The Age of PageRank is Over | Kagi Blog

[rcjordan]

"PageRank broke because nobody links to or curates content anymore. If they do, it is mainly for commercial benefit, not based on merit, which was the essence of both the original web and the algorithm."

https://blog.kagi.com/age-pagerank-over

[ergophobe]

I admit that I sort of skimmed that, but the author never seems to come right out and mention that there was a predictable cause and effect and the Page Rank algorithm is what destroyed the collaborative web.

Prior to that, even if my site was ad-supported, there was no negative impact in search results to linking out. Once I needed to hoard PR, then I was under pressure to not link out. So over time, sites with commercial intent basically stopped linking to other sites.

Of course, basic conversion optimization would result in fewer outbound links too, but there was a time when most SEOs and most clients would avoid outbound links purely due to rankings. I think that was the biggest driver, leading even sites without commercial intent to become parsimonious with links. So bloggers stopped linking to bloggers for fear of losing PR, which has really hurt bligging as an art.

Unfortunately, that signal has also corrupted the web. If you go to a recipe page, there's 1500 words of fluff before the recipe.

I have to say, this is an area where I was completely wrong. 15 years ago, I would argue that Google would get better at an increasingly rapid pace and by 2025 (which seemed so far away), it would not be possible to spam a search engine, because the AI would *understand* the page and the question and would be able to tell, algorithmically, which pages genuinely had the best information for a given query.

My big surprise on the web is how little search has improved. I do think it has improved, but the quantity of pages has drowned out many improvements. I remember telling people that "When kids born today are in college and we tell them that we used to find information by typing 'keywords' into a search box, they are going to laugh and wonder how we ever found anything." Well, those kids are fast approaching college and I still find most things by typing keywords into a form box.

I should keep a catalog of all my predictions... I don't think I'm a very good futurist (though, to be fair, I don't think many futurists are very good futurists).

Now I think that the search engine that I was certain would be here by 2025 (2030) at the latest will not, in fact, be here soon, perhaps not in my lifetime. Upon reflection, I realize it's one of the most difficult problems, harder even then self-driving cars and we all know how that's going.

[Brad]

To me, making webmasters afraid to link freely, is Google's greatest crime against the Web.  Involuntary Webslaughter at best but part of it was premeditated with nofollow schemes and sowing FUD.  This is why I do not see requiring Google to share it's index with other search engines as a means to break Google's search monopoly, as doing any good.  Google's index is huge but in many ways it now sucks.

[BoL]

Article seems to use broad strokes talking about how the web has changed and will continue to.

Fully agree about PR 'bastardising' the link graph, bet the toolbar didn't help on that front!

I think the closest Google were to solving it was with G+ and the advent of structured markup. If you could tie authors to articles then you sort of have a handle of intent, and the links between these 'verified' authors gives a lot more context. Harder to trust with just schema alone.

[ergophobe]

>>premeditated with nofollow schemes

I would disagree with that. I think they naively did not think about how easy PR was to game and nofollow was an attempt to get the horse back in the barn (as was authorship as BoL said). But PR was the original sin. Once they rolled out nofollow, the damage was done.

But of course, nofollow made things worse, since people started using nofollow on all outbound links not just untrusted links as was the intent originally.