Google Blogoscoped

Forum

Badly Ranked Wikipedia Watch  (View post)

Seth Finkelstein [PersonRank 10]

Friday, December 1, 2006
13 years ago7,709 views

"searching Google for wikipedia watch does bring Daniel’s site up on top"

I think that's a DMOZ whitelist (though I'm not completely sure)

Seth Finkelstein [PersonRank 10]

13 years ago #

<em>"and warns of "SEO superstition"... the attitude that something must be buggy with Google if your site doesn't rank well in it"</em>

Actually, by "SEO superstition" I mean more the phenomenona of bad theories in general.

Tadeusz Szewczyk [PersonRank 10]

13 years ago #

I get Wikipedia Watch at one for wikipedia watch. Besides it's a rather new site and Google does not trust new sites. It prefers sites that has been there a while.

"Online Since: October 13th, 2005" according to URLtrends.com

As you can see with this tool
prinformer.com/index.php?u=www ...
the site just received their PageRank of 5. Some data centers show still the old data set (PageRank 0) so that's why sometimes you can not see it in your Google Toolbar depending on what date center handles your request.

So this site experiences the so called "sandbox effect":
webpronews.com/insiderreports/ ...
Her the Wikipedia link, ;-) :
en.wikipedia.org/wiki/Sandbox_ ...

Seth, you don't need to ckeck results manually, try one of these:
blogoscoped.com/archive/2006-0 ...

Indeed in this case we see that the Google sandbox really works. We have a guy here gaining publicity by bashing a great project for a year now. That's no reason to show him as a very important ressource. If he establishes a thorough watch site that stands the test of time – great, then Google will reward it in due time.

Otherwise anybody with a new site would outrank people that are experts for years. And you don't want dirty, stinking, ugly smirking SEOs to manipulate Google ;-) in order to pay for their expensive black hats, do you?

Seth Finkelstein [PersonRank 10]

13 years ago #

It's been more than a year. And the site's gotten a lot of "organic" links. It *SHOULD* be out of the sandbox by now. It's not a matter of not getting the ranking boost from being a trusted site. He's way, way down.

Yes, I could have used a ranking tool, but I wanted to check by hand to make sure that the result was really that bad.

mc [PersonRank 3]

13 years ago #

Its a shame Google don't do RantRank, as I'm sure Brandt would come out on top for every search

SirNuke [PersonRank 1]

13 years ago #

Just to note, this guy also runs Google Watch (google-watch.com/). He also been complaining about Google supporting Wikipedia by lowering his results for quite some time.

I think he's an arrogant [personal attack removed] who doesn't deserve to be listened to, but that's just me :)

Seth Finkelstein [PersonRank 10]

13 years ago #

Indeed he has been complaining.

But, are we agreed that his SERPS are unreasonably low?

This is what I'm trying to get at – his *theory* is probably wrong. But there's a *problem* somewhere, independent of the personal issues.

Dan Tobias [PersonRank 6]

13 years ago #

I don't know... just what is the "reasonable" positioning for a whiny, annoying malcontent who likes to bash stuff?

Amit Patel [PersonRank 1]

13 years ago #

superstition? or facts?
Google has a lot to hide regarding how (and why) they rank things.

One documented case: idcide.com/affair/

Ionut Alex. Chitu [PersonRank 10]

13 years ago #



"Google's central PageRank system has been criticized. Some, such as Daniel Brandt, call it, "undemocratic". Common arguments are that the system is unfairly biased towards large web sites, and that the criteria for a page's importance are not subject to peer review. PageRank is a largely automated system which is impartial insofar as it knows no personal bias."

en.wikipedia.org/wiki/History_ ...

Daniel Brandt [PersonRank 3]

13 years ago #

Dan Tobias asks, "I don't know... just what is the 'reasonable' positioning for a whiny, annoying malcontent who likes to bash stuff?"

At the same time, Mr. Tobias as a Wikipedia editor constantly opposes any effort by me to get Wikipedia to delete my biography:

2005-12-20: "We should basically ignore all his sniveling..."

2005-12-30: makes false allegations about the legality of NameBase on this forum

2006-04-07: incorrectly claims that I edited Wikipedia under a look-alike misspelling of my user name

2006-09-08: on my notability: "...our bar for inclusion is generally lower than most of the mainstream media."

2006-09-09: "I'd like to see his whacked-out ideas tested and defeated in court."

2006-11-07: "Yet again he's featured in a major mainstream news outlet... he continues to get himself increasingly notable, during his quest to get himself removed from here as a non-notable person."

Today he insisted that my Ninth-Circuit-reversed conviction for public draft resistance in 1969 continue to be mentioned in my Wikipedia biography, even though Jimmy Carter restored all civil rights for draft violaters in 1977. How many employers do you know who read resumes past the point where "convicted felon" appears? Of course I leave it off my resume, but then how may employers "google" applicants? A huge percentage, according to the New York Times, and it is increasing. In Finland it's illegal for prospective employers to "google" applicants, but why should I have to move to Finland?

So now Mr. Tobias suggests that burying my Wikipedia-Watch site in the rankings is an appropriate response for Google, while Wikipedia maintaining my spin-infested biography, where it shows up as number one in all the search engines, is entirely appropriate, even though it is harmful to me. In his eyes, my Wikipedia biography now stands as an appropriate punishment for my efforts to get it taken down. Even Jimmy Wales would choke on that one, because it violates a new Wikipedia policy on biographies of living persons. This policy is too little, too late, as there are 142,766 biographis of living persons on the English Wikipedia. And with Wikipedia's toleration of editors like Mr. Tobias, it's almost certainly the weakest link for Wikipedia in terms of potential legal problems.

Who is the whiner in this case – Mr. Tobias or me? If both of us are whining, who has the most cause to whine?

Philipp Lenssen [PersonRank 10]

13 years ago #

> a new Wikipedia policy on biographies of living persons.

Do you have a link?

Daniel Brandt [PersonRank 3]

13 years ago #

en.wikipedia.org/wiki/WP:BLP

"Biographies of living people must be written conservatively and with due regard to the subject's privacy."

It's still very weak on invasion-of-privacy issues, because it is concerned mainly with defamation of character. But Florida, where Wikipedia's servers are located, has invasion-of-privacy laws in addition to libel laws.

Martin Porcheron [PersonRank 10]

13 years ago #

Let's not forget, this is the guy who hates Google because he didn't get a high pagerank after spending a few hours making a site which he claimed was worthy – so he started an anti-Google capaign.

Seth Finkelstein [PersonRank 10]

13 years ago #

Umm, does it bother any of the carpers that they are in fact agreeing with Daniel Brandt, saying that Google rankings are political? I mean, I keep saying basically that Google rankings aren't political, so there must be some sort of algorithm problem here, likely related to anti-spam measures. People keep popping up and replying, *in effect*, no, Google ranking are *political* – though they don't phrase it that way. They say Brandt is [various negatives], therefore presumably it is right and proper to have a low ranking. But this literal ad-hominem argument (rare!) only makes sense if Google's ranking are political. This does not seem to have occurred to the people making it.

The Wikipedia Watch site ranks highly for relevant keywords on Yahoo and MSN. This indicates the site does have good linkage. It's been up for more than a year, so "sandbox" shouldn't be in effect. So there's a problem. It would in fact be quite disturbing if ad-hominem was the answer.

Daniel Brandt [PersonRank 3]

13 years ago #

Mr. Porcheron says, "Let's not forget, this is the guy who hates Google because he didn't get a high pagerank after spending a few hours making a site which he claimed was worthy – so he started an anti-Google capaign."

Ahem, some of the participants on this forum are meaner than the six long pages of Talk comments associated with my Wikipedia bio. NameBase, which I felt deserved a higher PageRank on its thousands of inside pages, was used as an example for an essay I wrote that criticized PageRank. My home page PageRank for NameBase, which was fine, wasn't distributing to deep pages very well. This was the first essay on a site I started in August 2002, Google-Watch.org.

NameBase was started in 1982. That's 20 years before I started Google-Watch. It was successful in the late 1980s among Washington DC journalists and researchers. This was before the web – I was mailing out disks. NameBase went online in 1995 via telnet, and was converted to the web and HTML the following year. So it was on the web before Google existed, and when the very first engines were getting underway. I wouldn't call that "spending a few hours making a site." Not even Wikipedia would call that "a few hours making a site." But that description does just fine among some readers on this forum.

Now you can see what I'm up against when it comes to certain Google-worshiping editors on Wikipedia. Why do I even try?

Ryan [PersonRank 0]

13 years ago #

Umm.. lets see

1.) if you were convicted of a felony, they have every right to list it. It's public fact. Don't dodge the draft next time.

2.) Who says your page is relevant? I'm assuming you mean the term "wikipedia" correct? Let's examine features a relevant site would contain:
   a.) actually wikipedia – nope
   b.) A recent news article concerning wikipedia – nope, you're not "news" in the sense CNN or new york times is.
   c.) multiple user reviews / discussion of wikipedia – nope.. you're just 1 person.

The main reason you rank so low is because wikipedia has a TON of sites, in a ton of languages. All of which are more relevant than a 3rd party site (for obvious reasons)

The secondary reason is you only offer 1 opinion, and it's not a recognized journalistic body.

Now, lets look at possible spam penalties:
You use the word wikipedia WAY too much in your writing. SO much that it starts to get awkward to read.
You have a LOT of content on one page. Break it up... but a lot of that content exists on other places on the web. With so many quotes in one place, you might appear to the bot to be a scraper

ok, so how can you be relevant?

Add user comments and unbiased reviews / discussion.
Link to relevant news about wikipedia from other sites... sort of like a wikipedia in the news aggregator.
Offer tutorials or help, or advice. anything other than one person's viewpoint.

The fact is simple: one person's viewpoint is hardly relevant when that one person isn't a member of the established press, or famous.

It's not that google is trying to shut you up. It's just that your site isn't relevant for the terms you're trying to show up for.

Let me refer you to an article I wrote about just that: shoutwire.com/comments/15638/I ...

Philipp Lenssen [PersonRank 10]

13 years ago #

But it's gotta take more than a repeated keyword to get you banned, if that's what happened (I don't know if that happened, but the PageRank is grayed, which is weird for a site that's a year old). E.g. if you link to a bad neighborhood, or you use hidden links, or you use cloaking, now that might be a reason – not sure if Wikipedia-Watch is guilty of any of that. I guess only Googlers know for sure?

Seth Finkelstein [PersonRank 10]

13 years ago #

Ryan – read the original posts in this thread. The explanation "isn't relevant for the terms you're trying to show up for" CAN'T BE CORRECT, because the site ranks highly for relevant terms in Yahoo and MSN.

People are so reflexively knee-jerk flaming Daniel Brandt that they just won't think this through.

But you may have hit upon something in bringing up the scraper-like penalty.

Ryan [PersonRank 0]

13 years ago #

Seth..

Google, yahoo, and MSN have different definitions of relevant.

It's not logical to say that because something ranks well in yahoo or MSN it should rank well in Google too. They don't use the same algorithms, they don't value the same things. Hell, MSN is the easiest to game.

I've never heard of Daniel Brandt before this, so I have no desire to flame him.

Pagerank itself is a measure of the sites that link to you. I don't see any sites. I don't see a lot of sites linking to this page. Definitely not enought o warrant a PR6 or higher that he probably wants. The ones I do see linking are relatively not high pagerank sites. They're pages like this (internal PRless pages of blogs) and there's a LOT of foreign language type sites linking too. (as i type this I see a PR5 for his site..)

It's about where it stands as far as PR is concerned. He's where he belongs.

PR is irrelevant though, it's only one small factor of many in the ranking algorithm.

to be clear though, what terms are wetalking about? because I don't see him in MSN or Yahoo either.

Seth Finkelstein [PersonRank 10]

13 years ago #

He's already got PR 5 at newer Google servers.

Yahoo follows Google ranking pretty closely. So close that a top-ten site at Yahoo should be high on Google, not doznes of positions lower.

[can you sue Wikipedia] (wrods, not quoted) is the killer for me – #1 Yahoo, #1 MSN, more than #300 on Google. That indicates something amiss.

Ryan [PersonRank 0]

13 years ago #

With that term.. he's #1 in google if you put "" around it.

which makes me think it's something algorithmic, not political.

Also. there aren't any (maybe 2 , tops) pages that link to that article. It's weird for a site to have all it's links to it's main page and not to any of it's internal pages. Especially for an article on a site.

with google being highly link based... is it any suprise that an article that has 2 links doesn't rank well?

next...

Ryan [PersonRank 0]

13 years ago #

PS, it might have something to do with the fact that he's telling robots not to cache his websites either.

Seth Finkelstein [PersonRank 10]

13 years ago #

" read the original posts in this thread."

Seth Finkelstein [PersonRank 10]

13 years ago #

If you put quotes around the words, you're forcing the algorithm to focus on that string, and there's only a dozen or so matches (11 non-similar).
That's not much of a test.

Of course it's algorithmic. The issue is that the site's somehow being penalized in ranking, but the how and why aren't clear.

The "sue" page should inherit some ranking by being linked off the main page. But that's not happening.

Beta [PersonRank 0]

13 years ago #

As for differences between Google and MS/Yahoo, I experience the opposite: I'm quite high on Google for certain terms and barely in the top 100 with MS or Yahoo. I have no reason to believe that both Yahoo and MS are trying to punish me. I guess the differences between search engines just can be that large. There doesn't have to be a conspiracy or manual adjustment behind it. Has anyone ever done a large study comparing SERPs for the big three, finding out what kind of differences in result can be expected?

Daniel Brandt [PersonRank 3]

13 years ago #

Google has just fixed the problem. I've put the new rankings in the last column at wikipedia-watch.org/goohate.ht ... and will probably take down that page altogether in a week or so. I had no communication from Google, and noticed the dramatic change by accident.

Chris [PersonRank 0]

13 years ago #

I'm not sure that it necessarily follows that since your PageRank improved, Google manually "fixed" the problem. I don't automatically reject that conclusion. I do, however, remain open to the possibility that some other factor (possibly the links that your site has gotten based on your complaints) affected your PageRank.

I'll reconsider my position if you have any proof except the order that things happened.

Daniel Brandt [PersonRank 3]

13 years ago #

My PageRank didn't improve. It is still PR 5 for all pages except the two newest pages, wikitort.html and psamples.html, which still have PR zero. That's no problem – I expect a several-month lag on the PR that is reported by Google. I'm talking about how 13 of my 14 keyword searches jumped in rankings over the past three days. The average ranking for these searches three days ago was 233. Now it is 17. This is all recorded on the link that I gave above. There's nothing about PageRank on that link. The entire chart is about rankings for specific terms on specific engines.

Seth Finkelstein [PersonRank 10]

13 years ago #

It wasn't PageRank – it was the issue where the site was "sandboxed", which means PageRank factors are not applied to the result.

The site is now out of the sandbox – whether it was automatic, because of the links from "trusted" sites, or manual intervention, can't be determined – either is possible.

But the *effect* is evident!

Daniel Brandt [PersonRank 3]

13 years ago #

They are calling the "Sandbox" phenomenon "Trustrank" over at Webmasterworld.com. There are a couple of things different about my case as compared to the usual sandbox situation:

1) It took almost 14 months for www.wikipedia-watch.org to emerge, which is the extreme end among those who have reported on sandbox emergence.

2) Most sandboxed sites show some PageRank a lot sooner than wikipedia-watch.org did. I got my PageRank of 5 just last September, which was two months after I complained on Danny Sullivan's forum. Before I complained in July, everything was PR zero, the deep pages were not even indexed according to the "site:" command, and a search for "wikipedia watch" didn't even produce my home page. Three days after I complained, the deep pages showed up but didn't rank on anything at all, and the "wikipedia watch" search showed up at number one, but nothing else on the home page ranked well. I already had over a thousand external backlinks to the home page, according to Yahoo.

I remember reading an essay about the sandbox phenomenon about ten months ago. One of the points made in that essay is that high-visibility sites can get out more easily, presumably because Google doesn't like adverse publicity. All my sites are nonprofit, none contain any advertising, I can get a story in the newspaper if I need to, and I know where to complain so that Google will notice my complaint (as opposed to sending an email to Google, for example).

Now here is where Seth might disagree with me. I claim that it was manual intervention at the Googleplex that got me out. My sandboxing was more severe than usual because of political considerations at Google, and it was released from the sandbox because of further political considerations at Google. I was in a position to make more noise about my rankings, mainly because the evidence of discrimination against me had accumulated to the point where it was overwhelming, and would interest some journalists. Google noticed my position, and acted to preserve its own interests.

There was another clear-cut case of manual intervention by Google against me in July 2004. Google trashed my "out-of-touch executives" Googlebomb, which had been at number one for months, and was mentioned in the New York Times in June 2004. My bomb on Google went from number 1 to about 800 overnight in July 2004. It still works at Yahoo! Danny Sullivan bent over backwards to argue that Google doesn't do manual interventions like the one I was claiming for my Googlebomb. Sullivan was wrong. This was a clear case of a "hand job" at Google.

It's high time for Google lovers, and Google itself, to admit what we all know in our gut. The entire ranking structure, and everything else at Google, is basically political. There is nothing sacred any more about any mere algorithm when you're sitting on a $147 billion market capitalization that is beginning to look like a bubble that could burst at any moment. You tiptoe lightly at the Googleplex these days. If Google's stock crashes, that's the end of Web 2.0 for everyone.

What are the implications of Trustrank in terms of search engine quality? It may be true that Google has to do this to get out of the "made for AdSense mess" that they created. But I know one thing – if I'm searching for something fairly recent, I have to use Yahoo or MSN now in addition to Google.

Trustrank at Google means I can no longer trust Google to be objective with its rankings. I'm also not inclined to start up any new domains anytime soon.

Seth Finkelstein [PersonRank 10]

13 years ago #

Daniel, as you know, I indeed don't think you're right about this overall. It's not because I ignore what you say because of your anti-Google perspective – indeed, I think that perspective is a source of finding things that other people would not investigate. However, after many years of activism myself, my view to the activist who claims persecution by The Powers That Be is "You should be so dangerous". I've seen repeatedly how easy it is to be MISTAKEN about the sources of bad things. This is why I try to separate out the facts from the theory. Often there's a much simpler explanation.

And I've *been* on the receiving end of abuse by (minor) Powers That Be. Someone has really got to have made some inside enemy to even be in the running for notice. Just being a loud critic won't even register.

Look, it's entirely possible that:

a) It was a manual change
b) Matt C. or whoever doesn't want to say "Oops, bug, sorry", because that'll be posted on your site as "GOOGLE SPOOK ADMITS CHANGING RESULTS!"

I don't know the exact reason you were sandboxed for so long. But I think it far more likely that it has to do with having lots of junk on the site (e.g. IRC logs) that somehow mistakenly made it look like a scraper or spam-site to the anti-spam algorithms, than that someone at Google personally decided to suppress it.

Dan Tobias [PersonRank 6]

13 years ago #

One should note how Brandt boasts of being such a high-profile activist that he can get mainstream-media press coverage any time he wants it, but still insists that Wikipedia delete its article on him because he's a non-notable private individual.

Philipp Lenssen [PersonRank 10]

13 years ago #

It's weird, I have 20+ sites but I never saw this sandboxing in action. However, when I publish a new site I also try to get the word out by emailing blogs, submitting the site into "submit an item" sites and such. Of course the site won't instantly jump to top, if ever, because it's new and may not be worth to rank well. Maybe that works against the sandboxing, by having good "seed" sites link to one's site. But really search engines have to figure out how to rank the site, not the webmasters... if my site doesn't appear after say 3-6 months (you gotta wait first, of course) I wouldn't try to figure out how to "optimize the search engine" (SEO) but I'd try to find people who might be looking for this site, e.g. BBC Click, CollegeHumor.com, or some very specific webmasters in this topic zone who might actually *enjoy* the site. This is NOT to say that from a google-watching perspective I'm not interested if there's a hint of results odditiy (I definitely am!).

This thread is locked as it's old... but you can create a new thread in the forum. 

Forum home

Advertisement

 
Blog  |  Forum     more >> Archive | Feed | Google's blogs | About
Advertisement

 

This site unofficially covers Google™ and more with some rights reserved. Join our forum!