I swapped to ddg about a year ago. I've noticed their search results are not the greatest either. There is so much spam and paid articles that make it to the top search results I often end up frustrated trying to find what I'm looking for.
Same, but lately Bing fucked with their pricing, and now the search results are usually garbage. I have to go re-search the query in Google way too often.
Fuuuuuuuck Amazon is even worse, if that's possible. You used to be able to search for a product and would get that as the first result or at least on the first page. Now, they just return shit that is vaguely related but is a paid sponsor or someone manipulating their search algorithms.
It's all trash these days. We're having to just avoid the worst of it while preparing for the rest of it to also suck so bad that we have to find even more alternatives, and then just throw our computers away at the end or at least go back to computing where it wasn't all about stealing data, long, long ago.
I get where you're coming from, but I also feel like I might get kind of bored and I wonder how I'm going to get that hot sauce from Alabama that I like. I'll give up on a lot of my ideals and trade a LOT of frustration for artisanal hot sauces.
I'm honestly amazed no one's sued google over the site telling them things like "Jump off a bridge if you're depressed" or that products have been recalled that haven't
Google wanted to organize all the world's information to make it useful. Sounds like a great idea destroyed by a profit motive getting in the middle. Try the same idea without the profit motive and you'll probably end up creating something people will defend endlessly, like Wikipedia.
What if your search engine asked you for a five dollar donation, once a year, the way Wikipedia does?
Search engines should be non-profit, honestly, they should be an extension of Public Libraries.
I haven't kept up on this controversy, but is this really all there is to it? People think the service is untrustworthy because the CEO tried a bit too hard to convince a critical blogger, according to the comments possibly due to neurodivergence?
Read the blog post. I can totally see where the writer is coming from. Vlad's email responses are pretty laughable with that added context. Not only did they tell him they weren't interested in engaging, but the responses he sent anyways thoroughly dodged the substance of their arguments. They did their research and came to pretty reasonable conclusions, and vlad totally talked around/failed to address the points that really matter. Kinda slimy ngl.
I can also see from the highlighted discord exchanges why they weren't interested in engaging with him further. Sure he "responds" to everything, but he doesn't really engage with constructive criticism. Whether or not the guy is autistic isn't really relevant to whether he can run the business competently and make sure questions are answered in an honest and transparent manner.
It's the equivalent of doing a Google search and then clicking on the "Web" tab for results.
Clicking on "Web" manually inserts the &udm=14 at the end of the search string. I believe udm is just classifying which type of search to run, and this is a basic "web search" with no gimmicks.
&udm=14 is fancy for nerds who think they're special but for anyone else it's just clicking on "Web" results.
This started to become noticeable years ago when Google decided to start censoring searches even with SafeSearch off.
I switched to Bing at that time, which was good for a while, but eventually they have started doing the same thing.
I can now no longer find a search engine that actually works to find me all the relevant results. I've tried all that I've heard of, and none will provide complete results.
The easiest canary in the coal mine for this is NSFW stuff. If I search for a popular character of which I know there's lots of porn/hentai, with SafeSearch off, and do not get a heavy mix of SFW and NSFW results, I know that search engine is messing with those results, and is also definitely doing it with searches that are not so obvious.
The top results have been useless spam for a decade or more at this point, and the only difference is Google sit there hoovering up the money instead. The money is in the way of search. Any popular search engine will end up the same way.
It's a shite situation, but until somebody makes a non-profit search engine and filters out spammy results, we'll continue to Google, scroll down two pages for a reddit link, and carry on.
CEO of Kagi has no boundaries and doesn't know how to take "no" for an answer. He thinks he can just browbeat other people into agreeing with him. It's a bad look and makes me have zero trust in Kagi as a product.
I said elsewhere just a few days ago: These fucking techbros need to realize that sometimes shutting the fuck up and letting your products speak for themselves is the better move than trying to browbeat everyone into acceptance of your view of your products. Prelovac's insistence on being heard by someone who has no interest in a conversation with him screams entitlement and a controlling attitude. People who can't take no for an answer generally are not good people. Not being able to accept someone else's lack of consent and wanting to bully them into consent is a bad fucking look, period.
When I read your post I was expecting something much worse than what you linked to.
It wasn't really all that bad. Imo, it was 1 or 2 emails too many.
Maybe I'm a little biased because I love the product so much. It's a fantastic search engine again and all of the AI extras add value and aren't obnoxious like everyone else's.
I am still happily paying for kagi even if the CEO emails people that write shitty blogs about them.
It feels like this is the only bad thing about kagi that keeps getting shared every time kagi is mentioned. Somehow it feels like one person and their followers trying to promote themselves as Karen's more than legitimate accusations.
A CEO getting involved with the end users is not a bad thing imo
I'm currently prioritizing DuckDuckGo over Google and others for a better quality of search results. There ain't much ads to mess your attention, and the results are quite good most of the time.
If you need to use Google, I'd recommend the &udm=14 trick, as demonstrated on the linked site. It goes straight to a "web only" filter for search results. There are some tricks and tips to set that to your default search option in settings, or you can get a browser extension.
It's an extremely bold assumption to make to presume that we even could? Like Ukraine wanting to not be invaded, or Rome wanting to not fall, or someone wanting to not die of cancer, or perhaps the best analogy: never exercising a day in our lives + eating however we feel like in the moment, yet wanting to not suffer the negative consequences of obesity, hardened arteries, etc.
I take the approach of the stoics: we brought this upon ourselves, allowing ourselves to be tricked by the "Don't be evil" slogan. I do not own Google - I have no stocks in the corporation, they are not running their server code on my machines, they use none of my electricity, I do not own the land they park their buildings upon, etc. - and therefore I have no call in how they choose to go about their business. They chose to enshittify, and I am not offered a say in their choices. Therefore "we" cannot "fix" this. Only they could, and only if they want.
But maybe we can build our own LLMs so that we never need to use Google to search for anything ever again, directly? For now, I use DuckDuckGo whenever I can, Google when I have to, and perhaps most important go directly to the site that I want if possible - e.g. wikipedia, wiktionary, stackoverflow, Reddit if I absolutely must, etc. We lived in a golden era of prosperity when we were allowed to "have things", but that is over - we did not take care of it properly, and it was taken over from the inside, as it was always going to be, our delusions to the contrary notwithstanding. Now, maybe we can be more realistic about our expectations moving forward.
There's also an ongoing project called Yacy which is trying to make a decentralized ever-growing index, but it's small enough right now that the results are still quite poor even when federating with other users until you spend enough time fine-tuning it, which most people don't want or have time to do.
That's neat - though I would worry that it could get polluted easily, e.g. China, Russia, and/or fascists everywhere would very much like to control the conversation, so as it got a wider userbase that would be the time for it to cease functioning, whereas before that it could be allowed as something to occupy our time. So much of our daily lives are impacted by such geopolitics that we don't even/often think of.
I hope that we (people writing FOSS) can explore the concept of voting more - e.g. how wikipedia does its edits with "trust actors", similarly the Fediverse (& searching) could have a much wider pool of trusted community mods (or potentially a lesser category of that, lacking full post-removal capability as current ones do) where mods could vote and once something went below a certain threshold (e.g. 5x more down- than up-votes from such community curators), then a flag could go up like "this post has been marked as containing potential misinformation - are you sure that you would like to read it?" By distributing the load like that, it could help make this place MUCH more active, by lowering the barrier to moderation as it conjoins normal reading activities with a very simple button bush for most people. (and then a higher category of mods can double-check the curator mods, etc.)
On the other hand, authoritarian developers are unlikely to want to extend the Lemmy code along those lines, and rather go the opposite direction so that anytime someone says something even the slightest bit against their party line, boop the person becomes insta-banned in every single community that they have ever interacted in, even if never having commented but solely voted in them (I am not exaggerating this up - for one see the notes for the upcoming v0.19.4 release allowing this capability, and multiple recent threads discussing how this has already happened to numerous people - e.g. here is one excellent accounting of that process).
So anyway, if someone is willing to build a good system, then the people do seem willing to take it forward - e.g. see how many have already done so to Google Maps with all the reviews & pics including menus & such to share with everyone. But ofc there will be resistance to now doing that again for somewhere else, and perhaps still yet again if that one fails, and so on in perpetuity.:-(
Great! Everyone should try the free tier just to have a look at alternatives. To be honest after a few weeks and several specific searches I'm not ready to do the switch though.
I was (trying to) use AI to help me with something on my PC a few nights ago. It was telling me to go menus and choose options that literally don't exist.
Wtf is going on with AI right now? About a year ago, I found it useful for certain things. But it seems to have totally shit itself recently, and is pretty much useless now.
Part of the issue isn't just that they added garbage, it's that sites got better at targeting the algorithm. It's like recipe websites that started to add stories because the algorithm prioritized that. No one was asking for it, but the algorithm was broken in unexpected ways and exploited.
So, someone could make a new search engine with a different algorithm. If it becomes successful though, it's only a matter of time before it becomes targeted and manipulated.
So, someone could make a new search engine with a different algorithm. If it becomes successful though, it's only a matter of time before it becomes targeted and manipulated.
I can think of a solution to this. Once your new search engine is starting to get targeted, you find out who is offering these SEO services. Then you hire a gang of thugs to beat the living shit out of them.
I tried using DDG for about 3 months, until I had to ditch it. I want to use it, but its results are even worse than Googles.
While on Google I might have to scroll past the ads, sponsored links and obvious AI blog results, the first page will usually at least contain a relevant result somewhere, whereas DDG simply doesn't bother to show any relevant results at all. Even on it's first page, it will stray far from the actual search term.
Just double checked the email confirmation from when I bought it in 2016. Most recent xpadder version is 2024.05.01.
Also if you're looking for controller remapping software (and you're on windows OS), check out DS4Windows. Contrary to the name, it works for most brands of controller, and it's got a lot of stuff that xpadder doesn't.
I landed on it for the better control stick adjusting (you can change dead zone/max zone, and also sensitivity curve). I stuck with it because it can use an emulated controller while hiding the actual controller from programs and games (which solves any double input jank from shit like steam).
Also it can also map gyro controls to mouse input, which is real neat for essentially using my joycons as a remote for my pc, and the occasion where splatoon style gyro aiming fits in games without being too barfy.
Last update a month and a half ago. Great for what it is, but ds4windows is my go to until it stops working for me. More complicated, but it offers a good deal more finetuning.
I need to take a screenshot the next time I see the "other users have searched" thing is almost a full page in length. It's happened a few times recently.
4get, SearXNG and Whoogle are pretty good proxies for most search engines. I've recently been enjoying 4get with DuckDuckGo, it works really well. No ads, no trackers, no bullshit. Just a lightweight page that allows me to search the web privately.
I've just switch from DDG to Qwant. I didn't do research about how they make there algorithm but DDG that IS to say Being was returning me unrelated stuff and even things I ask not to see in my request. The result is not amazing quality but no more amazingly bad quality.
I want to learn more about search algorithm, what there referencement politic is and what use of AI they have. It not possible today to pick a search engin algorithm randomly and get pertinent résultat like it was two years ago.
Searching "SpaceX starship" on Google just now in Firefox with adblock origin, not logged into Google, on mobile I get:
Summary of the vehicle from Wikipedia with relevant pictures on the first card
The SpaceX page for the vehicle on the second card
Relevant videos on the third card
SpaceX's twitter on the fourth card
Doing the same in mobile chrome, logged in, no ad block:
Adverts for starship toys in the first card
Relevant news stories in the second card
Wikipedia summary
SpaceX's starship page
Relevant videos
SpaceX twitter
So if you're getting other crap, use ad block. I presume the comic was made while Google was adding LLM results, which they aren't doing now. I can't say I noticed the LLM results while they were included.
If search engines really wanted to fix this--and it would require Google and Bing taking action, not smaller search engines--they could de-list any company that engaged in SEO so that nothing from that company showed up in searches at all.
That would just change how SEO worked, since the old methods would be less optimal. Search engines have to rank results somehow, and whatever that is there's going to be some way to game it.
Maybe I should have said that any attempts to game ranking--versus actually creating content (e.g., no repackaged content)--would result in being blacklisted until such content was removed.
Gemini is basically an alternative web, so it has its own protocol, own server and browser software, and an own content language (heavily inspired by Markdown).
Everything is built to be much more lightweight, so certainly in the spirit of solarpunk.
But big downside is that "the thing you want" is honestly probably not on Gemini at all, because it does not share content with the HTTP web.
That does mean that all the ads and bullshit are gone, too (and cannot be implemented beyond static images, sponsored articles and I guess, ASCII art).
But yeah, at this point, it's most useful for reading (and writing) blog posts, and having fun with the technology itself. It does have a lovely community for that.