| 2:23 pm on Jun 19, 2006 (gmt 0)|
|I think that they are just trying to be too clever and from where they started from a simple easy to use search engine has just turned into a pile of useless information that searchers can't even find what they are looking for. |
While I agree the current results are crap, I think we should keep in mind that the spammers are (in large part) respsonsible for the current state of things. The recent story about the spammer who injected google with billions of pages of junk is a prime example. The bad guys always seem to stay one step ahead of the cops...
| 2:32 pm on Jun 19, 2006 (gmt 0)|
Google is a Goliath that is fighting an army of Davids.
The talent and resouces being thrown against it are likely too much for it to handle.
| 2:35 pm on Jun 19, 2006 (gmt 0)|
I'm seeing different results to deafult uk dcs
Desktop Machine is picking up 126.96.36.199 and my laptop is (wirelessly) on 188.8.131.52.
Which one is the Turd? I'm in one of them.. (gulp)
| 2:38 pm on Jun 19, 2006 (gmt 0)|
you cant put the backing off of the throttle on indexing down to spammers...not re-spidering pages all looking for new ones for months on regularly updated pages and sites (that arent huge) is a total change in the direction of search...
| 2:43 pm on Jun 19, 2006 (gmt 0)|
Aha - the same dc returns competely different results on both machines for the same key term.
They certainly dont seem to be organising information - more revolving it!
| 2:47 pm on Jun 19, 2006 (gmt 0)|
I think that we have to wait.
for example I get different Google IP from my home ( the IP starting with 72.14. ) and from my office (184.108.40.206.).
I noted that this IP 220.127.116.11. has a cache older than the 72.14 ..... .
I think that the fresher ( 72.14 ) index is cleaner: less spam compared to 64.233 .... .
| 2:52 pm on Jun 19, 2006 (gmt 0)|
Spammers took over Googles serps in the adult sector 5 months ago - Big Daddy arrived.
I'm not surprised to see spammers winning the battle because Google took 5 steps back with Big Baddy.
Spammers will continue to make an inroads into Big daddy because Google is not going to fix the problem.
| 4:21 pm on Jun 19, 2006 (gmt 0)|
|Which one is the Turd? I'm in one of them.. (gulp) |
18.104.22.168 is Copra
22.214.171.124 is Turd
| 4:26 pm on Jun 19, 2006 (gmt 0)|
|I think that the fresher ( 72.14 ) index is cleaner: less spam compared to 64.233 .... . |
There are two different (at least) data sets on 72.14*
You'll have to be more specific to have a useful conversation.
| 4:32 pm on Jun 19, 2006 (gmt 0)|
Sid! happy day - I'm in Copra, for now - Cheers.
| 5:20 pm on Jun 19, 2006 (gmt 0)|
Back in there as well, well to be precise I was always there but they were discounting the homepage and its inbounds.
Credit due they fixed it. As we knew they would.
| 7:13 pm on Jun 19, 2006 (gmt 0)|
Why does Copra show results for my site with a title from DMOZ, and Turd show results for my site with a title as it is online?
| 7:14 pm on Jun 19, 2006 (gmt 0)|
hehe...copra retreating again (is monday you see)
| 7:18 pm on Jun 19, 2006 (gmt 0)|
"Why does Copra show results for my site with a title from DMOZ"
I sayed previously here that many sites have DMOZ titles and that is not right those results you keep on worshiping is just another test that IMO has to do with recalculating trust rank and page rank ,as well withi ndexing priorities (that means DMOZ listed pages have more weight)IMO again.
| 7:40 pm on Jun 19, 2006 (gmt 0)|
"Copra results in retreat"
Can't see that from what I'm looking at still on 35 DCs.
| 8:11 pm on Jun 19, 2006 (gmt 0)|
> You'll have to be more specific to have a useful conversation.
but now it displays old cache.....
Google is a joker at the moment...
| 9:18 pm on Jun 19, 2006 (gmt 0)|
As mentioned by Adam recently, and as most of you might have noticed, our good friends at Googleplex are having problems.
Lets give them sometime to fix things and wait and see what will come up in August 2006 where most of Google's employees are expected to be back from their loooooooong vacations ;-)
| 9:21 pm on Jun 19, 2006 (gmt 0)|
>> "My positions are junk, but I still get a lot of traffic" <<
>> "I have high rankings, but I am not getting any traffic" <<
You have only been looking at about one-third of the Google datacentres that actually exist; so, while you were seeing one result, potentially two-thirds of the world were actually seeing something completely different.
Today I have been working on a list of Google DataCentre IP addresses [webmasterworld.com], and the new list is nearly three times the size of anything that I have seen published anywhere else.
| 10:50 pm on Jun 19, 2006 (gmt 0)|
Maybe I'm totally dumb but can someone show me again what is TURD and what is COPRA?
| 10:55 pm on Jun 19, 2006 (gmt 0)|
i saw copra retreat to 29 DCs, now it is up to 42 DCs.
long live copra...lol.
| 11:40 pm on Jun 19, 2006 (gmt 0)|
< You have only been looking at about one-third of the Google datacentres that actually exist; so, while you were seeing one result, potentially two-thirds of the world were actually seeing something completely different.
That's one explanation. The other explanation is that a site's really deep pages are no longer ranking.
For example, I have about 20 pet searches that I check to monitor Google SERPS. On all of those searches, on both Copra and Turd datacenters, my site is doing great.
However, overall Google referrals are down about 50% since Friday.
My site happens to do a ton of traffic on obscure, long tail type searches. However, it seems that these searches are no longer referring traffic, while the larger, more competitive searches are.
My sense is that this has something to do with the bizarre stuff that's going on with supplementals and site: counts. Or the fact that others have managed to inject billions of stolen and/or autogenerated pages of crap for every conceivable search into the index.
It will be interesting to see if this is just a "bad data push" or really is an update pertaining to how Google manages obscure searches.
| 2:10 am on Jun 20, 2006 (gmt 0)|
You think now, that Google is such a large company, that some people would have left google, and have given a hint on the general workings of the company.
Im not talking trade secrets, or the intricate details of their algorithm, just general concepts, as it currently stands, such as the number of datacenters, average number of computers in each datacenter, volume of data being managed, diaster recovery, etc. Hey this is google, not the NSA. This information should be out there.
| 3:17 am on Jun 20, 2006 (gmt 0)|
I agree, hey I know a few guys over there and know a few things but it would be better if an official channel of communication was out there so that everyone could understand the basics of how Google uses IPs and their resolution to the versions of their index.
As much as I can say is that the IP addresses that are resolving to different versions of the index are (allegedly I must add) a network set of computers that were previously not public until they were expanded since bigdaddy - however the association to "data centres" is a different concept that I don't think is the same anymore.
| 3:34 am on Jun 20, 2006 (gmt 0)|
I am noticing the quality of the search results are improving, and the size of the results sets is decreasing for the SERPS I watch on the main live google.com search.
My site disappeared from two very competitive searches last week for about 4 or 5 days. Then it reappeared over the weekend. Now the size of the result sets is decreasing. Decreasing in one case by 25% or 100M pages (was around 400M, now its around 300M).
I agree with the theories that index is being rebuilt as my observation is that the quality is slowly but surely coming back.
| 3:53 am on Jun 20, 2006 (gmt 0)|
I have to ask, with all those dc's how do you get your work done while you're watching them?
Sorry I just had to ask.
| 5:15 am on Jun 20, 2006 (gmt 0)|
|Today I have been working on a list of Google DataCentre IP addresses, and the new list is nearly three times the size of anything that I have seen published anywhere else. |
The SERPs at <the online tools> now reflect all of these datacenters. (Normally I would not have posted this except that it is directly relevant to this particular discussion.)
[edited by: tedster at 3:58 am (utc) on June 23, 2006]
| 8:18 am on Jun 20, 2006 (gmt 0)|
<An online tool> shows 97 dc's without copra.
[edited by: tedster at 3:59 am (utc) on June 23, 2006]
| 8:34 am on Jun 20, 2006 (gmt 0)|
but a lot more showing the copra results
| 8:41 am on Jun 20, 2006 (gmt 0)|
There are now at least four disting datasets.
Here are current examples of each:
126.96.36.199 Copra Variant
188.8.131.52 Original Turd
184.108.40.206 Turd Variant
| 8:46 am on Jun 20, 2006 (gmt 0)|
Pete looks like youare in love with copra :)
| 8:52 am on Jun 20, 2006 (gmt 0)|
I like the turd results better myself.
Separate matter: Any speculation what the new Google DC in The Dalles, Oregon might be all about?
Is this just another way to distribute search muscle,
to cut power costs etc, or is it for something new? -Larry
| This 179 message thread spans 6 pages: < < 179 ( 1 2 3 4  6 ) > > |