Just checking in response to this Google question and I am seeing PR=0 for all internal DMOZ pages that i;ve checked (not an exhaustive sample but random checks). Is this a sign of things to come for DMOZ? Or is this just another Google glitch?
You're right, CReed. The problem of no PR pages doesn't exist if you drop the www. Looks like DMOZ could use some help from a webmaster with a little knowledge of SEO, PR, and mod_rewrite...
The one category where the PR just doesn't seem quite right to me is here - and if you check Google's Directory the same category is PR0 and has been PR0 for some time.
That is strange, CReed. The parent category is PR=6 and a couple of sub-subcategories have decent PR (for example, this one is PR=5), but most of that branch is has no PR, at least as displayed on my toolbar. My guess is that Google got complaints alleging that the category (probably particularly this subcat) has no other purpose than manipulation of PR, so they reacted accordingly.
I was not offering my own opinion but hypothesizing that Google has gotten complaints about it. Personally, I'm agnostic about that category. I don't edit there (at least not on purpose), but I don't object if other people want to spend time maintaining it. Knowing, as I do, that the category is criticized in webmaster forums such as this one, I assume that Google has received complaints.
I didn't even know it existed until today. I guess my initial reaction is it seems a bit self-promotional but I don't personally object to it. It's not hurting anyone like some of the other categories I am continuing to attack. Bottom line for me is I don't care if it stays or goes. There are a lot more important issues for DMOZ to be worrying about.
I see things slightly differently Orlady, though it makes little odds. I don't get the feeling that Google is driven by complaints. They have to struggle against the manipulation of PR. And with all those PhDs available, I doubt if they would need anyone outside Google to point the finger at the obvious. Google's current battle against ODP clones is part of a wider anti-spam front. It has switched off PR for the entire directories category of its own ODP clone: http://directory.google.com/Top/Computers/Internet/Searching/Directories/. It went through a phase of switching off PR for the equivalent category in the ODP, but that now has a little PR, while Sites Using ODP Data has PR7. The change is fairly recent.
Remember of course that whatever the PR of the ODP category, it doesn't automatically mean any of that is passed along, if Google chooses to discount it. PR isn't a measure of outgoing links - it's a measure of links pointing to that particular page. The DMOZ page could be PR10 but if Google wishes it could easily blobk it from passing any PR whatsoever to other pages.
I don't think that's exactly how it works Minstrel, though the end result may be the same. The calculation of PageRank is automated. If a page has PR then that is passed on via links on the page (as long as the process is not blocked by nofollow or whatever.) Unless ... If Google wishes it can impose a penalty on a page or entire site by taking it out of the calculation of PR. So for example Google imposed a penalty on SearchKing for selling links. So SearchKing had zero PR. The incoming links to SearchKing counted for nothing, because SK was out of the loop. So let's suppose SK had a link from a Dmoz page with PR6 (as indeed it did until recently), that link passed on no PR at the time of the penalty because it couldn't do. SK was out in the frozen cold. I would guess that Google has been busy imposing penalties on directories that it sees as link-selling. It might well prefer that kind of direct action to the indirect approach.
Normally, yes. But as you yourself go on to say, that is not always the case: Not quite. In that case, the capacity for DMOZ to pass on PR wasn't affected and wasn't the issue. Outgoing links from SK were prevented from passing on PR. If memory serves, the SK site itself was also penalized (which is not the same as discounted). That's not a guess. Google has explicitly said they are doing that. It's not necessarily a matter of a "penalty" per se. As you indicate, Google can and does discount bought-and-paid-for links and does so regularly. They also can discount other types of links they don't feel are "natural" and I believe they do that regularly. They also can and do apply duplicate content filters to sites like DMOZ clones. The difference is that the site from which the links are discounted does not look any different (unless of course the site really is penalized, in which case it disappears from the index entirely). Thus, the fact that a page in DMOZ has recently increased to PR7 means nothing in itself, since as I said PR is affected ONLY by incoming links.
That's an interesting thought Minstrel. You are right that Matt Cutts has been warning that buying links will do people no good. Is there a new mechanism for discounting them? The simple way to do it is to impose zero PR on the link-selling site. As far as I am aware that is exactly what Google did to SearchKing. Simple and effective. If a site has no PR, it can't pass it on. However it is possible that Google has come up with a more subtle method that leaves PR intact on the linking site. I can't say I have heard anything about it though. And would this be a sufficient deterrent? If people are buying links on high PR pages, then the most effective deterrent is to remove the high PR. What I have heard about is some directories going PR0. How many I wouldn't know. I haven't been paying close attention.
Minnie, a LOT of categories in DMOZ don';t have PR. Remember that the directories/ category had a BLANK N/A PR up until some time go ? Now it's 0.. I think it will never increase. Several other categories lack PR (intentionally).
That is a penalty. The offending site itself has PR removed and is removed from the index. That's not what Cutts is talking about when he talks about discounting links, though. He's talking about blocking the page/site from passing PR on via it's outgoing links. http://www.mattcutts.com/blog/tell-me-about-your-backlinks/ http://www.seroundtable.com/archives/002818.html http://www.free-seo-news.com/newsletter177.htm#facts
Ah! That's very interesting. I had read the stuff on Mr Cutt's own blog and the comments by Rustybrick, but not the crucial last item. [Edited to add] Found the original source: http://www.mattcutts.com/blog/2005/09/page/2/ where Matt Cutts also talks about SearchKing.