Re:DuckDuckGo removed the ability to filter search results
by 1vuio0pswjnm7 on Monday April 24, 2023 @03:56AM
(#35684821)
"Kagi Search includes anonymized requests to traditional search indexes like Google and Bing as well as sources like Wikipedia, DeepL, and other APIs."
If many users all send their search queries through a proxy that forwwards them to, say, Google, then one can argue these queries are "anonymised" to some extent.
This has been tried. But AFAIK, no one has ever tried to charge money for it.
In 2003, there was Scroogle which lasted for almost 10 years and was reputedly processing hundreds of thousands of searches per day.
Scroogle used to have its own Wikipedia page before it was deleted then merged into the "Criticisms_of_Google" Wikipedia page.
In 2010, there was GoogleSharing, a Firefox plugin. It did not last very long and unlike Scroogle I cannot recall any explanation was ever given why it was shut down.
Criticisms of Scroogle or GoogleSharing might include having to trust the people operating the proxy servers.
For example, if operator gets a strongly-worded letter on law firm letterhead then will he share a user's search results. What if he receives a directive from some legal authority demanding that a user's queries be monitored and/or preserved. Will he notify the user. Will he fight such requests. How good are his lawyers. Does he even have any that are prepared to handle these matters.
All these issues apply to Kagi as well.
Being a paid service might actually improve the trust problems, assuming there is a contract with the user. The last time I looked for Kagi's terms, it was not reassuring. What does Kagi promise to do or refrain from doing. If Kagi agrees to zero liability for almost anything, then to me that's a red flag. Why would anyone trust him. At the very least, users should be able to sue for up to the amount of the fees they have paid if Kagi breaches the agreement.
NB. I am not for or against Kagi. I like the proxy server idea and wish there were more options to use proxies. However I do not agree that merely paying for web search solves all the problems. I wrote scripts for own use that search multiple indexes, filter out garbage and combine results into a simple, minimalist SERP with unprefixed URLs suitable for a text-only browser. For me, that addresses the paid placement ads and ranking algorithm problems. As for the privacy problems, I cannot find anything that promises Kagi will not collect data about users. The same problem of trust as in 2003 through 2012 exists today. Arguably things have gotten worse. As we've seen from so-called "tech" companies offering "pay to stop the ads" subscriptions, paying fees does not stop the data collection. It does not solve the privacy problems unless the search engine operator is actually bound by an agreement or laws that prohibit it from collecting data or resticting its use. And even with that, how does one ensure compliance.
If many users all send their search queries through a proxy that forwwards them to, say, Google, then one can argue these queries are "anonymised" to some extent.
This has been tried. But AFAIK, no one has ever tried to charge money for it.
In 2003, there was Scroogle which lasted for almost 10 years and was reputedly processing hundreds of thousands of searches per day.
Scroogle used to have its own Wikipedia page before it was deleted then merged into the "Criticisms_of_Google" Wikipedia page.
https://www.cnet.com/news/privacy/privacy-what-should-google...
https://en.wikipedia.org/wiki/Talk%3AScroogle
In 2010, there was GoogleSharing, a Firefox plugin. It did not last very long and unlike Scroogle I cannot recall any explanation was ever given why it was shut down.
https://www.forbes.com/sites/andygreenberg/2010/10/04/a-bett...
Criticisms of Scroogle or GoogleSharing might include having to trust the people operating the proxy servers.
For example, if operator gets a strongly-worded letter on law firm letterhead then will he share a user's search results. What if he receives a directive from some legal authority demanding that a user's queries be monitored and/or preserved. Will he notify the user. Will he fight such requests. How good are his lawyers. Does he even have any that are prepared to handle these matters.
All these issues apply to Kagi as well.
Being a paid service might actually improve the trust problems, assuming there is a contract with the user. The last time I looked for Kagi's terms, it was not reassuring. What does Kagi promise to do or refrain from doing. If Kagi agrees to zero liability for almost anything, then to me that's a red flag. Why would anyone trust him. At the very least, users should be able to sue for up to the amount of the fees they have paid if Kagi breaches the agreement.
NB. I am not for or against Kagi. I like the proxy server idea and wish there were more options to use proxies. However I do not agree that merely paying for web search solves all the problems. I wrote scripts for own use that search multiple indexes, filter out garbage and combine results into a simple, minimalist SERP with unprefixed URLs suitable for a text-only browser. For me, that addresses the paid placement ads and ranking algorithm problems. As for the privacy problems, I cannot find anything that promises Kagi will not collect data about users. The same problem of trust as in 2003 through 2012 exists today. Arguably things have gotten worse. As we've seen from so-called "tech" companies offering "pay to stop the ads" subscriptions, paying fees does not stop the data collection. It does not solve the privacy problems unless the search engine operator is actually bound by an agreement or laws that prohibit it from collecting data or resticting its use. And even with that, how does one ensure compliance.