tedster
msg:4331073 | 8:53 pm on Jun 25, 2011 (gmt 0) |
It does sound like spam, but detecting it has nothing to do with Panda.
Panda is about detecting shallow content and that lies in between spam (Matt Cutts' team) and regular algo (Amit Singhal's team). Because Panda's target lies in between the two area, they are collaborating on the algorithm change that deals with it - and focusing on that for this entire year.
Self linking pages are an interesting spammer tactic that have been cropping up more and more. Maybe there is a big loophole here - or maybe there's so little content online for the terms where I find this method that, spam penalty or not, it's still among the few relevant results available and ends up ranking by default.
|
Planet13
msg:4331079 | 9:16 pm on Jun 25, 2011 (gmt 0) |
| ...or maybe there's so little content online for the terms where I find this method that, spam penalty or not, it's still among the few relevant results available and ends up ranking by default. |
|
Well, for the example I was looking at, that could definitely be the case.
still, you have to think that any page that links to itself nearly 400 times (using different anchor text), it must have SOME negative consequences when it comes to flowing page rank, right?
maybe it's just a temporary loophole...
|
lucy24
msg:4331086 | 9:46 pm on Jun 25, 2011 (gmt 0) |
Worried query: How is this different from the ordinary internal link ("such-and-such vaguely related subject, which I'll blather about elsewhere on this page")? Or, I guess: how can a computer tell the difference?
|
tedster
msg:4331088 | 9:51 pm on Jun 25, 2011 (gmt 0) |
The page links to ITSELF 394 times, and each time the link uses different anchor text. There's nothing ordinary about that, and it's certainly not done for visitors. Clearly it's an attempt to game Google.
I would think machine intelligence could easily spot that pattern - and indeed I have seen pages in that vein get penalized.
|
netmeg
msg:4331097 | 11:08 pm on Jun 25, 2011 (gmt 0) |
| The page links to ITSELF 394 times |
|
Narcissism?
|
lucy24
msg:4331109 | 12:29 am on Jun 26, 2011 (gmt 0) |
| The page links to ITSELF 394 times |
|
You mean the actual html says <a href = "http://www.example.com/the-page-you're-already-on.html">?
Yeah. I think your average robot could figure that one out. Though note that in gwt, the page that lists "anchor text" makes no distinction between your own site- or page-internal linking text ("elsewhere on this page" or "in my article on obsolete widgets"), and linking text from other sites ("this incredibly fantastic site I just discovered" or "avoid this site like the plague").
|
Planet13
msg:4331115 | 1:41 am on Jun 26, 2011 (gmt 0) |
| You mean the actual html says <a href = "http://www.example.com/the-page-you're-already-on.html">? |
|
Yup!
And it is just a list of "search phrases" people used to find the page. One per line.
I've seen something similar where someone scraped a bunch of QUESTIONS from Q&A forums related to a subject and posted all those questions on one page - without providing any of the answers.
Ranked number one for LOTS of those questions, despite not having any answers.
|
tangor
msg:4331142 | 5:16 am on Jun 26, 2011 (gmt 0) |
Single page anchor text spam. Panda is not seeing that at the moment. (Panda is a warm-fuzzy G modification). Raccoon will (mark my words!) dig deeper as raccoons are not just warm, but are fuzzy. The following Aardvaark rollout will deal with inter-domain anchor text AND networked anchor text...
|
tedster
msg:4331149 | 6:05 am on Jun 26, 2011 (gmt 0) |
| Panda is not seeing that at the moment. |
|
I don't think Panda is intended to see that kind of thing. It's core spam, not shallow content that falls in the "no man's land" between the spam team and the main algorithm team.
I agree, the day's of this kind of spam are numbered, but we just aren't there yet. But for anyone who hasn't noticed what jumps out here - note that on-page anchor text seems to be a strong relevance factor for the page where it appears, not just the target page. In fact, this factor sometimes ends up making the "wrong page" rank higher that the "right page" which is just one click away.
|
mirrornl
msg:4331264 | 8:17 pm on Jun 26, 2011 (gmt 0) |
| note that on-page anchor text seems to be a strong relevance factor for the page where it appears, not just the target page. In fact, this factor sometimes ends up making the "wrong page" rank higher that the "right page" which is just one click away. |
|
already for more than 5 years this is the case
|
wheel
msg:4331276 | 8:55 pm on Jun 26, 2011 (gmt 0) |
300 times might be a bit much, but I used to do this once or twice on my homepage, and it worked wonderfully. Ranked on terms for no other reason than I dropped a self-referencing link on my homepage.
|
|
