Forum Moderators: Robert Charlton & goodroi
As reported by rustybrick below, Matt Cutts and Google now do understand what happened and they are backing out of it, rolling out the changes through their many data centers.
And this brings up a bigger - and for me, even better - question. What were they trying to do? Is Google now continuing to do it, but without "the mistake"? How much "forced position" work does Google do in the top positions, especially now that Universal Search is the rule of the day?
We begin with observations from our community, as the signs of the rollback began to appear a few days ago:
[edited by: tedster at 9:00 pm (utc) on Jan. 29, 2008]
We've got action.....almost all datacenters are reverting back to non 6 bugged positions.
I'm still curious as hell what caused this whole thing in the first place.
[edited by: Timetraveler at 7:28 am (utc) on Jan. 25, 2008]
[edited by: tedster at 8:43 pm (utc) on Jan. 29, 2008]
who thinks it was simply a glitch in data
I'm still going with my earlier theory.
This had all the markings of a normal testing roll-out when it began.
Then suddenly it got switched to all the DCs with no rhyme or reason.
Again, this was right before Christmas. aka. people aren't really focused on work. They'll focused on holidays, vacations, etc.
If human error is a "glitch" then you can call it that.
But from observing the past week, it seems obvious to me that the "test" never got finished or it was forgotten or whatever.
If you eliminate the 3+ weeks were nothing happened and things were "stuck" the entire testing proceeded like most test.
Test starts on a few DCs ---> Rolls out to more.----> Moves to live google.com off and on for a day or two to get actual results ---> results slowly but surely start to roll back onto less DCs until the results are folded in or discarded.
This usually happens in a week/week and a half. -- not a month+.
which is why i even started this stupid observation :P
Could even be that somehow we got to 'see' the re-ranking filter that Tedster has talked about, but that the operation, which was due to continue and unfold, was inaccurate. This might also why there it was not acknowledged at Google, why it wasn't stopped if there were errors part way, as doing so could shed light on elements of the process of ranking?
order seems to have been restored, but i could not even begin to offer a clue as to what has been observed. pages i left alone, pages with massive link changes, and pages whose on-page content was churned, all return to previous states of grace, at least for last two days.
if google was trying to get the attention of the complacent, it worked in this little corner of the net.
I'm wondering out loud whether there was any connection between this test (if that's what it was) and this new positioning.
I should also add that not all of the #6 results I've watched moved back up. Some have dropped down further.
[edited by: tedster at 8:44 pm (utc) on Jan. 29, 2008]
[edit reason] moved from another location [/edit]
Matt Cutts confirmed the roll back - it seems like this was all a mistake.
<rollseyes>
Gee, it's almost like he read this thread and decided to ask somebody what was going on.
And he's said nothing for the past 4 weeks?!
<insert "mass Google Noise grumbling to those who know why i'm grumbling" here>
Moral of the story; Don't rush to make changes when these aberrations occur.
Hope everyone gets back to where they were, and we can all look forward to fretting over the next Google burp.