6 ms·
Not to nitpick, but... If you take the top several dozen or so most-blocked domains from the Chrome extension, then this algorithmic change addresses 8
by bryanh 16y ago
Not to nitpick, but...
If you take the top several dozen or so most-blocked domains from the Chrome
extension, then this algorithmic change addresses 84% of them, which is
strong independent confirmation of the user benefits.
Well, what if they are mistakenly penalizing Common Jack's blog because it gets re-posted across the interwebs? I bet no one is manually blocking Common Jack's blog, so that factor is lost.
Example: let's say Google just randomly penalizes 84% of ALL domains. Chances are that they intersect with, you guessed it, 84% of the Chrome extension's data. Independent confirmation!
- mtkd 16y agoThe problem for Google is that any non-algorithmic approach (using a feedback mechanism) will just be abused by the SEO industry - they'll just crank up mechanical turk to start reporting anyone ranking above their clients.
- bryanh 16y agoI'm not advocating that, I'm just saying that the "independent confirmation" tells us little about the innocent sites that get negatively affected. They are, by definition, excluded.
- Dylan16807 16y agoAnd this somehow affects only 12% of searches?
- lutorm 16y agoTrue. It tests for completeness, not purity.
- invisible 16y agoThat is not what that said at all. It said that of the 36 or so domains that were the MOST blocked, 84% of those sites were lowered in the search results. Which means 4 or so crappy domains escaped being penalized. With that high of a "win" they might affect 1 or 2 domains on the entire Internet that should not be affected but I am sure it will eventually be corrected for those domains. This also means that many more than 36 domains will get punished for poor tactics.