| 12:10 am on Jul 17, 2006 (gmt 0)|
I'm seeing fixed site: searches on that datacenter, but no change in rankings, and the absolutely dreadful developments of the past week in terms of garbage blogs being everywhere is horrible there.
| 12:17 am on Jul 17, 2006 (gmt 0)|
>> 126.96.36.199 <<
That'll be gfe-eh.google.com then.
Do you get different results between when using the IP address, or the GFE-based URL?
| 12:28 am on Jul 17, 2006 (gmt 0)|
I’m seeing on that IP, supplemental problem fixed, root at the top, and little change in SERP position (for the site). It’s still basically buried for all relevant query’s, although there is some minor improvement. I have one keyword phrase back. On others I’ve moved from page 25 to 14, before June 27, site was on page 1.
| 1:20 am on Jul 17, 2006 (gmt 0)|
Yes, I get the same results on gfe-eh.google.com as well. What is gfe-eh?
| 1:23 am on Jul 17, 2006 (gmt 0)|
Yes, let's hope that does not spread. At least with the other datacenters we can easily see a problem when an index page is not at the top. With this one, the index is at the top, but the problem is still there in ranking, and that is the thing that matters.
| 4:40 am on Jul 17, 2006 (gmt 0)|
I don't have the same ranking problem on the gfe datacenter. It's not as good as it was before, but it's acceptable. I'm praying that it, or something similar, spreads.
| 7:24 am on Jul 17, 2006 (gmt 0)|
I have one term that I follow which seems to be sensitive to recent changes. That 2 word terms goes back in at #1 on the 72 DC we are discussing and whilst I would like to say that the rest of what moves for that term results in a better SERP, in fact it is no better than the rest.
I have a site that has pages that do very well on Ask and to a lesser extent on Yahoo. The best rank it has had on Google (except for occassional blips) is in the 60s range. On that 72 DC it is at #11 for one of my main target terms. That site has both www and non-www resolving to the same pages but all of the internal and external links are absolute URLs in the form [mysite.com...] so only the index page is crawlable as www and non-www.
| 7:26 am on Jul 17, 2006 (gmt 0)|
|I'm praying that it, or something similar, spreads. |
FWIW I think that requesting divine intervention is the strategy most likely to bring success ;)
| 8:17 am on Jul 17, 2006 (gmt 0)|
Well, now I see the problem. I have been wondering what are these ghost results that everyone is reporting which is very different from any of the DCs, but can only be seen on google.com or any of its search feed partners.
While searching from Bangalore, India, all the results I was getting on google.com were exactly the same as those on McNameless or the results on any of those DCs. Those results were great for me. But my log reports just didn't show any change in traffic from Google. Then I searched using an IP anonymiser and the results were different. Everything is the same, same DC etc, but the results are different. I would be curious to know which other countries where google.com results are same as those on its DCs.
[edited by: McMohan at 8:19 am (utc) on July 17, 2006]
| 9:02 am on Jul 17, 2006 (gmt 0)|
on gfe-eh.google.com i see my hijacked site back on top when I do a site: search, but I see another site wich works good with a PR6 lost 2/3 of its pages and all supplemental results, so realy we are not any further to any solution here.
| 10:17 am on Jul 17, 2006 (gmt 0)|
Good lord even more blog garbage dumped in during the past few hours.
It's like down at the plex instead of flushing they are siphoning....
| 8:07 am on Jul 18, 2006 (gmt 0)|
System: The following message was spliced on to this thread from: http://www.webmasterworld.com/google/3012804.htm [webmasterworld.com] by tedster - 10:09 am on July 18, 2006 (EDT -4)
These two DC groups show significantly different results.
72.14.203.* is with the majority of DCs.
72.14.207.* may be a vision of what is to come so it would be worth analysing what is different between this and the majority.
I watch a couple of terms that appear to be sensitive to the differences between these two and loads of terms that don’t produce any significant differences. Focusing on the ones that highlight some differences.
72.14.207.* handles semantics differently. Using a stem of one word of a sensitive 2 word term results in my #1 (non stem) ranking page being dropped to #36. On 72.14.203.* that page moves from #6 to #5 and is joined by an inset page. Also 72.14.207.* asks “did you mean split word secondword”.
I also have a site which appears to have had a penalty applied by Google for the past couple of years, it is closely linked to another of my sites and both www and non-www resolve to the same pages although only the index page non-www is listed because of careful use of full absolute URLs. This site is not in the top 100 for my sensitive term but comes in at #38 with the stem version of the term on 207. On 203 it is not in the top 100 for either the standard or stem term.
Do you have a sensitive term that produces different results on these two DC groups? Does it confirm, add to or reject my findings?
| 6:20 pm on Jul 18, 2006 (gmt 0)|
Ahhh Sid, on that dc, 188.8.131.52 and 184.108.40.206 I'm back to the first page for my preferred term, and rank much better for other terms. Wikipedia and the yahoo directory are still on the first page but Amazon and the other silly edu and gov sites are gone. Much better results I must say NOT ONLY because I'm back but because many irrelevant sites are gone.
If you remember, the present results had started showing on a handful of dcs, 2-3 months ago then they spread all over gradually.
So I'm hoping these results will also propagate too sooner or later.
| 7:14 pm on Jul 18, 2006 (gmt 0)|
Blackhat seo seems to be the way to go these days with Google's current ranking algo.
I'm through within the broken system it's now time to game it.
| 7:18 pm on Jul 18, 2006 (gmt 0)|
Maria"If you remember, the present results had started showing on a handful of dcs, 2-3 months ago "
yea the kopra coprolagnus you mean LOL
| 7:26 pm on Jul 18, 2006 (gmt 0)|
Yeah exactly toothake lol - did you check your results on the 72.14.207.* dc's? Are there still gov sites etc in your area?
| 7:31 pm on Jul 18, 2006 (gmt 0)|
no just wiki crap
| 8:43 pm on Jul 18, 2006 (gmt 0)|
For 72.14.207.* try gfe-eh.google.com and gfe-eh2.
For 72.14.203.* try gfe-ro.google.com and gfe-ro2.
For 66.249.93.* try gfe-ug.google.com and gfe-ug2.
[edited by: g1smd at 9:05 pm (utc) on July 18, 2006]
| 8:55 pm on Jul 18, 2006 (gmt 0)|
My default datacenter is [220.127.116.11...] whenever I go to google.com.
I don't understand why I always get the same datacenter,
does geo location determines the dedault datacenter ?
| 12:56 am on Jul 19, 2006 (gmt 0)|
and nothing from Matt yet...
| 1:16 am on Jul 19, 2006 (gmt 0)|
What would Matt say? I am seeing some SERP titles that were screwed up get fixed. Nothing update-like but after all these months stuff is getting fixed. No more vacations for Matt!
| 1:43 am on Jul 19, 2006 (gmt 0)|
Matt talks more when there are interesting things to say. "We are screwing up big time right now" isn't exactly interesting to talk about (heh, or news...)
| 1:46 am on Jul 19, 2006 (gmt 0)|
|Matt talks more when there are interesting things to say. "We are screwing up big time right now" isn't exactly interesting to talk about (heh, or news...) |
Unless he is really into self-effacement
| 8:40 am on Jul 19, 2006 (gmt 0)|
I wonder if this second refresh of data MC talked about has even happened yet? - May have happened with no-one noticing anything significant.
| 10:32 am on Jul 19, 2006 (gmt 0)|
Anyone has any idea about what error slipped in the algorithm?
| 12:01 pm on Jul 19, 2006 (gmt 0)|
That is the million dollar question - it appears all theories to do with age, affiliate links, unique content and anything else are flawed. It is either some sort of spam filter gone wrong, or a genuine error. My personal hunch is it's the latter and we just have to wait for Google to correct it.
| 1:00 pm on Jul 19, 2006 (gmt 0)|
I am seeing site: search fixed, as well as an accurate count of indexed pages - ever!
Of course rankings are down as they were, and supplementals fixed during Jagger are still there too.
| 7:59 am on Jul 20, 2006 (gmt 0)|
Google.co.uk is not showing cache links or similar pages links next to their search results today.
| 8:17 am on Jul 20, 2006 (gmt 0)|
Mods apologies for using actual terms but it is essential to illustrate the point and they are pretty general.
Go to McNameless and search on 72.* for
Why is there a difference between 72.14.203.* and 72.14.207.*?
Now search for
Why are all of the 72.* the same for this term?
The reason I ask is because I think that 72.14.207.* is what we are going to get propogate next and if we understand the differences then we maybe have a chance of tweaking our sites to suit.
| 9:09 am on Jul 20, 2006 (gmt 0)|
As Google is updating its index "constantly" as and when new pages, new links, spam etc are found what exactly is meant by the term "Data Refresh" these days?
| 9:10 am on Jul 20, 2006 (gmt 0)|
>>>>Google.co.uk is not showing cache links or similar pages links next to their search results today. <<<<<
It is from where I am sitting.... (London)
[edited by: Ellio at 9:11 am (utc) on July 20, 2006]
| This 179 message thread spans 6 pages: < < 179 ( 1 2 3 4  6 ) > > |