searchengineland.com » Durova

The SEO Value Of Featured Pictures At Wikipedia

Durova — Tue, 09 Oct 2007 12:49:22 +0000

My last article about Wikipedia—An Untapped SEO Opportunity: Image Link Love From Wikipedia—prompted an Orwellian mantra: instead of “four legs good, two legs bad,” people have started saying “images good, text bad” and wondering why the difference.

The short answer is the Wikimedia Foundation created a white hat opportunity for SEO to generate outgoing links via image uploads. The strategy hinges on a copyleft licensing option called CC-by-sa, and the decision isn’t accidental. They’re glad to trade some link love in return for quality content. Or, as Wikimedia communications committee member David Gerard puts it, “Releasing control over your stuff is hard. But consider the advantage of having a picture that isn’t a blurry, red-eyed piece of suck.” I chuckled at those words until I viewed the link David supplied as an example. Really, Emmy-winning actor Richard Schiff ought to have a portrait at his biography that looks better than the one on his driver’s license.

Guys, if your wife or girlfriend is the jealous type it may be wise to wait until she leaves the room before opening this next link. One celebrity who gets the idea is Michele Merkin, who hosted The Next Best Thing on ABC last spring.

There’s an important detail on that page if you can pry your eyes away from Michele for a moment. Have a look at that gold star in the far upper right corner: Wikipedia calls this a featured picture. Yes, the lady has brains. Michele is the first celebrity to gain featured picture status at Wikipedia by releasing a professional portrait under copyleft license. Readers of my last column won’t be surprised to learn this has earned her publicity at some high level Wikipedia articles.

By releasing high quality portraiture and gaining featured picture status, Michele’s image will probably be durable at all three of those pages. Featured pictures almost always are the winners in editorial debates about how to illustrate an article. That’s great… erm… exposure for her.

Another benefit that goes to featured pictures is a day on Wikipedia’s main page. I’ll crunch a few numbers to demonstrate how valuable that is. Recently I was on Alexa tooling around with their search function to see what a monster Wikipedia had become. You’ll either love these stats or hate them.

Wikipedia: traffic rank 9
Nytimes.com: traffic rank 216
Wallstreetjournal.com: traffic rank 1,098
Britannica.com: 4,097

If you combined all three of those venerable publications they’d bring about one tenth the traffic at Wikipedia, which gets 40 million unique visitors a month. 52% of the overall traffic goes to the English language edition where the most popular destination is the main page. According to Cary Bass of the Wikimedia Foundation, that page gets 3 million total hits a day.

All of this points to one simple conclusion: in late 2007 a scroll-down spot on Wikipedia’s main page is considerably more valuable than below the fold placement on the front page of The New York Times. If your goal is publicity then Wikipedia’s main page is a great place to be. Hardly anyone in the business community has tapped this opportunity yet. Most of Wikipedia’s pictures of the day are amateur creations or historic public domain material. This doesn’t need to be the case: businesses and entertainers commission large amounts of professional photography, and Wikipedians appreciate quality content.

So, here’s how to take advantage of the featured picture program. To get started, brush up on the important licensing information from my last article. Then browse Wikipedia’s featured picture category and read up on Wikipedia’s featured picture criteria. If you think your material is good enough then make sure you have a high resolution image of at least 1000 pixels in height or width.

Next, head to Wikimedia Commons and upload your images. Write up a self-nomination at the candidate page. Volunteer editors review each nomination and reach a consensus decision. As an example, here’s the discussion that promoted Michele Merkin’s portrait. After an image gets accepted you can write up some accompanying text and ask for a turn at picture of the day.

Wikimedia Commons also runs its own parallel featured picture and picture of the day programs. You can try for these also if you like:

Why are images so welcome while SEO pros walk on eggshells at the text side of Wikipedia? There are more ways to run afoul of site standards with text, especially since the style of professional press releases and advertising copyrighting is entirely inappropriate for an encyclopedia. Many desirable images are proprietary, which means business people are often uniquely positioned to provide encyclopedic content that way. David Gerard estimates that 10% of the image uploads at Wikimedia Commons violate copyright. The site’s regular volunteers undertake the Herculean labor of identifying and deleting that material.

In the long run, more freely licensed material will reduce the inflow of copyvios and that will make a lot of volunteers happier. The chief obstacle, as SEO professional and Wikipedian Jonathan Hochman notes, is persuading a client to loosen some control over copyright. “If you have a performer this could be a great strategy, or if you promote tourism to an area. The much bigger opportunity is to seed the Net with good quality reproducible images.” Businesses often own several nearly identical versions from the same photo shoot but use only one of them. A CC-by-sa licensed upload of one of the alternates may earn featured status on Wikipedia, while the client retains full copyright on the primary version.

Durova is the pen name for Lise Broer, a Wikipedia administrator who confronts some of the site’s most disruptive editors. After graduating Columbia College, Lise attended film school and also served in the US Navy. The Let’s Get Social column appears Tuesdays at Search Engine Land.

An Untapped SEO Opportunity: Image Link Love From Wikipedia

Durova — Tue, 11 Sep 2007 12:37:23 +0000

Suppose I told you about a way to create viral links on Wikipedia without raising objections from the site’s volunteers. Would I have your attention?

The key is images.

Wikipedia suffers from a shortage of images for use as illustrations and the SEO profession has yet to recognize the opportunity this presents. A modest investment of time learning site standards and licensure options can yield substantial benefits in website traffic, brand recognition, and customer loyalty.

According to Cary Bass of the Wikimedia Foundation, “The Wikimedia Foundation appreciates good quality, freely licensed images, and credit where credit is due is never an issue. Many of our images fall under an attribution license, which often takes the form of a web link back to the releaser’s web page.” So one mutually beneficial SEO strategy is to locate appropriate articles that lack images and upload a targeted set of images for them. As long as these uploads benefit the encyclopedia and the approach doesn’t come on too strong, site volunteers welcome the material.

To see what this means, let’s start by examining a missed opportunity. One of the world’s leading champagne producers is the Taittinger label. They also have some of the most interesting cellars in the business: part twelfth century monastery, part fourth century Roman chalk quarry. I visited the place when I was in France and enjoyed it for the history even though I’m not much for champagne. A good set of copyleft uploads from this firm could suit a variety of winemaking articles and probably appear elsewhere on topics as diverse as French history and Gallo-Roman culture. Unfortunately for both Taittinger and Wikipedia, the lone image on the article about this vintner is an amateur snapshot of the parking lot.

Taittinger is a leading champagne producer that operates in historic structures from Medieval and Roman times, but due to licensing issues Wikipedia’s article depicts only the parking lot.

A glance at Taittinger’s rival Moet et Chandon gives a better impression. The article about this firm, which is the maker of Dom Perignon champagne, contains four respectable images. That establishes an adequate brand presence on the article page, yet a close look shows how this company still misses out on several potential opportunities.

Moet et Chandon, makers of Dom Perignon champagne: the firm could get more traffic and brand recognition with a coordinated strategy.

Of the four Moet et Chandon article images, only one generates an outgoing link to the firm’s web site and that image is under full copyright. Fair use rationales don’t allow for reproduction at related pages where an astute marketer wants to appear and full copyright prevents an image from being housed at Wikimedia Commons. We’ll learn more about Commons later and how it facilitates viral links, but for now it’s enough to note that this company is in six different language editions of Wikipedia but only two of those six languages have an image-based link to the company’s site. Among the other three images on the page, two are under GDFL license and generate no outgoing links to the firm. The third has a CC-by-SA 2.0 license with an outgoing link to the photographer’s web site. Although the brand exposure probably still benefits the company, this image may be a legal gray area since Moet et Chandon owns an underlying trademark.

The bottom line for our discussion is that businesses can generate synergies by selectively re-licensing some images for use at Wikipedia and its sister projects. The first people to understand how to do this effectively are going to gain exposure at high level articles. An innovative SEO approach for either Taittinger or Moet et Chandon, for example, could position an outgoing link by uploading an image for use at Wikipedia’s article about the Champagne wine region or the history of wine. According to Larry Pieniazek, who volunteers for both Wikimedia Commons and Wikipedia, preferential treatment goes to the first appropriate image that gets suggested for an article. Wikipedia articles aren’t galleries so later candidates need to be substantially better in quality to take the place of an existing image.

Opportunity for high level placements exist across the topic of winemaking. The article American wine has no image at all. That’s right: if you represent an American vintner, your client’s product could become Wikipedia’s lead image of American wine. Other language editions would duplicate that image along with its outgoing link, and other downstream users would continue to spread the brand presence and linkage. A range of other articles lack images:

Regardless of your client’s specialty, a lot of low hanging fruit probably waits to be picked.

Hosting your images

The ideal location for image uploads isn’t Wikipedia itself but a sister project called Wikimedia Commons. Commons hosting makes it easy for an editor from the French Wikipedia or the Japanese Wikipedia to reuse an image they find at the English language Wikipedia. Of the three Commons regulars I consulted—Bass, Pieniazek and Brianna Laugher—all advised against massive uploads: Wikimedia Commons isn’t a free web host. Yet they all reacted with enthusiasm to the idea of twenty targeted uploads for specific articles that lacked images. A vintner could supply photos of grapevines, harvesting, fermentation, tasting, and finished product.

Sometimes, as Laugher explained, multiple images of the same subject are welcome if they emphasize different aspects of a topic. Laugher recommended the Mayflower search engine to scan Commons content and suggested the Commons help desk as a good place to seek advice.

Pieniazek added suggestions for categorization. “The best way to categorize them is to do it by what seems to come naturally… grapes under grapes, the process stuff under winemaking, and wines under wines. The category system is pretty good even though it could be better. Announcing on the Village Pump and asking for help would be a good thing in this case too.”

Appropriate licensure is important. Wikimedia Commons volunteers agree that CC-by-sa 2.5 or 3.0 is an attractive option for SEO purposes because it can stipulate that downstream users replicate an outgoing link to the client’s site as attribution. Learn the details before taking the plunge: some form of dual licensure might be the best option for a client’s needs and some of the other Creative Commons licenses are not accepted at Wikimedia projects. Credit normally appears on the image page rather than in an article caption.

A checklist for Wikipedia image SEO

1. Do a survey to identify specific articles that lack quality images. Your aim should be to provide unique content where it did not exist before. Create a list of images that would each enhance at least one article.

2. Arrange the appropriate licensing and announce it on your client’s web site.

3. Upload images to Wikimedia Commons. Include a statement about licensing status and an outgoing link as verification. Categorize the images.

4. Over at Wikipedia, notify the appropriate WikiProject with a list that correlates available images to suggested articles and request assistance from volunteers. This indirect approach is diplomatic for conflict of interest situations.

5. Be low key and tasteful. Wikipedia is an encyclopedia.

My next column will explain more ways to benefit from image uploads, including tips for landing a spot at some of the best real estate on the Internet: Wikipedia’s home page.

The Right Way To Fix Inaccurate Wikipedia Articles

Durova — Tue, 07 Aug 2007 12:51:03 +0000

Suppose your company, boss or political candidate discovers that their Wikipedia article is wrong, or has subtle inaccuracies that nonetheless paint them in an unfavorable light? Most people unfamiliar with how Wikipedia works consider only two solutions: edit the article or sit on their hands. Unfortunately, neither approach typically results in the optimal outcome: a factually accurate profile containing trustworthy information.

Search marketers and reputation management professionals should know that there are legitimate ways to correct errors in Wikipedia. Knowing the right way to fix things is even more important now that Wikipedia results frequently appear in the top listings of Google search results. The good news is that Wikipedia actually offers a broad range of options for correcting inaccurate or negative entries, and even better, all are easy to use and take little time to implement.

My last column looked at examples of inappropriate editing originating from a United States Congress IP address—meaning one politician’s staff was attempting to use Wikipedia for less than ethical purposes. This time we’ll confront the opposite problem: an anonymous vandal inserted false information to the biography of United States Congressman Steve LaTourette of Ohio. For four months, Congressman LaTourette’s staffers were aware of the falsehoods but did nothing to fix them because, as spokeswoman Deborah Setliff told the Plain Dealer of Cleveland, they feared a PR backlash if they edited the page.

The most serious problem occurred in the second paragraph. According to the Plain Dealer story:

“LaTourette’s anonymously authored biography on one of the world’s most visited Web sites claims he once disrupted a law school assembly honoring England’s Prince of Wales.”

The exact text as it appeared in Wikipedia was:

“A graduate of the University of Michigan, LaTourette studied law at the Cleveland-Marshall College of Law and had the dubious distinction there of disrupting a school assembly honoring Prince Charles, the Prince of Wales. LaTourette was roughly removed by the Secret Service.”

The really damaging aspect of that allegation is how it bears a tangential resemblance to the truth. There actually had been a student disturbance when Prince Charles visited that law school. LaTourette was enrolled at the time but had nothing to do with the incident.

Wikipedia and its volunteers do care about edit vandalism and the Biographies of Living Persons policy makes this problem a special priority. LaTourette’s staff could have e-mailed the Wikimedia Foundation, either directly or via the Open Ticket Request System (OTRS) that creates a tracking number for each query.

As a Wikipedia administrator I see the opportunity to go deeper than OTRS and fix the underlying problem: this article obviously wasn’t being watchlisted. Watchlists alert active editors of changes to particular pages. These are among the most powerful tools for combating vandalism. To solicit more volunteer watchlisting, LaTourette’s staff could have contacted two projects that are interested in the article: WikiProject Biography and WikiProject U.S. Congress. Most article talk pages contain links to one or more WikiProjects. A good general contact point is Wikipedia’s Counter-Vandalism Unit. Inappropriate edits usually vanish within minutes when enough editors watch a page. Best of all, the site’s volunteers will solve future problems while you sleep.

Wikipedia also maintains noticeboards to address specific issues. Here’s a short list that every search marketer or reputation management professional should keep for reference.

Requests for Page Protection: if a series of controversial edits hit an article within a short time, a week or two of edit protection usually resolves the problem
Administrator Intervention Against Vandalism: administrators prevent continued disruption by blocking persistent vandals
Biographies of Living Persons Noticeboard: an informal alternative to OTRS for biography articles
Administrators’ Noticeboard/Incidents: addresses urgent problems that don’t fall within the scope of other noticeboards

Site administrators insist on reports that include page diffs like the one displayed above for the vandalizing edit. These are accessible through the tab at the top of each article. Here’s the history of the Steve LaTourette article.

Each date-stamped line provides a (last) option at the second column from left. Selecting that leads to a visual display of the difference between that page version and the previous one. That, in Wikipedia jargon, is the diff. It shows exactly what happened, which account or IP performed the edit, and when the change occurred. Cut and paste the relevant diff URL whenever you need to present evidence. Standard wikimarkup is to enclose URLs in single brackets.

Now here’s where this knowledge becomes especially valuable: a little wikisleuthing sometimes turns up other interesting information that a reputation management professional can put to creative use. From the diff of the vandalizing edit I get a full list of this IP address’s contributions.

That shows a pattern of gossipy edits to biographies, mostly of Ohio politicians. Some Wikipedia vandals exhibit a pattern of ideological or profit-motivated edits. If I had noticed this IP during its spree of March 6 and March 7 I would have blocked it from editing for a while. Any editor can issue warnings for clear policy violations. A word of caution: no matter what your opinion about a user’s conduct, keep the legal angle offsite. Wikipedia doesn’t mind if you actually take someone to court, but threats of a suit have a stifling effect on discussion and could end your site editing privileges. Other strategies may yield swifter and more satisfying resolutions.

Congresswoman Stephanie Herseth of South Dakota got an unexpected boost to her reelection campaign last year after an anonymous vandal attacked her Wikipedia biography. Several strange claims entered the article including a baseless charge that she was pregnant by a nonexistent staffer. It’s uncertain whether the opposing campaign coordinated the vandalism, but shortly afterward its campaign manager sent an e-mail to several of the state’s bloggers that cited the vandalized Wikipedia biography and added an accusation that Herseth was a “home-wrecker.” Rather than damaging Herseth’s reputation, the tactic backfired on challenger Bruce Whalen to such an extent that the Rapid City Journal editorial board called for a public apology from the Whalen campaign. Herseth won the election.

Durova is a Wikipedia administrator who confronts some of the site’s most disruptive editors. She uses a pen name to avoid harassment in real life. After graduating Columbia College, Durova attended film school and also served in the US Navy.The Let’s Get Social column appears Tuesdays at Search Engine Land.

SEO Tips & Tactics From A Wikipedia Insider

Durova — Tue, 17 Jul 2007 15:35:50 +0000

I am a Wikipedia administrator, and I specialize in complex investigations. When Jonathan Hochman suggested I write an article for Search Engine Land, he mentioned that this publication and its readers regard Wikipedia as a search engine. It probably comes as no surprise that my spine stiffens at that concept, but media professionals and Wikipedia volunteers seldom understand each other. So I’ll illustrate my perspective with an example: let’s have a look at some politicians.

I cite this instance because it was not one of the cases I handled and it has already been in the news. On January 27, 2006 the Lowell Sun ran a story about Wikipedia’s biography of Representative Marty Meehan, reporting that an IP address which originated from the United States House of Representatives congressional offices had erased the congressman’s broken term limit pledge from the article. Here’s the edit, which Wikipedia jargon calls a "page diff".

The story became national news when people uncovered other congressional attempts to spin Wikipedia biographies of sitting legislators. That attention led to a fresh pledge by Rep. Meehan that his staff would stop editing Wikipedia. Several of his embarrassed colleagues issued similar assurances.

Rep. Meehan has since become Chancellor Meehan of the University of Massachusetts, but that IP address continues to be shared by various congressional offices. So we’ll see whether those pledges have stood the test of time. The block history raises my eyebrow.

A few clues and clicks from there I reach the biography of New York Rep. Carolyn McCarthy and locate an edit by that IP address dated 19 June 2007 that deleted a well-referenced but unflattering section describing an MSNBC interview that had exposed her ignorance about an assault weapons ban she had proposed. Other dubious activity from June 2007 includes blanking vandalism to the biographies of two Tennessee state politicians: Matthew Hill and David Davis. Someone who had access to a congressional office computer didn’t want the public to read properly cited information about their ties to the pharmaceutical industry.

I uncovered this information in ten minutes and my sysop tools weren’t necessary for any of the research. You can see for yourself:

None of those inappropriate edits remained in Wikipedia’s live version very long. Site volunteers reverted most of them one minute after implementation; the longest endured for nineteen minutes. What is notable for this discussion is how each action gets logged in site histories where it remains a public and voluntary disclosure. Anyone with an Internet connection can find the rest of that trail; it remains fresh as morning snow. A good follow-up exercise for SEO professionals would be to track the edits that led up to that IP address’s April 2007 blocks for link spamming the congressional black caucus website.

As an administrator I am continually surprised by how often I see edits that serve little purpose except to place the editor at risk for adverse news coverage. Few of the people who have a professional interest in knowing how Wikipedia operates actually possess more than a superficial understanding of its workings. Copyleft licensure, for example, requires that each action contain an authorship notation and ensures the information remains freely reproducible. Activities at Wikipedia are transparent, yet many individuals who have a professional reputation to protect behave as if their actions were guarded by an opacity the site does not possess. Except for article deletions and occasional courtesy blankings, Wikipedia’s archives remain public and accurate back to December 2001.

For each case that actually makes the papers I see perhaps two dozen that look newsworthy. Only a corresponding ignorance among investigative journalists has shielded the rest. The press has a habitual reliance on tips yet they have little inherent need for tips when the story relates to Wikipedia. In my estimate they will soon learn how to research leads for themselves. Wikipedia’s prominence as the Internet’s most popular nonprofit volunteer-driven site usually means a story cascades into international coverage once it breaks.

In January 2007 Microsoft learned a hard lesson in black hat tactics. Blogger Rick Jelliffe reported that Microsoft had offered to pay him to edit Wikipedia’s Open Office XML page on their behalf. The strong appearance of impropriety quickly turned that revelation into international news.

One point that was lost in most news reports was that Microsoft might actually have had a legitimate point about the balance of coverage on Wikipedia’s Open Office XML article. The folks at Microsoft could have improved that page while avoiding a PR debacle. Never mind what the Telegraph story says: I do not issue site bans simply because an editor has a conflict of interest. Nor, to my knowledge, does any Wikipedia administrator. We do ask people such as yourselves to act with discretion. So to address the original matter about Wikipedia being a search engine, site policy explicitly disavows that purpose.

Wikipedia’s Conflict of Interest guideline dovetails with several policies and experienced editors have written essays to provide supportive advice. These ought to be must-read material for any Search Engine Land regular. Rather than regurgitate what they already say I’ll summarize what I consider the most important points and add some suggestions of my own. These tips are not the official position of the Wikimedia Foundation, yet there’s a good chance that business editors who run afoul of policy will encounter either me or someone I’ve trained.

Wikipedia white hat activity in a nutshell: Designate a particular individual to be the Wikipedia liaison. Have that person register an account and declare the conflict of interest on the account’s user page. Then post suggested changes to article talk pages. For a variety of reasons this approach is safer and results in more durable changes than direct editing in conflict of interest situations.

Eight underused Wikipedia white hat strategies

Provide line citations. This is one point where an SEO professional’s interests often coincide with Wikipedia’s goals: factual verification is important to the project. When used judiciously, citations can be the most durable way to send traffic to your client’s website. Focus on topics where that site is strong on content and compliant with Wikipedia’s Reliable Sources guideline. In many instances a client’s site may be a self-published source, which limits how it can be used as a reference. Sourcing is welcome at articles that are already flagged with requests for citations. In other instances it is better to post preformatted citation suggestions at article talk pages. Supply text that summarizes the referenced content when making a citation, use wikimarkup, and conform to whatever citation format is already in use at the page. Act with care in order to avoid Wikipedia’s spam blacklist or criticism from volunteer editors. Tailor each suggestion to relevant content in the particular article.
Use edit summaries. These are courtesies to other editors who review your contributions in history files. Edit summaries are also an effective feedback to self-limit against link spam. If you can’t think of anything better to write than "inserting outgoing link", it’s time to rethink your practices.
Seek mentorship. Wikipedia’s Adopt-a-user mentorship program helps new users adjust to site standards. This can be particularly useful because some people come to the site with misconceptions drawn from inaccuracies in mainstream press reports about Wikipedia. Experienced editors and administrators interpret formal mentorship as a positive sign.
Get to know the "what links here" tool. Each Wikipedia page has an index that controls the internal incoming traffic. Wherever a Wikipedia article has a durable outgoing link to your website, examine its incoming links list for any obvious omissions, then run a text search on those omitted pages. If the title of the other article already exists in unlinked form, go ahead and link it. Otherwise compose sample linking text and propose it at that article’s talk page.
Utilize article categories. Another route that drives internal Wikipedia traffic is its category system. For instance, Wikipedia’s "Search engine optimization" article has four categories: "Internet advertising and promotion", "Internet terminology", "Search engine optimization", and "Internet marketing by method". Learn the category structure and add categories as appropriate at articles that link to your client’s site. Categories and wikilinks are also good ways to locate unreferenced article passages where your client’s site might become a citation.
Develop a watchlist. This tool provides swift notification of talk page replies and article changes. Use it to keep in touch with other editors. Watchlists also alert you to obvious vandalism and a track record of reverting vandalism helps you earn the respect of other Wikipedians.
Contact WikiProjects. WikiProjects are coordinating centers where Wikipedians with similar interests plan and prioritize article improvements. When you propose a major edit to an article talk page, a polite query for input at the relevant WikiProject can generate help and feedback.
Accept feedback. If your links are getting reverted and experienced editors are making complaints then you’ve probably misunderstood site standards. Pay attention to what they tell you, ask questions, and adjust your approach.

Although Wikipedians are understandably skeptical about conflict of interest editing, SEO professionals who respect the site as an encyclopedia rather than as a quirky search engine can earn acceptance from its volunteers. Look for approaches that reconcile your goal of sending traffic to websites with Wikipedia’s goal of being an informative and reliable first stop for research.