“Should we not be buying VW, BMW, Siemens and Bayer technology and products today because they participated in holocaust and directly collaborated with Hitler?” – CEO of Kagi when given feedback re: Brave partnership

  • noodlejetski@lemm.ee
    link
    fedilink
    arrow-up
    58
    arrow-down
    4
    ·
    edit-2
    10 months ago

    their CEO also moderates the discussion on Kagi’s Discord. he’s been removing criticism by queer folks, while - the last time I’ve checked before I’ve left their group - keeping replies like “stop shoving LGBTQ down our throats”.

    https://nyan.lol/@zicklepop/111716138453426218

    edit: also their response regarding a request to add a “don’t do it” widget to suicide searches basically boils down to “but if we make a moral choice now, we’ll have to do more moral choices in the future!”, which is… suboptimal. https://kagifeedback.org/d/865-suicide-results-should-probably-have-a-dont-do-that-widget-like-google/

    • Deceptichum@kbin.social
      link
      fedilink
      arrow-up
      29
      arrow-down
      2
      ·
      10 months ago

      Was thinking about biting the bullet and getting Kagi, but this is sign enough not to let alone the OP.

    • hydroptic@sopuli.xyz
      link
      fedilink
      arrow-up
      29
      arrow-down
      2
      ·
      10 months ago

      Ah fuck, I’ve been using Kagi but of course their CEO is a reich-wing asshat.

      Anybody know of any other options? I’m not going back to Google.

    • dmnknf@lemmy.ml
      link
      fedilink
      arrow-up
      12
      ·
      10 months ago

      Wow, this topic in the edit is bad, and I wasn’t aware of it. Just canceled my subscription. Now I have until 14/01 to choose an alternative.

          • ⲇⲅⲇ@lemmy.ml
            link
            fedilink
            arrow-up
            4
            ·
            10 months ago

            For me, it is doing a good job, and it’s pretty fast. I think in the past was slower and with many issues, but it improved a lot. I never tried Kagi and I don’t think it’s rational to “login” and identify myself to be able to get results, even if it’s working better, everyone should be able to get good results without needing to pay.

            • sudneo@lemmy.world
              link
              fedilink
              arrow-up
              4
              arrow-down
              1
              ·
              10 months ago

              everyone should be able to get good results without needing to pay.

              Until this stuff is funded with public money, it’s not really doable for such a compute and storage intense task.

              I am perfectly OK with paying for good software, until then. I also agree with the principle of aligning interests of users and the search provider by having the users pay. Other models (ads, sponsoring) creates incentive to favour those who pay. The other reasonable model is donation, that can work potentially, but it has its problems.

              • ⲇⲅⲇ@lemmy.ml
                link
                fedilink
                arrow-up
                3
                ·
                10 months ago

                But https://searx.space/ is like Lemmy, you can donate to those instances to help them to keep it on. It is not really a search engine, so the power usage isn’t that big, it uses other search engines to get the results, the difference is that the search engines like Google, DDG, Bing and etc don’t know who did those requests. The quality isn’t bad neither, but I can’t really say the difference as I never tried Kagi. 🤔

                • sudneo@lemmy.world
                  link
                  fedilink
                  arrow-up
                  3
                  arrow-down
                  1
                  ·
                  10 months ago

                  In fact it’s not comparable, because this is a metasearch engine. Kagi has quite many unique features and besides that it’s great in surfacing small websites (for which it mostly uses its own crawler) and downranking pages full of tracking. They are just different and the Kagi model is the most reasonable, in my opinion, for what it does (search engine).

    • Nate Cox@programming.dev
      link
      fedilink
      English
      arrow-up
      12
      arrow-down
      1
      ·
      10 months ago

      God damn it.

      I am a paying subscriber to Kagi because the search results are excellent and there are no ads, so of course you show me a thread on “we should maybe add a small message to suicidal users telling them there is help for them” which then reads like a truth-social propaganda thread, filled to the brim with “helping people is a slippery slope!!! muh freedoms!!” arguments.

      • sudneo@lemmy.world
        link
        fedilink
        arrow-up
        7
        arrow-down
        13
        ·
        10 months ago

        This is something I really don’t get.

        • it is unclear whether anybody in history has ever been helped by that kind of message.
        • it is kind of a religious morality that suicide needs to be prevented and that if someone wants to do it, it’s because they are not in control. This doesn’t mean it’s wrong in absolute sense, but it’s very opinionated.
        • realistically speaking, there is no need to “search” how to suicide.
        • trying to conclude what you want to do, rather than what you want to know (I.e. search) is IMHO exactly against what kagi’s idea is. It’s a service that does only what it is asked for, and doesn’t try to “know” you, as a customer or user. No text editor prompts you to suicide hotlines by analyzing the text you are writing, and we would consider it extremely weird if it did. However, with search we get used to the tool trying to guess what we want to do because Google does know you, I think the beauty of Kagi is going in another direction.

        But let’s assume that all the previous points are invalid, and - for a greater good - it’s worth displaying a message when someone is looking at suicide-related topics. What about “how to kill someone”, " how to rape", “how to […]” with the hundreds of things that are universally considered wrong? And even worse, what about the thousands of things that are not universally considered wrong, but that some group thinks are wrong? “How to change sex”, " how to blow up a pipeline", etc.?

        This I think was their point in that conversation, and I agree with. The moment you try to interpret what the user wants to do with the info they ask you, and you decide to assume responsibility to change the user’s mind, there are hundreds or thousands of instances in which users or groups of users will demand you take a position for what they believe is right. Instead I think a search engine should stop at providing information relevant to your query and not assume what you want to do with it. It’s not its place to correct people’s behavior or educate people. The public education system should do that, the healthcare system should ensure people have the right support. A search engine is (or better, should be) basically like a librarian, or a library index, you ask what you want and they point you in that direction. They don’t try to guess why you are looking for books about torture or environmental activism.

        This at least is my perspective.

    • HeavyRaptor@lemmy.zip
      link
      fedilink
      English
      arrow-up
      2
      ·
      10 months ago

      Reading about the suicide thing: I might try kagi again this year. This is the exact type of software I want. It should only perform its intended function (returning relevant results to any query) and not try to influence me into following someone else’s moralisations.

    • 2xsaiko@discuss.tchncs.de
      link
      fedilink
      arrow-up
      3
      arrow-down
      1
      ·
      10 months ago

      Well, I was giving him a few days to backpedal but it seems like he’s not going to do that. There goes my subscription, back to ddg I guess. Since they are supposed to be 100% subscription funded (they still are, right?), this is one of the few companies where that hopefully might actually have a noticeable effect if enough people care about it.

      edit: also their response regarding a request to add a “don’t do it” widget to suicide searches basically boils down to “but if we make a moral choice now, we’ll have to do more moral choices in the future!”, which is… suboptimal. https://kagifeedback.org/d/865-suicide-results-should-probably-have-a-dont-do-that-widget-like-google/

      Meh, good for them tbh. I find these messages to be incredibly patronizing and somehow I doubt you can find a single person who will say “google posting the suicide number has made me reconsider killing myself”

    • hersh@literature.cafe
      link
      fedilink
      arrow-up
      6
      arrow-down
      4
      ·
      10 months ago

      As a Kagi subscriber, I’ve been very happy with their transparency in general. The feedback site is open to the public and Vlad and other staff members regularly engage in conversation about possible future features, limitations, and even business decisions in the Discord. It’s been refreshing.

      …which makes the response to this issue all the more frustrating and disappointing.

      I think Vlad’s comments in the original feedback thread were fair enough, but then later, in the Discord, I saw a lot of “let’s move this to a private chat”. They even changed their General channel to “slow mode” to prevent live conversations as this topic became hot. Now I see they were also deleting threads?! Ugh. That’s not transparent at all. Not what I expected based on my previous experience with Kagi.