Hacker News Reader

Comment by siliconc0w | Hacker News Reader

It's hard to tell if it's Google's dominance causing search to suck or just the web ecosystem naturally optimizing for clicks. I will say most queries with even vaguely commercial intent do seem to get the max or nearly a full page of sponsored links. I don't recall what it was like years ago but I'm pretty sure it is worse. I also think it's pretty clear the platform take-rate needs to be generally lowered across the board, likely making it illegal to lock customers to a store or require payments go through the same store. This is a tax on basically everyone and provides very little value.

View on HN

>I don't recall what it was like years ago but I'm pretty sure it is worse.

It's absolutely worse today, how much worse depends on how far back you go.

Before 2005 Google didn't put ads in search results, AdWords results were separated off to the right side of the page to make them clearly distinct from organic search results.

Kagi stands to me as proof that search can be better if it isn't run by an ad company. Kagi deliberately downranks sites that are loaded with ads and trackers, and the quality of their results is much higher, with much less SEO blogspam. Unsurprisingly, having a high concentration of ads and trackers is highly correlated with having content that is intended to maximize views rather than provide utility.

Google has a conflict of interest here: on the one hand their users need information, but on the other Google is the primary vendor of advertisements on these garbage clickbait sites. It would not surprise me if Google intentionally upweighted results that have their ads in them, and they certainly are not downweighting them the way Kagi does.

If Kagi were as popular as Google, you'd quickly see pages appear "without ads", that looks suspiciously like what an ad would say. The problem is SEO, and it affects the mainstream search engine because that's what's worth optimising for.

For instance, Google product search results would likely be greatly improved by always showing relevant reddit posts first. But this would last a very short amount of time until reddit would be completely drowned in junk. (I've seen people argue that this is already the case, but searching reddit still tends to work well for the products I search for)

We can be happy that Google search hasn't gone this way, and as a Kagi user you should be glad that it's not a more popular engine, as the quality of results likely highly depend on that just fact.

In a way the ads-filled product search results are better, because at least they're honestly trying to sell you the product you are searching for rather than making fake comparison sites pretending to be genuine content.

> that looks suspiciously like what an ad would say ... fake comparison sites pretending to be genuine content.

This is already a thing under Google's system: if a result isn't an ad-ridden mess, it's content marketing trying to sell an adjacent product.

You haven't addressed the conflict of interest that Google has as the largest vendor of ads on the internet. Right now, Kagi's sole job is to help me find what I'm looking for. If SEO started to game their current algorithm, the incentive would be for Kagi to improve their algorithm and win the meta game. If they fail to do so, their payments will dry up as people look for an engine that will.

Google has no such incentive. If their algorithms consistently drive traffic to pages that have their ads on them, Google is incentivized to keep that traffic flowing, regardless of how SEO'd the content is. The user isn't Google's customer, the advertiser is.

In addition to what lolinder writes Kagi already gives me what we have askes for from Google for a decade or two: a way for us to punish creative SEO artists by individually downranking or outright block them.

If they start popping up in Kagis result they will bother me at most two or three times before I hit the downrank or block button and they are gone.

Why Google hasn't managed to create something similar during the last two decades I leave as an exercise for the reader.

> In a way the ads-filled product search results are better, because at least they're honestly trying to sell you the product you are searching for rather than making fake comparison sites pretending to be genuine content.

In what universe can for-profit behavior be trusted or accepted at face value as genuine? Also, maybe I'm using it Wrong™, but I don't use general search engines to look for products. I'm not shopping all the time, so why should we be advertised to all the time? I use a blocker and aggressively eradicate ads from my experiences, precisely because they are useless to me, take up valuable screen real estate and bandwidth, and are security and privacy vulnerabilities.

It wasn't so bad back when it was just a JPG and a link, but the telemetry that gets baked into it is creepy as well. For-profit Internet behavior needs to rein in its snooping if it wants to gain trust.

I interpreted differently, in that it’s easier to filter out the explicit ads than the sites that purport to being honest review pages but actually have affiliate deals for every product

How does kagi do with "corporate blogspam" - things like tutorials, guides, intros, etc created by a company as a funnel for conversions.

You know, things like the fusionauth OAuth guides (not trying to call them out).

I ask because it's a weird area where it can be shit and predatory, or a general good thing.

It's present on Kagi as well, but less so than on Google.

Try Kagi. They are weeding out the bad results waaaay more aggressively than Google. It helps a lot. They even allow me to immediately boost or block whole websites! Something absolutely trivial Google just refused to do.

The problem here is authority. What makes you an authority on a subject? Books written? Credentials (diploma?)? References?

These things aren't all true in real life, and on the internet it's even easier to fake all of that.

In the beginning Google went by 'amount of text' (which is why you read everybody's life story on how they had sex next to a magma flow in Hawaii which created the idea of doing burnt ends on a pizza stone), references (remember all those backlinks which would boost your page). Credentials are a lot harder to track. There is no API at my school where you can put in my name and see my credentials (and all you'd see is 'we got governments subsidies for this person, which is why we let him pass').

And all of this worked at the start. Knowledgeable people naturally had references on the web, as only knowledgeable people were on the web.

Now we literally have people writing text purely to make it look like their website has more content on them, even if it is remotely related to what they serve.

Great example, even posted here: https://news.ycombinator.com/item?id=36740400

Or any coding problem where you end up on logrocket / some partitioning app / some generated website with which has an 'answer' for KB0000000 -> KB9999999...

I work (partly) in SEO. One of the main issues that affects search and will get even worse in the future is the post-scarcity environment of the web when it comes to information. Say you had five websites for that each have identical cookie recipes. If you optimize search rankings by click-through rating, eventually all the sites will have similar copy, titles, and page descriptions. If you optimize for bounce rate, you knock out sites that get to the point but also knock out sites that crappy and have bad ui/ux. What about core web vitals? Well again, everyone will catch up eventually. Anything that can be gamified will eventually be gamed.

Now instead of 5 sites with identical recipes, what about 500? 5,000? 5,000,000? How can you even rank them in a meaningful way and does it matter to the end user?

There will be too much of a supply of information, especially when ai ramps up, that no one website will have unique value outside of local significance or if it was made by your mom or someone you like. I think it will be crazy to see what the web looks like 10 years from now. It could be vastly different.

You could stop working in SEO and just design shit to be responsive to typical boolean search like Tim intended. But dear God, people might be able to find things then, and where would you fork in the distracting cruft?

The problem never was or will be Search. The problem is advertisers hijacking the verb of "Search" to weight it for those willing to pay.

Librarians and archivists have had search solved for the last century. The only people who have a problem with that implementation are the ones who want to convince you the answer you're looking for is them.

This expectation that everything can be automatically done by algoryths is the core if the problem. its like a form of insanity, it doesn't work but we keep trying.

Why are top 10 results for 'apply for visa to Moldova' a scam? Is it really impossible to give government officials an ability to provide input for official business, like applying for a passport?

Could you not hire like 2 people in each makor country to keep track of websites for important government services?

Could you not take community input like reddit does, or wikipedia?

seriously sometimes i think the people making the smartest algorythms are tge ones with biggest tonnel vision, and are incapable of examining alternative solutions.

...It isn't the algorithm that's the problem. It's people invested in becoming a value extracting diversion for an answer that'll be around.

Lets take your problem. Applying for a visa for Moldova. Let's say that there's a site hosted at visa.moldova.gov with a form to apply.

Eventually, someone is going to integrate a bunch of data broker services to auto-fill that form, then data broker to other tourism related businesses trying to compete to plan your stay.

It is very much in their interest to ensure that no one sees visa.moldova.gov, but instead sees one of their onboarding funnels.

Thus the well around a well known onboarding point to a critical government service is poisoned.

>>What makes you an authority on a subject<<

the ability to make accurate predictions of outcome by way of knowledge, observation, and logical deliberation.

You seem to have misread the question. How does Google, an automated service, "make accurate predictions of outcome by way of knowledge, observation, and logical deliberation"?

they dont, and thats a major issue.

After switching to Kagi, I absolutely see this as Google intentionally making search awful. They are catering to the web ecosystem that is optimized for clicks.

The old Internet is dead. There used to be people building web pages just because they are passionate about something.

Now any imaginable niche subject is drowned by a deluge of blogspam articles.

Interesting people moved behind walled gardens because who is going to bother building a webpage when you can just post a pic or 140 characters on a whim and move on with life.

Please try search.marginalia.nu or even Kagi.

I was about to give up but these to have given me hope that search isn't unsolvable.

well building the webpage is easy, but paying hosting and keeping it updated is a side jobby i dont need once I've reached 30 and have kids

It's hard to tell if any given technology has led to real productivity increases, or just led us to hallucinate productivity increases given the number of little tasks we can fire off with technology. This has been studied; there's too many variables to say "email increased profits 500%!"

If it's too complicated to say one technology is where all the value lie, it's going to be extremely complicated to say Google's existence is meaningful.

We can lean on the fundamental social truism; it's a front to act as a bloom filter for the US dollar, and busy work for people that might otherwise topple the government.

Reply on news.ycombinator.com