4 ms·

99% of websites won't implement, promote, or get engagement with this universal button. And so then the way to rank is to get in a group chat with all your SEO

99% of websites won't implement, promote, or get engagement with this universal button. And so then the way to rank is to get in a group chat with all your SEO buddies and trade ~~backlinks~~ upvotes with each other. Or make bot accounts that appear real. Even if this idea worked, the idea of giving Google even more consolidated power/knowledge over the open web is horrifying to me.

rglover 2y ago
So those sites will be ranked accordingly. Upvote farming could be mitigated by vote frequency. That sort of behavior would be obvious in the data as you'd see significant voting activity over some time range (which wouldn't look like organic search behavior). You can't outright avoid the consolidation of power with search engines, unless (and I think this would be smart for someone to do), you create a decentralized version of what I described that's part of an alternative search engine.
- cjblomqvist 2y ago
  Google has been playing this game for 20+ years and failed hard the last 10+. You think you could easily solve it? I'm not saying they're applying the best of ideas and doing it well, but I'd wager none with good experience in this field would bet it would be that easy. I think it's more akin to hackers and defenders, each surpassing the other one in a never ending race.
  - rglover 2y ago
    > Google has been playing this game for 20+ years and failed hard the last 10+. You think you could easily solve it? Yes. Google hasn't really produced anything of substance post-2010. They fell into the typical corporate bureaucracy trap. The rejection of ideas like mine are likely even more common internally in Google. In an org of that size/stature, people jockey for position, they don't create (and that's not an opinion, it's objective reality). O.G. Goolge would try anything (I remember they offered radio station automation software briefly). Could I be wrong? Sure. But if public opinion of my core product is waning massively, I'd rather trust the mad scientist than the Wharton dork who's focused on his L-status. Worst case scenario the failure is a rounding error on my balance sheet.
- crowcroft 2y ago
  Maybe, although I think your understating the complexity that filtering through data at Google scale can involve. When you have enough users in a system many many 'normal' people will look and behave like bots/fraud, things that should be obvious are often not. I wonder if Stumbleupon could have achieved something similar to what you're proposing if they had tried to build a search engine.

rglover 2y ago

So those sites will be ranked accordingly. Upvote farming could be mitigated by vote frequency. That sort of behavior would be obvious in the data as you'd see significant voting activity over some time range (which wouldn't look like organic search behavior). You can't outright avoid the consolidation of power with search engines, unless (and I think this would be smart for someone to do), you create a decentralized version of what I described that's part of an alternative search engine.

cjblomqvist 2y ago
Google has been playing this game for 20+ years and failed hard the last 10+. You think you could easily solve it? I'm not saying they're applying the best of ideas and doing it well, but I'd wager none with good experience in this field would bet it would be that easy. I think it's more akin to hackers and defenders, each surpassing the other one in a never ending race.
- rglover 2y ago
  > Google has been playing this game for 20+ years and failed hard the last 10+. You think you could easily solve it? Yes. Google hasn't really produced anything of substance post-2010. They fell into the typical corporate bureaucracy trap. The rejection of ideas like mine are likely even more common internally in Google. In an org of that size/stature, people jockey for position, they don't create (and that's not an opinion, it's objective reality). O.G. Goolge would try anything (I remember they offered radio station automation software briefly). Could I be wrong? Sure. But if public opinion of my core product is waning massively, I'd rather trust the mad scientist than the Wharton dork who's focused on his L-status. Worst case scenario the failure is a rounding error on my balance sheet.
crowcroft 2y ago
Maybe, although I think your understating the complexity that filtering through data at Google scale can involve. When you have enough users in a system many many 'normal' people will look and behave like bots/fraud, things that should be obvious are often not. I wonder if Stumbleupon could have achieved something similar to what you're proposing if they had tried to build a search engine.