Quote:
|
Dave, I followed that entire thread at wmw. I think most anyone with a bit of web savvy could figure out the process of redirect 301\302 or meta refresh hijacks from reading that thread. Makes me kind of wonder why everyone is making such a big deal about posting step by step instructions.
|
I don't have any problem with these type of instructions being posted. Dave, if you want to go into depth, feel free.
Here's the point. I've talked with Google on this subject and have been told they don't see it as a big problem, if much of one at all. If that's indeed the case, then posting the instructions do nothing. Put them up, so everyone has a clear idea of what the concern is, but we're all safe in the knowledge that in reality, Google says this isn't a problem. There is no loaded gun to worry about.
If it IS a problem, then posting the instructions still doesn't hurt anyone but the people who would ultimately misuse them. That's because if the technique works as described, then Google will have to come up with a solution to it in short order.
FYI, my understanding of the way it operates is that if you redirect a URL that you control to point at someone else's URL -- and if your URL is of a higher PR value than the URL you point at, then you can trick Google into thinking YOU have the original URL.
So say we have this situation:
Page A ranks tops for "cars" and has a PR value of 6
Page B is someone who copies Page A and gets indexed, so they appear as a mirror of that page. Google sees this but sticks with Page A because that page has a higher PR value than Page B.
Now Page B gains PR score of 7 somehow. They then redirect to Page A. Google sees this and decides that Page B must be the best URL to show. So the top listing gets replaced with Page Bs address.
Now, what good is this to Page B? After all, it's still pointing over to Page A. Well, the redirect stops, so that the traffic for this "cars" listing can be pointed elsewhere. And eventually, if Page B's content is different than Page A, Page A should pop back up. That's also why this doesn't appear to be a major problem for many people -- it seems to be a short term thing that's going to work more in cases of pages with low page rank values. Amazon is unlikely to get hijacked.
That's my understanding without digging through the notes I have from talking with Google and others about this. It's been in the works for me to do my own article on it. Others can feel free to dive in and correct me or enhance as appropriate.
And I'll say it again -- the point of the thread or posting steps is
not to help people hijack pages but to determine if this really is a problem people need to worry about.
If so, then it will undoubtedly get corrected once we get some real examples of it out in the public. And the fact that GoogleGuy is going to help encourage people to at least provide more private examples is to be applauded. I for one would like to see the situation solved finally one way or another, rather than the months -- literally months -- people have been discussing it.