Welcome to WebmasterWorld Guest from 184.108.40.206
>> Homepage Gone <<
You need to check very carefully that it isn't indexed at an alternative URL:
For a lot of sites, some other variant is indexed and it can take a bit of searching to discover.
Also, make sure you are very intimate with all of these searches:
because there are some hidden surprises (nasty ones) to be found there for many sites.
Oh yes, I do most of my searches with &num=100 on the Google URL, and I always check both with &filter=0 and without it.
[edited by: Brett_Tabke at 1:29 pm (utc) on June 17, 2006]
Something's definitely badly wrong in the plex - it seems some things are tripping over other things, bugs on bugs.
Interestingly a new site we put up with one IBL has cached fully it's 378 pages in 2 months.
Shame about the rest - i think the "plex" can't currently handle complex sites right now [ forgive the rhyme ].
Keep on smiling - it's the only way to go :)
Some of this stuff is just too weird to be important. For example, I'm looking at one single word search in the name of all the sites in question where some pages (all in DMOZ) have DMOZ titles and descriptions, some have DMOZ titles but meta descriptions, while others have meta titles and meta descriptions. No rhyme or reason to the mix.
It seems like a good bet that we will just have more or less gibberish serps for a week or so more, then settling down, then Matt coming down from from Black Mountain...
Ive been posting to that effect also a few times and have conclusive proof.
I cleared all cookies, internet files etc. A friend of mine who lives about 25 miles away from me in the UK did the same. We then logged onto THE SAME data centre number and did about 8 keyword search requests. The results were different in both locations.
Make what you want of that but maybe its a feature of the new infastructure that Google results are different over the data centres depending on your location so that sites get overall less traffic as its spread about more.
Just thought I'd drop by and say that we now have 37 DCs on Copra results. The highest yet even showing on 72.** DCs what is happening is this the end and it is all proprogating and why is it always at the weekend the copra results show on the 72.** DCs are they testing things on them?
Let's wait and see maybe celebrating tonight I think.
Might be a random fluctuation, or perhaps they're finally starting to use the Copra DCs' for ordinary searches performed by real users (as opposed to just being run in the background, and available via McDar and other DC-specific searches).
If so, this could have a noticeable impact on some sandboxed sites, and others which will suddenly become more visible.
Copra results are being seen by real users and have been all along AFAIK. If you look at the thread on UK default DCs you will see that we have had a mix of good and bad results as the defaults.
Its wrong to assume that those DCs have not been serving someones standard results since the time they were spotted on McDar. I've noticed that I'm getting occassional surges in traffic (and troughs as well worse luck) and assume that this is when a good (Copra) DC is being aimed at a major UK ISPs customers as the default Google search.
Almost a half of them have reappeared back in the index in the last few days (but as many are for PDF files, I cannot ascertain their cache date - I guess that they are re-cached with dates that are newer than 2005 June but I cannot prove it).
The problem is, that the new pages that have been added back have incorrect titles. The bot is grabbing the title tag, then grabbing the first part of the text that is on the page and cramming it into the title until it's full. Since I repeat my title on the page, it looks something like: "Welcome to my webpage! Welcome to my webpage! Blah Blah text from the page" for the titles on everything.
I'm sure, of course, that this is not a bug but rather a reprioritization of the Titling feature, and that if I would simply get better backlinks and stop spamming all the time I could get real titles back.
Strangely, the 72.14.207.x datacentres still show just the old one-word titles. These are pages that have just reappeared in the index. They were for duplicate content Supplemental Results that Google appeared to throw away last month. I hoped that they had gone forever - now, just a month later, they are baaaack (but are not back in all datacentres)!
<edit>the pages that had been dropped for the last month now also appear with long titles in 64.233.187.x in just the last few minutes. The data is spreading.<edit>
[edited by: g1smd at 3:00 pm (utc) on June 17, 2006]
Brett Has chosen this thread as a Featured Home Page Discussion. Thanks Brett!
And of course; Congrats to all kind fellow Datacenters Watchers!
Back to those unpredicted DCs which has been the case for at least the last 6 months or so ;-)
Within the sector I watch (related to online advertising & marketing), I see the following DCs set represents to some extent more relevant search results
Of course, you might have different opinion than mine, depending upon which sector you watch.
They will show the redirected URLs as Supplemental Results for several years, usually with a very old cache copy attached to them.
As long as the new URLs for the pages are indexed and do rank, and as long as anyone that clicks on any of those old links does get to the correct new page on your site, then you have nothing to worry about.
At that time, there were a lot of theories going around, but in the end all settled and homepages returned.
Having said that Florida hit a couple of months later....
I am also trying to get rid of a subdomain that no longer exists. Has been gone for over a year. However, the items are still showing up in supplemental.
If something shows in the supplemental - can it cause your site to have a penalty?