Forum Moderators: Robert Charlton & goodroi
And is the news on "nofollow" attribute handling as well as the new GWT interface announcement and rollout in fact here to distract SEO world on what is really happening?
I might be wrong but what I am seeing is the following:
1) May/June changes in SERPs - just read a long discussion on a separate thread here
2) The new GWT, which not just introduced a new interface many talk about, but in fact changed how inbound (and perhaps outbound) links are counted (e.g. no more 301 inbound links) and perhaps other things?
3) Big discussion about nofollow on internal links + PR sculpting, a discussion which to me seems to be out of proportion to acknowledged influence of page's PR in relationship on how the page ranks in SERPs
On top of this, I am seeing in new GWT that Google suddenly "found" a large number of old links of ours, which were present before we did changes in URL structure and is trying to crawl them. Only noticed because they got reported as errors in crawl. I am not talking here about a link or two, but about hundred or so links using old syntax we abandoned and which were never before in GWT as either internal or external links. These links are nowhere on our site nor are being linked from "wide world", but suddenly there is a surge of them being crawled, whereas these have certainly not being crawled since at least Feb.
It almost seems to me that Google pulled these "old" links up from its own data storage from somewhere and is retrying them.
And further, we just got a surge of Google alerts for few articles we syndicated over 18 months ago and for which we already had Google alerts coming 12 months ago or earlier. So why again the same alert?
Without wanting this to sound like a conspiracy theory, but I am wondering whether the announced change in nofollow handling and the new GWT interface are there to distract from other perhaps major change that is happening right under our noses?
Kind of like world politicians, who, when they want some changes to go unobserved, they make them at the times when there is some other major news that distracts general public.
Plus old links being re-crawled...
And there is a question already asked on GWT thread about keywords missing from the new GWT - with the comment that they should have the data already, so why it is not there?
Or maybe it is so simple as for Google to release a new version of (GWT) product in such a "buggy" state and hence things are not showing?
Just makes me wonder - too many coincidences maybe.
JohnMu: ...we built a new backend for gathering links data as part of our new UI and we decided not to follow 301s. The reason we did this was a lot of users were trying to go to the source page that we mention has the URL and were unable to find it since we were showing the final destination URL rather than what was specified on page.We understand this change has solved issues for some users, but created new problems for others and we are going to look at ways to solve this problem.
Webmaster Help Forums [google.com]
Webmaster Tools gathers links data from the main data set at Google. This comment in its full context makes it pretty clear that it's only the reporting that has changed. Not following 301's at all would be a HUGE change - and a stupid one, too. Google is many things, but they're not stupid.
Wouldn't one normally assume that these users own both domains and that they would know if one of them redirects 301 to the other? This would be much more common scenario then a third party redirecting 301 to your domain hence leaving you oblivious to where the link to your page is.
On the other hand, if 301 is not followed when getting links data, does this mean that anchor text of redirected link will also not be reported to belong to a target domain? And what about passing PR?
Just wondering...
Wouldn't one normally assume that these users own both domains...
There is all manner of funny business on the web. Consider one example - incorrectly configured click counting scripts on external links. There is also a kind of disruptive webmastering that happily experiments with 301 and other redirects, just to see what kind of mayhem they might be able to create.
But getting back to your overall point, there is certainly a lot of churn and (from what I see) spidering right now. Google has had such periods in the past - and apparently we've got one now. It's the beginning of summer and in years past that was also a time when we saw similar things.
I believe if the changes to this attribute were being folded in, they would be done so in manageable increments as it is quite likely flipping those type of changes over wholesale would cause some very strong changes.
Today I had a hit from googlebot on default.html - I've NEVER had a page named that on ANY of my/customer sites.
I'm quite prepared to believe there are a lot of other instances I haven't seen - I only caught this one because it triggered an attack trap on the page name "default" (a common exploit/scrape attempt).
And... Just checked this month's good-bot log for google.*404 and got 309 of 'em. My sites don't lose pages very often so aakk9999 could be right about a re-scan of old pages.
Mind you, the same scan for msnbot got me about 540 hits and yahoo 430. Not that either is a surprise.
Also note, whether a nofollow attribute tells a search engine that the page linked to is "unimportant" is not the change that went live. The change is in how a nofollow link affects the PR vote of other links on the page. I still use nofollow for internal links to utility pages that I don't want to see in sitelinks, for example. It's easier for me than hawking over the sitelinks and deleting them in WMT.
We've got two other [webmasterworld.com] threads [webmasterworld.com] where we are discussing the nofollow - so let's not have a third.
So now we return this thread to it's regularly scheduled topic: Is Google doing a major crawl right now to re-evaluate links?
Thanks for your reply, tedster. I read about and had also read the full blog post of Mr. Cutts. I fully understood the changes that Matt discussed, prior to my post here and I appreciate that you shared your knowledge about the matter.
[edited by: tedster at 6:19 am (utc) on June 21, 2009]
[edit reason] member request [/edit]
Then again, there's the packrat syndrome: I got it all, I want it all, and I want it all the time, if only I have the time to check to see if I have what I've forgotten I have.
At what point does information glut break the machine? Not sure, but I suspect it's not that far away given the MFA and scrappers, etc. out there. And the gamers, too.
One thing they "could be" doing is getting more current data to see how webmasters are responding to the announcements about PR sculpting, nofollow, javascript links and so on - "Are webmasters going crazy with changes?"
But even that seems not so likely to me. Most webmasters do not follow Google's or Matt Cutts announcements - many major corporate web teams (and certainly smaller site owners) don't know who Matt Cutts is. Google goes through cycles of heavier spidering and then lighter spidering throughout the year, depending on plans and needs that we'll never know about. This is more likely to be just that kind of business-as-usual, I think.
Also of interest is suddenly the Google blog widget that shows incoming links to blogs in the Wordpress admin dashboard, which has not worked for some time in not showing any links for many users, suddenly lit up on all my blogs (6) and all the links I have been getting are suddenly coming in like crazy.
