We were promised Strong AI, but instead we got metadata analysis

Re:We were promised Strong AI, but instead we got metadata a... by Mediterraneo10 on Monday April 26, 2021 @09:01AM (#26942060)

Google page ranking depends on how long people spend on the webpage, so if you have to wade through longform text, it mistakenly assumes that the content is quality.

Note that most recipe blogs are fake. Someone wrote the original content long ago. Then, someone else hired a copywriter off a freelancing platform to change the words of the text just enough to avoid copyright violation, then put up a new copycat website loaded with SEO and ads. Look closely and notice how the author’s bio says e.g. "Born and bred in Lousiania and I love to share southern cooking", but the text contains grammatical mistakes typical of Eastern Europeans or Southeast Asians.

The ecosystem is already so advanced that new copycat recipe websites are often based on previous copycat recipe websites.

[ Parent | on HN ]

Re:We were promised Strong AI, but instead we got metadata a... by criddell on Monday April 26, 2021 @09:20AM (#26942228)

> Google page ranking depends on how long people spend on the webpage, so if you have to wade through longform text, it mistakenly assumes that the content is quality.

Is this true? I have a hard time believing it because it's a pretty naive assumption. A site that lets me get what I'm after quickly is generally going to be what I want.

[ Parent | on HN ]

Re:We were promised Strong AI, but instead we got metadata a... by jcfrei on Monday April 26, 2021 @10:15AM (#26942716)

Google analytics is running on lots of pages. Would be pretty surprising if they didn't use the insights they get from there, such as time spent on the site, buttons clicked, how far they scrolled etc.

[ Parent | on HN ]

Re:We were promised Strong AI, but instead we got metadata a... by ThalesX on Monday April 26, 2021 @12:25PM (#26944415)

I worked for a startup obsessed with vanity metrics. They would freak out if the average time spent per article dropped down, even if our conversions increased. Just an odd bunch of people. They went down with the company while doing nothing but micro-optimizing for bad metrics.

[ Parent | on HN ]

Re:We were promised Strong AI, but instead we got metadata a... by mortehu on Monday April 26, 2021 @09:33AM (#26942318)

Presumably they are using bounce rates, i.e. how often you visit result N+1 after visiting result N. I hope time spent on site N is not a large factor, because it's so easily gamed.

[ Parent | on HN ]

Re:We were promised Strong AI, but instead we got metadata a... by ggggtez on Monday April 26, 2021 @11:51AM (#26943882)

I sincerely doubt it.

How many people keep open 50+ tabs, or open a page and walk away?

I'm sure that time spent on a page, if it's given any weight, is given barely any. I'm sure returns to search to click a new link is a much higher indicator that the user didn't find what they wanted.

[ Parent | on HN ]

Re:We were promised Strong AI, but instead we got metadata a... by BlueTemplar on Wednesday April 28, 2021 @04:08AM (#26966952)

Google doesn't literally know how long that page is going to be left open in a tab or window. IIRC they do know when you're back to Google trying the next search result ?

[ Parent | on HN ]

Re:We were promised Strong AI, but instead we got metadata a... by FredPret on Monday April 26, 2021 @11:34AM (#26943648)

I just had a vision of a perverse instantiation AI that takes over and turns the whole universe into copy-paste recipe websites

[ Parent | on HN ]

Re:We were promised Strong AI, but instead we got metadata a... by sjg007 on Monday April 26, 2021 @09:15AM (#26942183)

There are a bunch of these all templated exactly the same way. It’s fascinating.

[ Parent | on HN ]