The Most PRIVATE Search Engine (2021)
And to circle back to web search, I got word that when netscape launches this week, yahoo will be the default directory. They've got their own button and everything what the hell is yahoo. Two guys stanford used to be called jerry and david's guide to the world wide web. These guides to the world wide web are called search, engines and they're. Like directories that list all the websites that exist websites are hosted on servers all over the world and when you use the web to visit one of these sites, your computer needs to know the specific address of where the website's information is located in order to retrieve it in the early internet days, it was really difficult to find all these websites, because there were no directories, you had to either type in an exact address or click on a link, or you wouldn't find the page.
So people started to create handmade lists of websites that they discovered he suddenly thinks he should index every website in existence. These were the earliest indexes. Someone just sent me a new link to put in our index he's been meticulously building a list of urls by hand that sounds awfully tedious. Then things called web crawlers were invented which were like little spider robots that crawled across the web and generated an automated index of all the websites. They could find it's a search algorithm, basically indexing the web, making it searchable this completely revolutionized how we access the web. Now, if you want to find a website, you can type in keywords, and these search engines will show you a list of websites that match your criteria.
It's the first site. You visit every time, you're online, it's the page that gets you where you want to go. While a browser is like a car that takes you to the website. A search engine is the map that shows you what websites are out there. The most popular search engines are bing, baidu, yahoo and, of course, by far the most used search engine. Google, almost 90 percent of internet searches in the united states use. Google. Google has become such an important part of our lives that we turned their name into a verb. You should google it, but how? Private are your searches when you use these engines?
Short answer is not at all your search engine collects your ip address. Information about the web, browser that you're using your location, search, queries and unique identifiers to your machine which are stored in your browser cookies. Many search engines are massive data collection tools by using them you're giving these companies a deep granular insight into who you are as a person, including where you're from your interests, your medical concerns or political views. This data is used to build personal ad experiences for you, but these companies also sell your data. The data can also be used to target you as an individual and manipulate and censor your internet experience. We have reason to believe that google is knowingly deliberately strategically manipulating people's thinking and behavior from the very first character people type into the search box, and, let's not forget that the government also has direct access to all this information thanks to programs like the nsa's prism program.
So, what's the solution, you should never ever use google.com. Never because it tracks you uh, you should use uh either something like duckduckgo or my favorite is called start page. That's pretty good advice, let's dive into some of these more private alternatives when choosing a search engine. There are lots of trade-offs to consider search. Engines like google are popular precisely because they return results faster and provide the most relevant search results. Their algorithms know so much about you that they're able to predict what you're actually looking for in your searches and they prioritize these results on top, while hiding less relevant results. But this also means that what you're shown is carefully controlled and can support any bias of their choosing.
Your algorithm suppresses negative search suggestions for one candidate, but you allow negatives to appear now and then for the opposing candidate. It's also worth noting, as we explore different options that there's a difference between a pure search engine and a meta search engine pure search engines, crawl, the web to build their own database of results and meta search engines rely on other search engine sources to produce their results. Using meta search engines can give you privacy without affecting the quality of the results, but meta search engines are still subject to the bias or even censorship of the search engine. It pulls. Data from some search engines use a hybrid approach, both trolling the web, for their own results and using other sources.
The second thing to keep in mind is that software that touts privacy as a boasting point, should be open source to prove it. Open source refers to software with source code that a user can inspect, but while being open source is necessary to be able to trust the privacy of a software. It isn't sufficient because the code needs to be simple enough, that you can reason about it or it needs to have undergone comprehensive third-party audits by those who do understand it. So keep in mind that open source isn't a panacea to cure your privacy woes. Let's start with duckduckgo, it's probably the most famous privacy-focused search engine with over 3 billion searches in 2021.
Unlike the big search engines, it doesn't collect ip addresses or user information, it does store searches, but not in a personally identifiable way, and they do this to improve things such as misspellings duckduckgo is both a search engine and a meta search engine getting its results from over 400 sources like wikipedia being yandex and yahoo. It makes its money from advertising and affiliates the ads are not personalized to you, but are delivered based on your search term. Only another important aspect of duckduckgo is that, because it doesn't track user behavior, it also delivers the same search results to all of its users. This makes such search results more neutral and prevents the issues that arise from personalized targeting.
However, this doesn't mean that search results are without bias, as it draws its results from other sources. It can still be subject to censorship. This happened recently when an image search on the iconic tank man returned no results on duckduckgo tankman shows an unknown person blocking a line of tanks in beijing the day after the tiananmen square massacre, and the image is banned in china on the anniversary of tiananmen square. This year. The image disappeared from bing's image and video search results, because duckduckgo relies primarily on bing for its image search. The image also disappeared from duckduckgo. This illustrates the issue of being heavily reliant on other search providers and in response duckduckgo's ceo said that they were now looking into adding additional sources.
It's also worth noting that dug go is only partially open sourced and isn't very upfront as to which components are open sourced or not. They say it's, so they can remain competitive and prevent spam, which is a fair point, but it would be better to be upfront about which parts are closed. Even so, duckduckgo remains a solid search engine and is a staple in my internet search activity. Metagar is an open sourced, privacy-focused search engine based in germany and funded by user donations. It uses the hybrid model of combining the search results from its own web crawler with those of other search engines. It makes your search query anonymous and then passes it to various search engines.
It also has an integrated proxy server which allows you to view websites anonymously. The receiving website and other third parties only see metagas proxy, rather than your ip address. It's also available as a hidden tour service for maximum privacy and it doesn't use any tracking cookies medicare does record your ip address and timestamp for a maximum of 96 hours, after which it's automatically erased. They say that their reason for doing this is to limit the number of search, queries per internet connection from our run-through of it most of metagail's search. Queries were delivered from bing or scopia, meaning that, although it promotes itself as hybrid, it still relies heavily on other search engines, but it's a pretty interesting option.
Of course. The best way to ensure search engines aren't logging. Your data is to host the search engine yourself enter crx, an open source, self-hostable engine. It's a meta search engine that sources its results from places like google, yahoo and bing, but crx makes sure it does this anonymously and does not share the user's ip addresses or search history with any of these other search engines, and it also blocks tracking cookies. Another benefit is that all search results give you a direct link to the respective site rather than attract redirect link and when available, these direct links are accompanied by a cached or proxied link that allows you to see the results page without having to visit the sites themselves with your unique identifiers.
The case links point to a saved version of a page on archive.org, while the proxy links allow you to view the current live page via a cx-based web proxy crx also has tabs to filter your search within specific domains such as images, maps, music news, videos and social media. Now, if you don't want to host your own crx instance, because you think it's way too complicated, you can use any of the many public instances available. However, that means trusting that particular public instance not to log your searches or requests. Crx also does get blocked from using google from time to time as it scrapes its results, but, for the most part it works.
Fine quant is a pure search engine launched in 2013 and based out of paris, france, it's one of the few eu-based search engines, and it's also one of the few on this list. That is a true search engine rather than just a meta search engine, as it relies on the company's own algorithms and indexes it processes over 10 million searches per day worldwide and has three unique home pages to begin a search. The non-personalization of search results means that all searches appear the same, delivering a more neutral point of view. Rather than trapping users in a filter, bubble, search, queries are encrypted and your ip is also disassociated from your searches like other privacy-focused search engines.
Quant doesn't use your search history to help deliver results as it doesn't retain user data, but this also means your previous searches aren't saved or remembered, which is the sacrifice made for additional privacy. One thing worth noting, though, is that quant does share some anonymized data with microsoft, to deliver contextual advertising based on your region and what you type into the search so take that how you will based out of the netherlands, the start page search engine allows users to obtain google search results, privately, removing trackers and without storing any search data. They have an anonymous view, browsing feature like crx, which allows users to search the web by proxy and not reveal unique identifiers.
They are also constantly innovating with apps users will be familiar with from google, but which are crafted to protect their privacy. Some examples include a private language, translator, private stock search, private currency, converter, private shopping and a region filter to let users customize their search results. Now there has been some controversy over start pages funding and whether it conflicts with their promise of privacy. In 2019, they received a considerable investment from a subsidiary of system, one an advertising company that once said, if we can gather as much data as possible, give it off to our engineers and data scientists and then manage the tool effectively. The business can quickly scale system, one is also an american-based company and in the us there are no comprehensive privacy laws like in the eu.
Now, when I reached out to start page, they assured me that the start page founders may unilaterally reject any potential technical change that could negatively affect user privacy. They also said that start page continues to be headquartered and operated in the netherlands. Ensuring all of our users worldwide are protected by dutch and eu privacy laws. This will ensure compliance with the european gdpr and dutch avg and will provide protection from patriot act. Regulations. Startpage remains passionately driven by the mission of providing quality, unbiased search results while respecting online privacy and never storing consumer data. Uk-Based mojig has been around since 2004, and their search results come from their own index of web pages that they created by crawling.
The web mojic has already indexed over 4 billion pages and retains its commitment to be independent from big tech. It also, interestingly, has an emotional search function where it categorizes content, using deep learning to five different emotions, love laughter, surprise, sadness and anger, allowing users to, for example, filter out sad news items, mojit doesn't implement user tracking and ip addresses are stripped and replaced with only a country code. The only time it does record your ip address is if the search query relates to illegal and unethical practices relating to minors, like other pure search engines on this list. Mojic's strength is also its weakness. Without relying on other search engines, it can claim to be more independent from big companies such as google or microsoft, but it also means that search results may not be as relevant, but testing out the search engine ourselves demonstrated reasonably accurate results with quite a different emphasis than google or microsoft.
Yasi is another open source search engine with a cool twist, and then it works on a peer-to-peer model. A yasi peer search will independently crawl through the internet, analyze and index any web pages. It finds and then store these results in a common index, which is shared with other yasi peers using p2p because of its p2p nature. Users are required to download software before using it to protect your privacy after performing a search. The words used are sent to appear in the form of distributed hash tables. Peers, then store crawled search results as cryptographic hashes, and these are all mixed in between peers, making it impossible to pinpoint search queries to a certain host.
The benefit is that there's no need to erase logs because there are no logs and there's no need to rely on a third-party server to run private search queries unlike google or bing, where the company managing the search results is open to subpoenas with yasi. There is no central authority, but instead thousands of servers in multiple countries providing results. This also means that yasi results can't be censored now. Unfortunately, while its architecture and p2p nature are laudable, its search results in our experience for the weakest among those tested, and there is a noticeable delay when searching, but I think that decentralized tools like this will become ever more prominent in our future as web 3.
0 evolves and that future is very exciting. To me, brave search is a newcomer with its beta launching in june 2021. It comes off the heels of brave's acquisition of tailcat, an open source search engine that forms the foundation of brave search brave, doesn't track you or your queries, nor does it log your ip or geographical location, though you can choose to search just your own region based on your ip, which is stored locally, without sharing with brave results, are delivered from its own index to ensure neutrality in its beta form, it still anonymously checks. Its search results against third-party results, such as google and mixes them in to improve results.
Brave also provides a results. Independence metric to show the percentage of search results that come from brave versus those from third parties. Brave search is not open sourced at this time and it's still in beta, so there are undoubtedly going to be bugs to iron out as it's tested, but search results were of a high quality. So what's the verdict boys, the verdict is many of these search engines. Do a lot of the same things and the reason why you might select one over another could literally be as simple as liking. One small feature over another choosing. Any search engine is like choosing a restaurant. Each one claims to be the best and will boast about its specials, but at the end of the day you should be the one to read the menu and choose what best suits your palette.
I got a whole new menu just for you, but if it's privacy you're looking for then, hopefully this video has given you a taste of what's on the menu.