Very much worth watching this.

Rgds
Richard

How would you measure the success of Wiki? Best proxy in my view would be usage. Of course others will disagree.

yes, google is providing upwards of 50% of wiki’s traffic. and yes, wiki is a v prominent outward link from google (1.87% of downstream traffic). regardless, i still fail to see the problem here, for reasons offered above. but also i just don’t see the logic of what you’re suggesting. yes, come the wiki search thingey (which sounds more like a directory than a true search engine – and we all know the problems inherent in directories) google could throttle back wiki’s traffic. but how would that hurt wiki? surely doing the opposite – driving up wiki’s bandwidth costs – would hurt more?

funnily, i find the wiki-damaging-google argument ( http://www.theregister.co.uk/2007/03/06/wikipedia_crisis/ ) more persuasive. though i’d be as unwilling to bet on it coming true as i’d bet on your theory.

btw – have you noticed the changes in amazon’s link structure? wiki’s place as the #1 serp could be under threat, especially if the same link structure crosses from amazon to imdb.

And I’ve read that:

Wikipedia received 54% of its traffic from Google

The above is based on UK traffic. Of course none of the above can be judged to be 100% accurate, but I’m happy to believe that some of Hitwise’s data is reliable.

If you would like me to dig up any other details just leave another ‘sheesh’ comment

hang on a moment yesterday you *knew* this figure. it justified your post’s unexplained leap of logic. now you’re just wondering what it is? sheesh …

I’m not attacking Wikipedia. I think Wikipedia is a fantastic resource and use it regularly. This is not a Wikipedia-bashing exercise. What I am saying is that Wikipedia may be over reliant on Google for their traffic. My apologies if I am sending mixed messages on this or my comments lack clarity.

Hi Andrew. The points you make are very valid, and my English-mother-tongue-blinkered brain now acknowledges the multi-lingual element of Wiki. I do still believe that it is not healthy for any site to rely on one source of traffic to any extent.

Google in itself is not the one masterminding influence on the Internet.

I have to say that as an aggregator of all the links on the Internet Google is very much a masterminding influence. What Google displays on it’s pages affects what people link to in the first place. You do realise that increaslingly people view Google as the gateway to the Internet. Just look at how many people type URLs into Google’s search.

If Google went away tomorrow, the next thing that people use to search or find will recognize the linkages.

Perhaps, perhaps not. All algorithms are different. If the other SEs gave the same weighting as Google to the factors now affecting Wiki’s rank we would have similar search results in Y! and live! Of course there are a number of other SEs, but none has been adopted in any significant way.

Hi Halfdeck. Again, this isn’t the rant of an anti-Wiki SEO. I really want to get across what I feel is the possibly unhealthy relationship between Google and Wikipedia. Perhaps all those links do justify the rankings, but I wonder just how much of Wiki’s traffic comes from Google alone? In a nutshell – would you trust Google if you were Wikipedia?

1. Do you realize that Wikipedia is a multilingual project of enormous scale? That there are 13 language editions with over 100,000 articles each, and for many of those languages, Google is not the #1 search engine? The multitude of ways that Wikipedia is being linked to across languages, cultures and geography means that we Anglo-oriented fellows don’t even begin the fathom the impact of Wikipedia’s inward-linking ecosystem.

2. Google’s PageRank is a reflection of what is being linked on the wild Internet. Blogs, articles, news organizations, etc. are linking to Wikipedia articles which is why it’s showing up in Google. Google is in effect a third party simply cataloging this relationship and providing it as search. Google in itself is not the one masterminding influence on the Internet. If Google went away tomorrow, the next thing that people use to search or find will recognize the linkages. Check out Clusty, A9 or anything else out there and see the results.

I see two basic reasons why SEOs aren’t fond of Wikipedia:

1) They are competitors. No matter what niche you’re in, Wikipedia is likely to show up on page one, pushing someone off the front page.

2) Due to their new nofollow policy, SEOs can no longer leech off Wikipedia to improve their search ranking.

In other words, Wikipedia takes an arm and a leg from SEOs and gives nothing back.

But for the average Google user, Wikipedia is often an invaluable resource, much more so compared to some of the junk out there people try to pass off as legit websites.

–

I also agree with David Gerard. If Google flicked a switch tonight and Wikipedia pages slipped to page 10, Wikipedia wouldn’t blink an eye.

“And I know for sure that you can find Wikipedia pages that rank well without even one external link pointing at them.”

Internal links on Wikipedia are pretty darn strong, considering that the site is a PageRank black hole and that millions of links point to the site.

Well as much as I think Wikipedia is a great project, it’s value can only be measured through its usefulness to users. Given that a large number of its users arrive via Google, any reduction in that traffic is going to have a knock-on effect on the success of Wikipedia.

Our ad revenue would drop to zero … from zero.

Obviously I’m well aware that Wiki doesn’t generate revenue from the site. But how can you measure the success of Wiki? I presume by how many people use it – that figure is very heavily dependant on Google SERP referrals.

After all, a resource is only really useful if people actually use it.

But the point I’m really trying to make is that the relationship between Google and Wiki strikes me as a wee bit unnatural.

If Google cut off all links to Wikipedia tomorrow … the project would continue writing an encyclopedia, and barely noticing. Except for the servers appearing to run a bit better without the added load. Our ad revenue would drop to zero … from zero.

Google links Wikipedia because they like its effects on their search results’ usefulness. Which is nice that Wikipedia is useful to people, which gives all this a point, but that’s about it.

What was your actual point again?

are you contending that the vast majority of wiki’s traffic is coming from google alone?

Yes, absolutely. I’ll try to dig up some stats on that (I’m sure I’ve seen them somewhere).

WRT the front door – that is a very persuasive argument which I’m sure is very credible, but I don’t believe that the majority of Google traffic through to Wiki is from ‘front door’ users.

and, as has been pointed out, many of use google as wiki’s frontdoor / search engine. so there you have to ask yourself, how much of the google traffic is genuine search and how much is just down to google having better ease of use than wiki’s own search?

Yep Wikipedia scores very well because of many inward links, but I have a niggling feeling that this is overstated in the ranks Wiki gets. No evidence, just gut. And I know for sure that you can find Wikipedia pages that rank well without even one external link pointing at them.

@Ken – Hi also. Sorry that was the point I was making. No other traffic source could take Google’s place. I don’t think Google would blank them, but they could twist a knob somewhere and very subtly affect the rank of many Wikipedia pages.

Bear in mind that Google is a very slick money-making machine (with a search engine). Wikipedia is working on its own search engine, which could draw traffic away from Google. Things could get interesting then.

I missed the logic leap that went from Wiki being a prominent outward link from Google, to Wiki being overly reliant on Google for traffic.

Surely one of the main reasons that Wiki scores so well is down to the number of inward pointing links? Most every blog somewhere on it has a link to Wiki. Wiki has one of the most distributed set of inward pointing links on the web that you can imagine.