241–250 of 299 posts
Earlier quoted context omitted.
I used Qwant a lot in its early phase, but then they decided to become yahoo-esque and also removed the lite version, hence I went back to ddg and brave.
What does yahoo-esque mean in this context?
The best way to stop SEO is to build a bot which detects commercial activity on the page (and/or availability of cart/payment controls), and commercial language in the text of the page (does the content look like an ad or product/sales page). Also combine with number of hops to a commercial page (with cart, prices, advertising material, etc), and create a distance metric. Offer the user a commercial activity slider.…
The presence of affiliate links should be the biggest indicator that the information provided is going to be heavily financially incentivized (ie. bad). Would be much simpler to calculate and detect. I would be super surprised if Google's algorithm doesn't already know how to detect affiliate links.
When I am looking up baking recipes or paleontology text on T-Rex, that should not be a commercial[izable] activity. Full stop! Commercial activity is like sexual activity in this specific regard: it's unwanted! 98% of my searches are for information that autists would pour their hearts into for free out of love, and that commercial interests will only detract from (such as pirated content and software cracks... lol wut did u xpect hm?). If I could pass a law which bled webmasters and search engines dry in fines until something like this is set up... I would! (Elect me BDFL of the world.)
The best way to stop SEO is to build a bot which detects commercial activity on the page (and/or availability of cart/payment controls), and commercial language in the text of the page (does the content look like an ad or product/sales page). Also combine with number of hops to a commercial page (with cart, prices, advertising material, etc), and create a distance metric. Offer the user a commercial activity slider.…
Maximizing ad profit on search traffic necessitates more paid ads, both on search results page and the web pages. They won’t say that directly of course but that’s what we’ve seen play out over last 20 years.
Vague metrics like “good user experience” get trumped by hard metrics like total $ / search.
Earlier quoted context omitted.
To properly compete with Google search you have to: 1. Create your own index/crawl and true search engine on top of it, rather than delegating that back to bing or Google. In doing so, solve or address all the related problems like SEO/spam, and cover the “long tail” (expensive part) 2. Monetize it somehow to afford all of the personnel/hardware/software costs. Probably with your own version of Google’s ad products.…
You missed out the only thing that's actually important: give users a compelling reason to use your site over Google. Google's dominance will never be upended by an incremental improvement.
Only people I saw manually typing the url are my office colleagues(mostly engineers of some sort), even management does the same(skipping the FB part obviously) and you can see it when they are sharing their screen and they want to search for something.
To pull people out of google moat, first step is becoming their default search engine.
I'm beginning to suspect LLMs' viability for search purposes is dependent on your existing search habits. 40% of my search queries are just copy-pasted error messages. Another 10% are business names, for the sole purpose of finding their hours or phone number. Less than 10% are complete clauses or sentences. I just don't see how an LLM fits my search habits. I tried using ChatGPT for search purposes, and it was dread…
So instead of getting a list of pages that contain that error message the LLM can evaluate which ones are closest to my configuration (are they the same OS, are variable parts of the error message the same or similar to mine, does the page have a solution or at least a discussion, or just one person posting the error with no other info).
Additionally search engines generally suck at negative phrases. For example I'm looking for some JS library that doesn't depend on React. Current search engines really suck at differentiating between "Foo for React" and "Foo, no React required". LLMs are pretty good at this.
That being said LLMs are also imperfect, so I wouldn't want the results that they reject to be completely hidden, but they can be ranked much lower. I think the biggest improvement would come from searches with many candidate pages (like looking for some library that fits my needs) rather than rare ones (like that error that has only been mentioned twice on the internet). But lots of times I am just looking for something with a very specific set of requirements that could be evaluated by an LLM (with pretty good accuracy). That can greatly improve my experience compared to getting all textually matching searches and doing the filtering myself.
The best way to stop SEO is to build a bot which detects commercial activity on the page (and/or availability of cart/payment controls), and commercial language in the text of the page (does the content look like an ad or product/sales page). Also combine with number of hops to a commercial page (with cart, prices, advertising material, etc), and create a distance metric. Offer the user a commercial activity slider.…
There’s a simple reason there’s so much SEO spam: Adsense. Maximizing ad profit on search traffic necessitates more paid ads, both on search results page and the web pages. They won’t say that directly of course but that’s what we’ve seen play out over last 20 years. Vague metrics like “good user experience” get trumped by hard metrics like total $ / search.
When I slide the Commercial slider to zero, only web pages which there are no obvious detectable ways for the host or author to make money from showing me the page should be returned.
Bad behavior stems from the drive to make money. Make a search where it is impossible to make money by being a result in the search (especially if a Commercial slider is slid to low or zero).
The best way to stop SEO is to build a bot which detects commercial activity on the page (and/or availability of cart/payment controls), and commercial language in the text of the page (does the content look like an ad or product/sales page). Also combine with number of hops to a commercial page (with cart, prices, advertising material, etc), and create a distance metric. Offer the user a commercial activity slider.…
Don't believe a single second the marketing speech about these 2 engines, they are both total crap trying to gain users with whatever hype subject is around. Qwant used to pretend being a champion of privacy, a "french made" tech, but with a search engine mostly based on Bing, lobbying with Microsoft against our interests, and with a boss sucking as much public funding as possible to finance luxury HQ locations and l…
If Google mangement can't see the user value of putting their own useful products in search, what hope is there for the rest of the world's useful information?
Earlier quoted context omitted.
>- and it fails the same in a loop Because that's the desired feature, not a bug. The system works just as intended. None of the people at the helm actually believe they can create competitors to US giants, the scope is to funnel public tax money into the right politically connected private industry pockets. It's just wealth redistribution with a veneer of "sovereignty", similar to large infrastructure projects, exce…
And usually at the end of a public infrastructure project you have something of some benefit to the public, even if there's bad RoI. With tech stuff that goes nowhere at best you have a jobs programme and short-term PR for some politicians. The fervent wish to only let deserving people get grants results in a huge amount of box-ticking and self-promotion that (in my experience) seems to select for self-promotion para…