This is a related discussion from about 5 years ago about how SEO is ruining search. Google still seems to have a thick enough skin and a monopoly to get away with crap even after so many years of ruining search.
This isn't even the worst example, since it does at least have the correct info buried amongst tons of Ai generated garbage, but I can't use this for reference, since it tells me 4 different drill sizes. I've had to switch back to a paper copy of the machinist's handbook, since I can't trust the internet to give me accurate information anymore. 10 years ago, I could easily search for the clearance hole for a 10-24 fastener, now I get AI junk that I can't trust.
How have we regressed to the point that I'm better off using a paper book than online charts for things that don't change?
Half of web in Russia is blocked. Literally, powers that be think of Russian tech companies as of their servants and nothing more. Yandex basically sold their main asset, domain name to other entity.
This, I am a terrible note taker. For years a huge part my knowledge and skills relied on "if I found that information once, I'll find it again". My brain compressed the information by memorizing the path to retrieve it again.
Now that does not work anymore. You know some information is out there, you found it once when google worked, now it's lost in the noise.
I'm learning to take notes again and organize them so I can search them easily.
For queries like that I now turn to Gemini / ChatGPT first. Of course, this is only a good idea if I have some way of sanity checking the answer. If I doubt the answer I get back I try Google search instead.
I really like Kagi's approach to this, which is to give a list of references. There's still no guarantee that the answer is correct, but you can at least check the references :).
You can ask a model to provided an analysis of its answer including a probability that it is correct as part of the prompt, helps with doublechecking a lot.
They're consistent to the model, particularly if you ask the model to rationalize its rating. You will get plenty of hallucinated answers that the model can recognize as hallucinations and give a low rating to in the same response.
Models can get caught by what they start to say early. So if they model goes down a path that seems like a likely answer early on, and that ends up being a false lead or dead end, they will end up making up something plausible sounding to try and finish that line of thought even if it's wrong. This is why chain of thought and other "pre-answer" techniques improve results.
Because of the way transformers work, they have very good hindsight, so they can realize that they've just said things that are incorrect much more often than they can avoid saying incorrect things.
Does that extra information come from a separate process than the LLM network? If not then, assuming the same output is not guaranteed from the same input as per usual, then all bets are off correct?
Sorry for the late reply, but if you read this, there is research that shows that prompting a LLM to take variety of perspectives on a problem (IIRC it was demonstrated with code) then finding the most common ground answer improved benchmark scores significantly. So, for example if you ask it to provide a brief review and likelihood of the answer, and repeat that process from several different perspectives, you can get some very solid data.
> How have we regressed to the point that I'm better off using a paper book than online charts for things that don't change?
because products that require iteration lend themself to subscription models which in turn mean a recurring revenue which is deemed superior to onetime payments for a 'finished product'.
We need to collectively stop using Google, but the alternatives are just not as good for some things.
The best one is probably Kagi, but let's be real: "normal" people would never pay for a search engine service. Well, "normal" people don't even know the difference between Google, Google Chrome and probably the internet.
I say this as someone who doesn't yet use Kagi but is increasingly warming to the idea: I think normal people may pay for a search engine, one day. People used to think the bottom had irrevocably fallen out of paying for media once internet piracy became a thing. Streaming services may be in an unappealing state now, but they at least showed that people can be persuaded to pay for something if it makes access to the things they love easier. We might be years away from it, but I wouldn't say never. And $10 to find things on the internet again seems like a more persuasive offer than what people are currently paying for streaming.
I wonder if there is a market for search engine where sites pay to be listed. Not a substantial amount like they would for ads. But a small compute fee for the spider plus a contribution for being indexed.
People will pay for clicks. If a small search engine with pay per list sends a lot of visitors website owners will for sure pay. The key I think is to have verified and unverified listings, and give people a free trial of being verified before bumping them out so they know what they're paying for.
My default is Duckduckgo but in the recent weeks they have massively upped how abrasive their ads are. Nothing wrong with there being search related ads but the sheer volume is getting out of control.
Hi, I work at DuckDuckGo - thanks for the feedback. Nothing much has changed in the past few weeks on our side, and so would love some more info so we can investigate. Do you mind sharing what country you are searching from, and whether you noticed on desktop or mobile? Also just FYI — you can turn ads off completely in settings if you want.
I am based in Australia. I am not sure if it is the quantity has increased, but their size definitely has. I usually have to scroll a full screen down before I start seeing non-paid results. I am on a desktop machine, mobile has been absolutely fine.
Also didn't know you could turn them off. I don't mind ads to support businesses when it is reasonable and DDG is a decent business. I just hope you folks don't try to keep up with the giants by playing their game. ;)
Got it, thanks for the follow-up, and for your support! It should be pretty rare that you would see no organic content above the fold. But we'd like to take a closer look, and if you have any particular queries where the ads are surprisingly tall, it would be useful to know.
With the forthcoming winter of synthetic content, we may easily find ourselves, in the coming few years, forced to resort once again to directories a-la AltaVista and Mozilla's. I really see no way Google would stop their ads activities, as these provide the financial backbone.
In a sense we resorted to the searchable message board, once an university homework assignment, in the form of HN here.
Rest assured the existing tech monopolies who are flooding the internet with hallucinated AI garbage will be there to sell us our own "Verified Human-Written Content" back to us.
Mind you this is then before the recent article that alleges Ben Gomes was pushed out of Google[0]. This was my feeling regarding that post, that search had been getting worse from before 2019.
This is a related discussion from about 5 years ago about how SEO is ruining search. Google still seems to have a thick enough skin and a monopoly to get away with crap even after so many years of ruining search.