeful in the long run that search index / results will become better as the core UX for most people becomes chat, search result pages become low human traffic (meaning ads are worthless), and search becomes one of many research tools for to the agents
I'm now looking for APIs to integrate with my custom / personal agent setup. I'm done outsourcing my UX to Big Ai/Tech. I don't think we should repeat the same mistakes of outsource a core human/digital UX to Big Ai/Tech. We (HNers) complain so much about all the bad stuff the prior iterations (social media, saas out the wazoo), are we going to repeat it again by defaulting to whatever they give us, misaligned incentives and all?
If you’ve got screenshots or details on your personal masterpieces I’m interested to see it.
The neutral reviews of Kagi I have seen don't lead me to believe there is some superior quality awaiting me if I just give them the money
> If you’ve got screenshots or details on your personal masterpieces I’m interested to see it.
Saying something like this, in this way, is not constructive and not how to bring people towards your way of seeing things. You, specifically with this comment, make me even less likely to use Kagi, and pay less attention to things you may say in reply or otherwise in the future
You may have misread the thread. I'm saying "I'm done outsourcing my UX"—the very thing that makes both DDG and Kagi make unsuitable to you—sounds like you're building something novel, even if it's exclusively to scratch your own itch.
If its ever in a place where you'd be open to sharing, that's the sort of thing I come to HN for.
[0] https://html.duckduckgo.com/html
[1] https://lite.duckduckgo.com/lite
I would pay DDG if they gave me an API for search, ideally pay-per-request. I'm not paying them for Ai, I can get that much better elsewhere
Agreed that a DDG API would be pretty great, though.
SERP, EXA, Tavily... they are becoming a dime-a-dozen with the Ai hype b/c (1) making the process of fetch/scraping easier (2) being the primary end user of search API (3) making the creating of such an API easier
I haven't dug in enough to any of them to get a sense of result quality or consistency
I would prefer if there was a way to have dark mode or similar things in it tho. It was really really fast.
You can search "!w Gabriel Weinberg" and it will open the Wikipedia article because of the leading exclamation mark and w. If a site changes their search url, you can submit the precise new pattern they should use for a redirect. If a new service pops up, you can use the same form to request a new search prefix. These form submissions could give someone at DDG an easy interface to verify quickly and approve or reject them.
These form submissions get ignored and have been for years at this point.
You can choose keywords that don't start with !, so typing them is easier than using Duck Duck Go's bang feature.
So if, for example, you wanted to make
> x <search_term>
and
> y <search_term>
both work the same, x and y being letters from 2 different alphabets but mapped to the same keys, you couldn't, without some JavaScript. If you just added those 2 keywords, even if you manually edited or created your bookmarks, one bookmark would override the other and the other would appear empty with no keyword.
The workaround I found was using a bookmark with this code in it (instead of the usual URL):
javascript:(function(){var keywords="%s";var mainURL="https://<URL>/";var searchURL="https://<URL>/<params>-"+keywords;if(keywords==""||keywords=="%"+"s"){window.location=mainURL}else{window.location=searchURL}})();
Where https://<URL>/<params> is something like https://example.org/search/q-.
It's slower and sometimes doesn't work if you type "y" and then the query too fast, especially if you're pasting the query. So sometimes it doesn't work and searches with the browser's default search engine for "y <query>".
Being able to put language-specific shortcuts on keys that change with the keyboard layout is damn useful.
Why does software have such powerful modes for Python and JavaScript, but never for French or English?
https://en.wikipedia.org/wiki/Special:Search?search=%s
https://en.wikipedia.org/wiki/Special:Search?&search=%s
Each of those can have a different keyword.
!gw - Google web-only search (&udm=14) - https://udm14.com
!kagi - Your best competitor. I like to switch back and forth and compare. They have !ddg which goes to you, so it's only fair.
(I have submitted these, I think more than once, but no response.)
btw - I enjoyed collaborating with you years ago to hack a way to get DDG working in Safari, before it was official. If I recall, you still had servers running in your basement at the time. I was really impressed with the lengths you were willing to go to make this work.
DDG now enjoys a privileged position as one of only five selectable search engines in Safari, which makes it a priceless gateway of sorts in a browser that is otherwise extremely locked-down and hostile to user customization. This is one reason why I think the bangs are really important.
https://search.vale.rocks
Like till now If I wanted to search something on reddit from duckduckgo, I would search "<search query> reddit"
But it was also an hit or miss sometimes so you are telling me that snaps can just @r <search query> and guarantee its from that is amazing!
Your list of resources feel good too, https://time.fyi and other tools are good too!
I would love it if your resources also included open source resources similar perhaps as I prefer open source tools mostly but even these resources are good too so thanks!
Just for anyone else who isn’t aware, the bang commands can be anywhere in the search string, and need not necessarily be at the beginning.
All these queries will take you to Wikipedia for the term:
"!w Gabriel Weinberg"
"Gabriel !w Weinberg"
"Gabriel Weinberg !w"
Many a times when I find the default DuckDuckGo search results inadequate and want to go to Google search, I just put a “!g” as a separate term anywhere within the search string and hit enter. This is especially useful on mobile where the search string may be a lot longer than the visible text box and I can’t be bothered to move the cursor.
The difference is especially noticeable for tasks like editing / deleting / selecting specific parts of long URLs, and on smaller phones where the iPhone space bar is smaller than on larger phones.
Hmm, when I added !mt more than a decade ago it went live almost immediately...
https://0x0.st/K4kE.png
Anyway the DDG html search site is next to useless now, and I'm fed up of the AI nags, so I'll be right behind you once I have searched for my coat.
Suspicious as heck to have enough money for supporting +300 employees plus all other operating costs without an obvious money cow for those costs.
Rather use Qwant, Brave or even Ecosia.
Who is "we"? Don't they get their results from Bing?
Eg, DDG always fail the "watch (specific movie or tv show) online" search query test. Many other search engines like Bing and Google also fail. It's a quick censorship influence test. And DMCA takedown requests have a clear track-record of being abused.
One search engine that succeeds is Russia's Yandex. I'm sure they censor plenty of things (eg, material sensitive to Russia), but that censorship set may not intersect with that Google, Bing and DDG's sets.
DDG results are mostly Bing results, so if a page doesn't show up on Bing, it probably won't on DDG either. That doesn't mean DDG themselves censored the results.
In terms of money, as the article notes we have 3% of U.S. search market share. That's a lot if you consider how much Google makes. Now, in part because of our search privacy, we make way way less, but it is still enough to be profitable. That said, that means we could be way way more profitable if we tracked people, which we don't.
That is not a serious review.
There is no way you are supporting +300 people per month, which equates to a minimum of 36 million USD per year without even counting infrastructure expenses, fancier salaries for management and associated costs for business building.
By all means feel free to present more detailed documentation about your expenses and sources of income. I'll wait, will even grab a chair to avoid getting tired after waiting a few more years.
All dodgy.
Also, NAD is a very serious review process as NAD refers any unsubstantiated claims directly to the FTC. Lots of evidence, vetted by a third-party.
Do you have plans to build an email service without the tracking? Would love to hear thoughts about this. There may be users willing to pay a small monthly fee for this.
It isn't full email, but removes trackers in your email before it gets sent to you, which is the most privacy-invasive part of email. More info at https://duckduckgo.com/duckduckgo-help-pages/email-protectio... and
I really like Kagi, becuase I can pay for the search and my searches aren’t being leaked to third parties.
For about first 5-10 years of its existence DuckDuckGo also promoted their use of Perl, and afaik they contributed to Perl development.
I don’t like the duck.ai interface much (choosing a different LLM is not easy once you’re already in a conversation), but I use it a lot more than I use the DuckDuckGo search engine (the results from the latter aren’t great).
Just like with DuckDuckGo search, where I start a search and then use the !g bang command to go to Google for better results if needed, I try duck.ai and then move to ChatGPT (without any account) when even the best models in duck.ai aren’t good enough.
For most simpler queries though — where I’m just looking to learn a bit about something as opposed to finding a solution for a specific (more complex) question or problem — duck.ai with its GPT 5 models are more than adequate (even the 4o mini is fine).
Its just the right amount of AI with all the other things and I can have a lot of freedom/customizability/block AI and they provide subdomains for a lot of things (I found out about noai.duckduckgo.com from here and other things too) and overall feel like its one of the best search engines.
I wish if they could create their own index tho because I do not trust microsoft so much.
I wonder why people still use google when there is duckduckgo. I suppose monopoly might be the answer but I wish if there was more awareness about duckduckgo.
They don't do that. That is a story that comes from a case of Bing, their upstream results provider, doing that, before quickly reverting the block.
Bing is just one of their sources. They run their own crawler. They source data from multiple 3rd party providers.
> I searched torrent sites on DDG ... it shows me torrent sites links
You didn't post your search so here is DDG results for widely used torrent sites (per reddit).
Interestingly, I did find that DDG does not scrub russian torrent sites from it's results. Which would make sense if DDG is trying to compete with Yandex.Because Yandex doesn't preemptively censor torrent sites, like DDG does.
Although I still agree that this is interesting discovery but I didnt know that these sites allowed to be indexed in the first place and that the issue was from ddg site, I always thought it was the opposite or didnt really give much too thought into it but interesting, I would love to discuss more about why you might think so this is the case or how it can meaningfully impact I suppose.
I think that the issue here could be that there might be some censorship from big upstream or that ddg does it to prevent any legal issues which I am sure can open a load ton of questions if you scrape something tangentially related to copyright/similar so it might be understandable why they might do this I suppose.
For us users, the point of search engines is to index the context of sites and tell us which one(s) have the info we're looking for.
There isn't much point to a search engine that just tells us where a site is. Even less to tell us for a site we already know.
This hash search gives one site: https://duckduckgo.com/?q=%229623748B411A3CD02DE8F332820C4C7...
This way also gives one torrent site: https://duckduckgo.com/?q=all+quiet+on+the+western+front+193...
[1]: https://1337x-status.org/
yes, yes, I know why it's named like that
Is a version of DuckDuckGo without Javascript. Very fast and compatible with minimalistic web browser like lynx.
Also https://html.duckduckgo.com/html/
Awesome stuff.
I was able to run modern firefox on that 1 gb puny laptop too but it took 800 megabits of ram but I was able to run https://pomodorokitty.com/ on it.
Although I occasionally have to use google for https://images.google.com/
Is there any way that duckduckgo can have something similar or perhaps there already is and I am not aware of it, either way, I would love to know more about it and have a nice day!
What I meant by the abilities of image search was being able to provide an image as a search query and then it can find similar images basically.
Like if I have a picture of wikipedia's logo (as an example) in my clipboard or as a png saved, is there any particular way I would be able to upload that picture from my clipboard/my desktop to duckduckgo and then duckduckgo can query for similar pictures basically similar to the camera feature of images.google.com which has this camera option where I can paste the clipboard image
This is honestly one of the biggest reasons that I ever use google (which is rarely nowadays but still) and I would love to use duckduckgo for this use case.
Sorry but I am unable to find it and I would love to know more about it/ ideas on it. And have a nice day sir and thanks for responding to my message! :D
I love DDG! However it's search term autocorrect is really broken when DDG decides for me I must not be looking for exactly what I typed.
It seems to assume that if there isn't a lot of results then it's my fault for typing a "bad" query? Sometimes it makes up a new search out of whole cloth in this case with terms not even close to what I typed.
I get the utility for detecting+correcting typos but DDG often takes this way too far. Posting this in friendly criticism, happy user otherwise!
I avoid DDG and all search engines that do this. Forcing me to manually sort through garbage because you didn't want to render an empty page is a huge waste of my time. If you don't respect my time, I'm just not gonna interact with your service.
Do better.
For context, my team lives in a timezone ~17 ahead of the parent company and they seem to really have difficulty grappling with the idea of not scheduling meetings with me for three in the morning.
(Seasonal related practices like taking holidays during our summer in the other hemisphere also seems to be perplexing.)
[1] https://duckduckgo.com/how-we-work
Funny how people are still making jokes about yahoo search, when it works better than google.