3 ms·

I much prefer to trust an implementation over someone's word. I use Signal because I agree with their stated goals, and their implementation seems to align with

I much prefer to trust an implementation over someone's word. I use Signal because I agree with their stated goals, and their implementation seems to align with those goals. I don't use Facebook because their policy is awful in the first place, and implementation is nonexistent. Kagi is in between. I am happy with their policy, but their implementation is lacking. This wouldn't be a problem, but they don't seem to admit that their goals are fundamentally in conflict with their mode of operation. This is the first item on their FAQ: "Our goal is amplify the web of human knowledge, creativity and self-expression and provide the user tools to fight against the web of greed, ad-tech and user tracking." To pretend that company policy is the best way to achieve that goal is insulting. We will never have an ad-tech and tracking-free future without technological measures. Offering a service on the basis of "Pay with your money, not your data" is a fantastic step in the right direction, but obviously not enough.

freediver 4y ago
Challenges in front of us are monumental, and we prioritize features based on impact to user base. It seems to be the case that full anonymity is not very high on the list for most Kagi users (but best search results in the world are, so that is where we spend most of our limited resources on). It is also not immediately clear how can anonymity be guaranteed from a technical implementation standpoint? Happy to understand more about what the ask is. We do say that our business model is the best guarantee - we make money from our users not third parties; our volume is too small to be interesting to anyone and will always remain small because of the paid nature of the product (in fact we are actively making decisions like increasing prices that are making it even smaller); if we ever sold a byte of data and that was found out, it would immediately kill the entire business; and we do not log searches because we have no use of them, they would only be an uneeded cost and liability. The motive simply isn't there. We understand that you have to believe that, but so you would need to believe any other claim including the full anonymity one. Nothing in the world can 100% guarantee that. As it is we carry the full burden of lack of trust in online technology, that we are not responsible for creating (and have actually founded to be an anti-thesis to). Thanks for supporting Kagi, and we ask for a bit of patience with this, we just started.

freediver 4y ago

Challenges in front of us are monumental, and we prioritize features based on impact to user base. It seems to be the case that full anonymity is not very high on the list for most Kagi users (but best search results in the world are, so that is where we spend most of our limited resources on). It is also not immediately clear how can anonymity be guaranteed from a technical implementation standpoint? Happy to understand more about what the ask is. We do say that our business model is the best guarantee - we make money from our users not third parties; our volume is too small to be interesting to anyone and will always remain small because of the paid nature of the product (in fact we are actively making decisions like increasing prices that are making it even smaller); if we ever sold a byte of data and that was found out, it would immediately kill the entire business; and we do not log searches because we have no use of them, they would only be an uneeded cost and liability. The motive simply isn't there. We understand that you have to believe that, but so you would need to believe any other claim including the full anonymity one. Nothing in the world can 100% guarantee that. As it is we carry the full burden of lack of trust in online technology, that we are not responsible for creating (and have actually founded to be an anti-thesis to). Thanks for supporting Kagi, and we ask for a bit of patience with this, we just started.