| This 208 message thread spans 7 pages: < < 208 ( 1 2  4 5 6 7 ) > > || |
|Why Is Wikipedia On Top in Search Results?|
So many searches now have Wikipedia at the Top.
I've noticed in many searches for business names or people over the past few months that Wikipedia is now the No. 1 or 2 result. Is Google promoting a non-profit site over others or just the huge traffic Wikipedia is getting?
|It may be a better reference source than most on the Internet, but it is not by any stretch of the definition an authoritative site. To be authoritative one must know what the qualifications are of the person doing the writing. Put simply anonymity and being authoritative are mutually exclusive concepts. |
This thread isn't about whether Wikipedia is an authoritative site that deserves praise or pooh-poohing by academic librarians and professors; it's about why Wikipedia is "on top in [Google] search results."
Again: Google is an automated search engine. It can't make an informed, human judgment about the credentials of every author or the credibility of every article. It can use PageRank, "TrustRank," on-page factors, etc. to determine the statistical likelihood of whether a given page from Wikipedia (or any other source) will supply information that will satisfy a user who searches on a keyphrase.
It's also worth noting that Google's search results aren't limited to five, ten, twenty, or even a hundred. Search on "history of rome," for example, and you'll find a Wikipedia in the #3 position (or the #5 position, if one counts indented secondary results), but you'll also find another 81 pages of results before you get a message that says "In order to show you the most relevant results, we have omitted some entries very similar to the 816 already displayed." If some users choose to go directly to the Wikipedia result (#3 or #5) and ignore all 800+ other results, that's hardly Google's fault. Google can't keep people from being brand-conscious or lazy.
Google ranks results. That is what we are discussing. My aunt Myrtle can supply information on any topic, but she shouldn't be at #3 in the Google results. Likewise a competely accurate two sentences citing the work of a Nobel prize winner should not outrank an authoritative 100 pages by that Nobel prize winner.
A PR6 Wikipedia page outranking a PR2 edu page is reasonable. Better SEO can explain why the Wikipedia will often rank higher than a niche authority site, but when the link power and PR are similar, ranking the Wikpedia above the real authorities is lame search engineering.
It's the search engine's business to rank results for their users, not just say "um, here are a pile of relevant pages including a relevant page on a big website".
Wikipedia articles are normally mediocre, above average but not exceptional. So is Google.
|Most of the wiki editors are an expert in the field. Not all of them but most of them are. |
How do you know this is true? How does anybody know this is true? How can you prove it to be true? The answers are: you don't, they can't, and you can't. The reason is that posts are anonymous. Thus it isn't an authoritative source. Yes the information is frequently pretty good, but it isn't authoritative. If you doubt this or question it then you do not understand the fundamentals of research especially where science is concerned.
Wikipedia is a good generalist site. For the sheer number of topics they cover they are pretty good, but Wikipedia is not a specialist site and at a specialist level Wikipedia frequently falls short and is nothing more than a tertiary source that simply regurgitates what specialist sites actually did the research to compile and publish. I should know because one entire section of every Wikipedia language is based on the data that I and three or four other websites spent years compiling from off line sources before Wikipedia even existed. Wikipedia editors did nothing more than go through and collect the data from my site and the other sites and then make their own clone of what we had already published. Early on editors had also copied some of my articles and claimed they were their own, which necessitated me filing DMCA take down notices and then chasing down the copies that had been copied from Wikipedia to other sites like About dot com, etc.
|There is a series of edits a subject goes through before it gets posted. Like I said, not perfect but a lot better than any other site I have seen out there for all the subjects they write about. |
|it's about why Wikipedia is "on top in [Google] search results." |
Wikipedia isn't on top because Google anointed Wikipedia. It is on top because as of its sheer quantity of pages AND because so many people link to its pages.
Oh and contrary to what some have stated the sheer quantity of pages doesn't make a site authoritative. A site can have only one or two pages and be THE authority for the topic in question. At the same time as Wikipedia shows a MASSIVE amount of content creates lots of fodder for the search engines and if some of that content is good that good content will generate a great number of inbound links that will help bolster ALL pages of the site in search results for relevant search phrases.
If this topic is about why Wikipedia is on top, the question has been answered. If web publishers want to know what they can do to reverse the trend of Wikipedia dominating SERPS the answer is stop linking to it. Collectively web publishers helped give Wikipedia its SERP power by constantly linking to it. I've been saying it for a long time and it is still true. If we don't want Wikipedia to eat our lunch in the SERPs we need to stop linking to it and those who have put links to it need to remove those links. Wake up and get smart folks, if you are a website publisher you need to fear Wikipedia.
|Wikipedia articles are normally mediocre, above average but not exceptional. So is Google. |
Indeed, sites like Wikipedia aren't about quality. They are about quantity. Or as Stalin stated "quantity has a quality all its own." Of course it was in Russian. Sites like Wikipedia are pushing the Internet towards mass market mediocrity because it is exceedingly difficult to produces massive quantities of high quality content but Google has a habit of favoring quantity over quality because quality is a highly subjective concept that automated processes can't really score.
Personally speaking I have found producing large amounts of mediocre content to be much more profitable than spending the safe effort and money producing a small amount of high quality content. I still try to produce high quality content, but I subsidize that content with mediocre content that actually generates the revenue to pay for the higher quality stuff.
[edited by: KenB at 12:59 am (utc) on Feb. 26, 2007]
How can you prove that anyone who posts to the internet is a subject matter expert? Anyone can publish online.
At least wiki has some very good editors who are interested in the subjects they post in.
Wiki also lists out a lot of references where you can further research a topic as well. Most websites do not do that!
Also you make a statement that websites should fear wiki? Why would we? They are just an informational site just like a lot of other sites out there. In other words, they are an online encyclopedia, thats it, no more no less.
If you want to outrank wiki, make sure you have more content than they do. There are many sites out there that outrank wiki. If your not, add more content and you will.
For my my main keywords, wikipedia doesn't rank terribly high (I don't know where it ranks actually, but not top 20). However, my site is linked to from the relevant wikipedia entry (yes, I added it, but I have a genuinely useful site on widgets) and I do get a lot of click through traffic from wikipedia. So, it seems to me it gets lots of traffic, but passes a lot of it out as well. Just my anecdote.
You can look at the source of the material (e.g. type of site it is posted on and who the by lines are. These factors can help one weigh what the motives are of the individual. For example on my site not only do I require bibliographies for articles I publish, but I also publish complete biographies of the writers.
|How can you prove that anyone who posts to the internet is a subject matter expert? Anyone can publish online. |
This doesn't make my site an authoritative site, but it does provide my readers with enough information that they can validate the articles for themselves and discern what the motives of the writer might be. There is no such thing as a non-biased writer, everyone has a bias and if they are anonymous we have no way of discerning their bias.
|At least wiki has some very good editors who are interested in the subjects they post in. |
Prove it. Who are they? What is their background? What are their motives? For every good editor how many crappy editors are there? How many have hidden agendas? How many think they know a subject but don't have a clue? None of this can be answered because the editors are anonymous
On this part I will say that most websites are crap (can I say that?). Really the Internet would be a much better resource if everyone held to higher standards. At the same time just because a Wikipedia article has a bunch of references doesn't mean that the article actually used said references. This is no different than high school reports where the student regurgitated information from a couple of sources and then padded the bibliography with "random" unused titles pulled from the card catalog.
|Wiki also lists out a lot of references where you can further research a topic as well. Most websites do not do that! |
|Also you make a statement that websites should fear wiki? Why would we? They are just an informational site just like a lot of other sites out there. In other words, they are an online encyclopedia, thats it, no more no less. |
They should fear Wikipedia because it has become the 800 lb gorilla with so many pages (some good, some mediocre and some just stubs) and so many links on vastly diverse subjects that its pages now have the ability to gain high rankings in search results regardless of the merit, quality or actual external back links to said wiki article. Simply put a mediocre article or even a stub on Wikipedia with no external back links itself can out rank and bury quality articles on other sites. Put another way, Wikipedia is bleeding traffic off of the authoritative sites that Wikipedia's editors used to produced their "own writings". All too often those Wikipedia writings are not based on original research but simply a rewriting of the research done by others.
The real threat Wikipedia poses to the Internet is that it will bleed traffic off of commercial sites that actually depend upon readers to generate the revenue they need to pay writers to do original research. Sure this might not matter for some pop culture subject. In some niches, however, that depend upon real research and reporting (including off line research) with the disclosure of sources and motives, the bleeding of traffic will reduce the quantity of quality original content that can actually be produced. Simply put Wikipedia will push out quality content and replace it mediocre that is written by unknown authors with unknown motives.
|If you want to outrank wiki, make sure you have more content than they do. There are many sites out there that outrank wiki. If your not, add more content and you will. |
You just don't get it do you? One can either focus on quantity or quality, but one can not produce mountains of content and maintain a high quality standard (especially not Wikipedia) unless they have a bottomless budget. Furthermore one can not be an expert on all subjects. The truly most authoritative sites on any subject will be very narrow in scope. They aren't going to cover every topic under the sun. They are going to focus on one thing and do that one thing very well.
These authoritative niche sites are now getting buried by mediocre articles on Wikipedia without any individual merit because Wikipedia has garnered hundreds of thousands of back links for inconsequential trivia and pop culture de jour.
I have been saying Wikipedia was a threat to web publishers for a few years now (ever since I first started to discover my articles being republished on Wikipedia verbatim) and now my fears being proven true.
[edited by: KenB at 5:30 am (utc) on Feb. 26, 2007]
Over 2 million backlinks, many backlinks to specific articles...
Plain and simple...
Let's move on, please!
None of this is ever a mystery.
It's not a mystery, but don't be simplistic.
Wikipedia articles often rank higher than better quality content with better quality backlinks. This is the problem. Google foolishly values the generic of the Wikipedia over webpages that have many more on topic links, more higher quality links, deeper content, and far more topical authority.
Google trusts mediocrity, not topical authority. This leads to not just the ranking of weak-ish Wikipedia pages, and not just the enormous mass of hacked edu redirect spam, but the overall mediocre quality of the non-spam results.
What irks me is that Wiki has often contained content that is paraphrased, with pictures, from a specialty site of mine. In other words, they have rehashed my content but because of wiki's PR and their new nofollow tags, my pages are sliding. I can see first hand how wiki infests the top rankings, slowly at first but like a infection it overcomes the host (oringal page with the information) to dominate the rankings.
Wiki often publish falsehoods that hurt real people. Take the nonsense they have put up about one group of islands in Asia which has been desparately seeking independence (and according to UN observers, legally and successfully declared it 30 years ago) from a invading pariah state. The entry was clearly made by a proponent of this state and Wiki refuses to correct falsehoods which have been shown to them using clearly documented facts.
It's a wiki - anyone can publish and anyone can make a change. No need to get upset about a "they", just do it.
> if Googlebot were smart, it would take "nofollow" with a grain of salt when crawling Wikipedia, blogs, etc. : i.e., by following the link anyway if the link led to a page at a "trusted site."
I've always wondered whether googlebot does that. The recent rise of Wikipedia since they started nofollowing suggests that they don't.
The nofollow decision by Wikipedia was a poor one though, even if they do have a major issue with spam.
Why do you say it was a poor decision by Wikipedia?
Sure it is bad for us but they win all the way.
Tapolyai says: "If you find an article that seems to have a slant, or lacks attribution, or is an original work, put it in the discussion. It will be struck."
I know of several examples where bias is complained about on the Talk page, but the article has not been corrected for months, and attempts to correct it are rolled back by the "owner" of the article. As someone else said, those who are more persistent can some times win out over those who know more.
DumpedbyG say: "Solution = Build your own Google custom search and exclude Wikipedia."
One of my problems with WP concerns their several biased or incomplete entries misleading people about politicians. I could certainly build my own custom search, but what about the thousands or tens of thousands of people who've been given the false impression of a politician because of a biased or incomplete entry?
europeforvisitors says: "Fortunately, Wikipedia articles typically include citations to serious research destinations that are acceptable. :-)"
Not if the "owner(s)" of the article don't agree with the viewpoints expressed in the linked site and refer to it as "linkspam". I have one of my sites in mind, one relating to a specific event. Despite having about 100 times more information than the corresponding WP entry, it was repeatedly deleted with one pretext after another by one of those "owners".
As for nofollow, not all the external links at WP have that tag. For instance, their Interwiki links don't have that, and there's at least one very well-known commercial site dealing with movies on that list. The top video sharing site was briefly on the list but was deleted out of copyright-linking issues. A far-left site has its own macro that isn't nofollow'ed. And, apparently one of WP's commercial sister sites isn't nofollowed, as are links to their foundation.
If you want to see WP political bias in action, choose a minor but controversial politician (choosing a liberal is by far your best bet) and see what his opponents say about him. Then, see whether that factual "dirt" is in his WP entry, or whether it's been sanitized for searchers' protection.
I'm willing to bet Google will end up buying Wiki!
|If you want to see WP political bias in action, choose a minor but controversial politician (choosing a liberal is by far your best bet) and see what his opponents say about him. Then, see whether that factual "dirt" is in his WP entry, or whether it's been sanitized for searchers' protection. |
Anybody can edit WP, why do you not just go and fix the political bias?
|I could certainly build my own custom search, but what about the thousands or tens of thousands of people who've been given the false impression of a politician because of a biased or incomplete entry? |
Why should this be our problem? If the tens of thousands of people want to read the biased articles let them.
I think your are more concerned that they are not reading your "unbiased" material.
[edited by: DumpedbyG at 11:49 am (utc) on Feb. 26, 2007]
|This is why I don't like wiki, the "content" is VERY unreliable |
You could say that about 95% of the content on the internet. Wikipedia is no better or worse.
If you can't stand the heat...
They are upset because their "informational" adsense pages have been replaced by wiki and they took a hit in the pocketbook.
[edited by: trinorthlighting at 1:13 pm (utc) on Feb. 26, 2007]
The one thing that google does is it ranks the answers.com lower. What about Yahoo? It promotes this answers.com which of the same content as Wiki, the difference is the ads. How much money does Wiki getting from Answers.com? Answers.com is #278 in Alexa?
Wikipedia is a great source for information as long as you are in that frame of mind.
Unfortunately, Google can only predict so much with the query you input. While they will provide informative results based on that interpretation, sometimes that interpretation is simply off.
While that's understandable -- Wikipedia to Google is like candy to a child now... Every possible thing they're looking for -- they have. Backlinks, PageRank, Users, etc. Based on an algorithm, Wiki has it all.
It'd be nice for searchers to customize search results, and I suspect that is a year or so off following the deployment and mass usage of the AJAX APIs.
Either way, currently -- it's bad for Google users.
> Why do you say it was a poor decision by Wikipedia?
Wikipedia is largely built out of knowledge scanned from other websites, usually people put references to where they discovered certain facts and those links should be counted by Google and other search engines when deciding a page's popularity. That is the basis to PageRank. By applying no-follow extensively Wikipedia, being a very important website, breaks Google's algorithm to some extent.
|They are upset because their "informational" adsense pages have been replaced by wiki and they took a hit in the pocketbook. |
Isn't that the case with 75% of the anti-Google posts around here?
|That is the basis to PageRank. By applying no-follow extensively Wikipedia, being a very important website, breaks Google's algorithm to some extent. |
Google decided to implement the no-follow rule, so Google broke their own algorithm not Wikipedia.
Yes, you can add this to the eBay/epinions/dealtime/kelkoo.uk crap that dominates the top 10 in many searches.
With each of these showing 2 results, sometimes you are lucky to see 2 real websites in the top 10.
I wouldn't mind it as much if these things were actually relevant, but they usually aren't. Seems 90% of the time, they don't have any reviews/info/etc on whatever the product is - nothing but a page full of ads.
Google COULD have good results if they would simply downgrade this kind of stuff...
Sites like wiki/ebay/amazon are very popular websites that people like to see in the results. I think they are ok. When people do not want to see these sites, they have the choice of not clicking on them. Google serves up a variety of sites in the top 10. No need to downgrade them, if people do not like the results they can go to page 2 or 3
You left out the important point that the Wiki "content" that is replacing these informational websites in many cases was gleaned and rehashed from the same informational sites Wikipedia is replacing. Simply put Wikipedia didn't get to its position by using original research. All too often articles on Wikipedia are a mediocre and non-original (in the strictest sense) regurgitation of the content Wikipedia is replacing in the SERPs.
|They are upset because their "informational" adsense pages have been replaced by wiki and they took a hit in the pocketbook. |
To me Wikipedia implementing the nofollow instruction on links smells of SEO. In my opinion, this has nothing to do with reducing link spam that has been cluttering up Wikipedia. I'm not saying whether it is a good or bad thing for Wikipedia to do (it is their purgative). I'm simply saying what I believe to be the real motivation behind it.
|Wikipedia is largely built out of knowledge scanned from other websites, usually people put references to where they discovered certain facts and those links should be counted by Google and other search engines when deciding a page's popularity. That is the basis to PageRank. By applying no-follow extensively Wikipedia, being a very important website, breaks Google's algorithm to some extent. |
To me, the thing that has always most scared me about Wikipedia is what is the real motivation behind its creation? I'm sorry, but I have an inherent distrust of the motives. They always say follow the money.
What annoys me with wikipedia is that many of the articles are very thinly disguised rewrites of the sources that they link to, and of course the Wikipedia rewrite scores higher than the original source.
Because they're rephrased and factual, the people who put it on wikipedia claim they can get away with it, and they probably can legally, but it's still clear who did the real legwork on structuring and editing the article together.
For me, discussing Wikipedia is the ultimate in discussion about authority: what is authority? Also, what is an "encyclopedia?" And should one ever quote from one?
When *I* got to high school, I was not allowed to use them as a source. So why is Google using it? Perhaps because *most* only read at an 8th grade level? I know that my niece in a highly respected school has been told, just as with any encyclopedia, she *cannot* use Wikipedia as a *source* for her writing. Amen, I say.
BUT, does that mean Google should not give it authority? Thus, back to, what it "authority?"
One thing that's not clear to me is if Wikipedia has achieved this status as a natural result of Google's algorithm, or if Google has manually given them some kind of "bonus" in the SERPS. One thing is for sure - they've been showing up a LOT more often lately.
I don't find it that bad, really. For the searches I do, Wikipedia normally has spam-free and accurate information. I laughed at BaseVinyl's post, but I can honestly say that I *do* often search Wikipedia first, simply because (again, for my types of searches) it usually has what I want to find.
I got in on the Internet toward the end of the days of "use-this-engine-for-this-search-type." It appears that those days may be coming back in a slightly different way. I think users are going to gradually start searching certain sites (as opposed to using certain engines) when they're looking for particular types of information. Firefox and search add-ins (and search keywords) make this a lot easier.
| This 208 message thread spans 7 pages: < < 208 ( 1 2  4 5 6 7 ) > > |