Examining Wikipedia's Bias

Oct 30th

Some people are up in arms about the idea of Wikipedia adding ads to their site. The issue is not that ads are hated. The true issue with the Wikipedia and advertising is this:

The issue is not targeting or relevancy... the issue is that some will feel it is bait and switch. That something they thought was pure and easy to believe in now suddenly is part of the real world.

The truth is that the Wikipedia has always been chuck full of ads. I am not talking about the link spam that people sneak in, or when people promote their own brands, I am talking about the mindset with which Wikipedia articles are drafted. Lets look at the search engine optimization article.

Classification:

First of all, lets start with the classification and associated fields:
Wikipedia classifies SEO as spamming height=
Even Google's guidance on hiring an SEO, which is quite biased (and self serving) in nature, probably is not as biased as the Wikipedia's classification of SEO.

Now lets compare that frame of reference to the opinion of Google's lead engineer in charge of search quality. From my interview with Matt Cutts, where I asked Is all SEO spam? His response was:

Absolutely not--I need to do a post about this on my blog sometime. Lots and lots of search engine optimization is white-hat and not spam at all.

The way Wikipedia classifies SEO is an advertisement biased against the entire field of SEO, and thus acts as an ad for search engines and pay per click marketing.

Accepted Types of Information:

I knew that directly linking to my site or directly marketing myself on Wikipedia was not going to go to far with them generally hating the field of SEO so much. On the other hand, I knew their vile hatred of the field meant that me mentioning Traffic Power and linking to articles about Traffic Power that link to my site would stay in that article forever. And they have stuck thusfar.

Framing:

The Wikipedia states:

When discovered, search engines may take action against those found to be using unethical SEO methods.

Why is ethics even tied to SEO techniques? Machines can't have ethics. When their results are inaccurate that must be the fault of some external third party with low ethical standards? What is that?

Participation:

From Rough Type:

"Wikipedia hasn't been a real 'wiki' where anyone can write and edit for quite a while now." A few months ago, in the wake of controversies about the quality and reliability of the free encyclopedia's content, the Wikipedian powers-that-be - its "administrators" - abandoned the work's founding ideal of being the "ULTIMATE 'open' format" and tightened the restrictions on editing. In addition to banning some contributors from the site, the administrators adopted an "official policy" of what they called, in good Orwellian fashion, "semi-protection" to prevent "vandals" (also known as people) from messing with their open encyclopedia.

There is a bias toward those who want to talk down or shine a negative light on the field of SEO while true topical experts are driven off. Google founders Sergey Brin and Lawrence Page mentioned Danny Sullivan in some of their early research, and yet Wikipedians ran him out of the topic. Danny is probably the single most authoritative voice on search.

Incentive:

If I know my words are probably going to get edited out of the Wikipedia where is the incentive for me to put effort into editing there when my work is much more likely to be respected and profitable if I put it on my own site?

Staleness:

Not only does the classification and writing within Wikipedia reinforce the worldview pushed by the most powerful associated commercial entity (Google), but the types of things that are reference worthy are "famous" SEOs, which is going to be inherently biased toward people who established strong brands many years ago.

Which of the listed famous SEOs have entered the field this decade? None of them.
Famous SEOs Listed on Wikipedia.
I have no doubt in my mind that many people newer to the SEO market than I know far more than I do.

Also as fields and language itself evolve will the large cross referenced content base that is the Wikipedia even be able to keep up with rapidly changing markets or linguistic changes?

General Factual Errors:

The SEO article on Wikipedia also states

Yahoo! and MSN Search do not automatically punish entire websites for small amounts of hidden text. Google's market share of daily searches has fallen rapidly from 75% to 56% over the past few years, as other search engines find many web pages that Google has banned and cannot display due to Google's severely limited index.

One would have to live under a rock, having no access website referral logs, the news, or financial markets to believe that Google has been drastically losing market share to competing search companies.

The ease with which people can edit the Wikipedia creates a bias toward quickly adding incorrect factoids, while discouraging true topical experts from participating, especially if their opinion is likely to get edited out if it does not conform to the flavor of the day group-think.

One simple fact that must be accepted as the basis for any intellectual work is that truth "whatever definition of that word you may subscribe to" is not democratically determined. And another is that talent, whether for soccer or for exposition, is not equally distributed across the population, while a robust confidence is one's own views apparently is. If there is a systemic bias in Wikipedia, it is to have ignored so far these inescapable facts.

Conclusion:

I know one article is a small sample, and am not saying that I think the Wikipedia is a bad source for everything, just that in rapidly changing fields of commercial interests the Wikipedia is one of the last sources I would trust for an accurate view of the market. It is more representative of an advertisement that the most powerful sources in a market tell people that they should be thinking about.

Amateurs vs Professionals & Advertising vs Content

Oct 20th

Brett Tabke recently created a supporter's only thread about the potential downfall of blog ad networks, claiming that they may end up undermining our ability to trust what we read. Bill Hartzer (who I am generally a big fan of) added

There are still unbiased sources out there, you just have to look for them.

I responded to the thread with (roughly) the following (edited on my blog for better formatting and grammar, and I added more depth to my opinions here):

There is no such thing as an unbiased source. Unbiased = unreal.

I think as user / consumer is transferred into a market participant beyond just what they consume that we will

  • see our own influence (and influences) better

  • take better care of our attention
  • be more likely to find things we are passionate in
  • get better at judging the intent of others
  • generally trust most things we see less.

While on the surface it is easy to paint that lack of trust as a negative thing, I think a lack of trust toward authority (ie: questioning what you see, why you were shown it, and who placed it there) is an important component in any functional society.

The only reason that learning to not trust what you see is a negative is because there is so much fraud in the world perpetrated by power source who only retain power through the ignorance of the average citizen. Why are most articles in the mainstream media about SEO usually focused on black hat techniques? Anything that challenges any established authority system is deemed to be wrong by default, especially when evaluated by existing sources of power.

Would I have joined the military if I knew more about the military industrial complex? Not a snowball's chance in hell. Should I be quiet about them doing illegal things like destroying some of my work records prior to processing me out of the military? Not a snowball's chance in hell.

I believe consumer generated media will transfer power away from macro-parasites toward creative and passionate individuals who are driven to change the world.

I also think that anyone who communicates, even if only for themselves, is selling something...even if that potential gain is just trying to understand our own faults and why we think the way we do.

On another front, which is more ethical and legitimate? Blindly trusting an ad system that promotes products you know nothing about and is pushed to no end by the goal of achieving an efficient market. Or, writing about things you know about, and occasionally getting paid for the value of your time, feedback, and influence?

How relevant is a Google AdLink with my name and brand name in it that links to a list of ads that does not even include me? How is that any more legitimate than getting paid to review things you find interesting?

Some of my other blogs have no commercial intent to them, but they still rank for a lot of things, and I still learn a bunch from other's feedback, and I also think I learn a lot about myself by reading how I was thinking when I was doing different things.

The biggest thing that is killing off traditional publishing is the lack of personality, lack of passion, and a lack of bias (or watered down pro corporate bias) which is contained in nearly every piece of content they create.

I would rather read a passionate author than one that abides by some arbitrarily crafted ethical standards. Would a newspaper ever publish a self analysis like this? No. And if I read a person long enough I can understand their biases much greater than I can by reading random published articles. And if earning the trust of readers is harder then it will be valued more.

The big reason that people are against open networks, paid placement, free markets, paying individuals what they are worth, or anything that redistributes power is that many of the most powerful sources in the world stand to lose power if we question authority. And so they must play down the roll of or try to undermine the credibility of competing business models (or anything that threatens the ideology they sell or their business model). Nothing new there, it has been going on forever (even if the sources of power do not hold themselves to the same standards they want to hold amateurs to).

History - a Cool Marketing Idea

Not sure if this idea exists already, but it would be a cool project idea for anyone ambitious enough to do it. What about creating a social network site that leverages famous poems, speeches, and quotations, and integrated them into the web by allowing submitters to add links to famous text that existed before the web. The links could show

  • how the meanings of words changed over time

  • how static human nature is
  • how politicians lie and lied
  • how religious material changes over time
  • how bogus and misguided most forms of patriotism are
  • whether cultural norms should change
  • or anything else you are interested in

There is a lot of marketing potential in history. Google realizes it, and is already exploiting it, but not to its full potential. Invariably traditional publishers are losing control due to network efficiency. Warner is already threatening to sue Google over YouTube, but YouTube just sold for 1.65 billion. I think historical text (and maybe personalized versions of it) is another vertical which is low hanging fruit like video once was.

Part of Google's move toward trying to be the default hard drive for different types of information is such that they can add context to whatever you are doing. Some of that context will be relevant ads, but the other piece of it will be useful related ideas based on other's usage data.

I think the best way to make the wisdom of the past appealing to a wide audience would be by making it interactive and showing how it is relevant to today. Some amount of that can be automated, but given how many people are trying to interpret the meaning of lyrics and how layered great writing is you would think there is a market for adding personalized or opinionated context to historical text.

Announcing ReviewMe!

In April I mentioned that I wanted to create some sort of a social network. I left that description intentionally broad such as to not tip my hand too much. But the idea was a social ad network.

I had a good idea, but hate the idea of having employees and running a company. I want to be able to travel and explore the world, so the idea required a partner. ;) I brought my idea to Andy Hagans, and he was up for running the show. The idea is to create a blog advertising platform that allows advertisers to contact related bloggers to ask if they would review their products or services. Our network is called ReviewMe, and will be launching soon!

While I got into the web as an SEO, I tend to think of myself more as a blogger and viral marketer than an SEO. Viral marketing was the idea behind ReviewMe. It took us a while to get the model and infrastructure down. Since we started working on the projects other blog ad networks launched and one even got VC funding, but I believe our model is going to be somewhat unique and offer a high value when compared to other businesses in the same space.

I think writing a lot and reading a lot about marketing put me in a unique position. After acquiring Threadwatch, while still building this site and others, I started getting pitched more and more frequently. It made me think that there could be a formal marketplace which made it more efficient to ask bloggers for reviews, and also removed some of the potential risks associated with pitching to bloggers. The last thing you want is a popular blogger calling you a spammer, because that stuff tends to rank well.

Four elements which will work nice in our network to filter out bad products and bad offers are

  • bloggers will disclose their relationship with the advertiser

  • bloggers only review things that are interested in
  • we encourage brutal honesty
  • the comment sections on popular blogs will help keep advertisers and bloggers honest

Fraudsters and advertisers with junk offers will not want to risk paying people to write reviews that may expose their business flaws. But, if you have a good product honest feedback and conversation about your business should only help you. Getting great feedback early on in a product's life-cycle can save millions of dollars in the long haul.

I also think this is a more efficient way of selling cost per influence than some of the other networks. Buy a sitewide ad in the right rail of a popular blog. Compare how much traffic that sends to the amount of traffic sent to links in the content area of the blog, and you will see that the influence is in the content area of the site, not near it.

In addition, I think the real value of blogs is the unique feedback you can get from the blogger. Do they like your idea or hate it? And why? What advice can they offer you on how to improve your business?

We are not going to try to create the largest and most efficient ad network in the world. That's Google's job. Rather than trying to squeeze a few more cents out of an ad space, our idea is to extend the value of advertising by coupling it with reviews and conversation on popular sites. Brand building is much more about conversation and community involvement than it is about targeting keywords and displaying ads.

ReviewMe hasn't even launched yet, and I am more excited about it than any site I have ever worked on. You can read Andy's official announcement here, and read up on the latest developments on the ReviewMe blog.

Research, Scraps, Ordered Lists, & Social Structures

I am still busy busy busy redrafting SEO Book and other content, but a couple recent comments made me want to make a quick post. On my post about why I thought it was alright to mention politics in work blogs Andrew Goodman came buy and left a gem of a comment

Been thinking this over. I don't often post on politics in spite of having an extensive background in political studies. Maybe that's because I learned you need to have eight chapters of literature review, history, and facts, and three chapters of case study, before you get to write the two chapters with the conclusions. It's easy to dump on obvious miserable failures -- much harder to imagine and/or implement a better or perfect world, at any level.

And given that the web encourages lousy content, having to throw away or leave unused a large amount of research is brutal if you are looking at content production on a ROI basis. To do so, you almost have to be certain that your research is going to be so outstanding that others notice it, or you have to be creating it out of passion without much regard to finance.

One of the reasons you see so many lists of items on popular blogs and social news sites is that they allow you to collect these random scraps, slap them together, format it, and GIVE THE IMPRESSION that the work is well researched and comprehensive, even if it was not. Little to no waste in the formatting, and rather than doing a lot of work that doesn't show you look like you did far more work than you did.

And the reason factiod posts do so well is not only that impression that they are a lot of hard work, but also:

  • they are at a low enough level that most people can understand them

  • people are attention starved, and the ideas are usually broken into small bits easy to digest
  • at least one of the ideas in the list will be easy to identify with (as an example, I once told a story of how I was an idiot and accidentally dialed 911...most people had no comment or interest. the only person who expressed interest later revealed that the did the same thing)

Right now I have roughly 50 or so draft posts saved, and whenever I want to I can finish one up or use chunks of it to help create content for another related post.

These scraps of knowledge (or factoids, if you will) are not only big on blogs and meta news sites, but also are largely what most any user generated content sites and what the Wikipedia consist of. I used to be (and maybe still am) so anti authoritarian that I view most everything that starts from bottoms up as being better than things that are top down, but in many spheres it probably does make sense to have human editors, human aggregators, and trusted topical authorities that exist somewhere in the middle.

Search isn't successful because the technology is so great, it works well because they do have human editors, and they use your and my links as signs of trust.

When you look at the Wikipedia page on SEO and read through the talk archives you will see that they ran off both Bill Slawski and Danny Sullivan. Is there another topical expert that could possibly be more qualified to talk about search than Danny Sullivan? Not that I know of.

Some of the best topical experts have no distribution because people can not understand them or identify with them. Other topical experts may be good at communicating their ideas, but still can only reach certain audiences due to the errors of authority structures. For example, imagine if everyone followed the law, would we still need police officers, or would the laws change to create the need for the job and creat some criminal class to control the remainder of the populous?

People who just reach a bit of authority often like to feel a self-aggrandizing level of importance, and use a mob mentality to express it. Self preservation and a sense of purpose are probably the two key goals to any social structure or any person heavily committed to one. The more you try to convince them they are wrong the more you get flamed to bits, even if all they are protecting are their rights to remain ignorant and feel important.
And wherever there is conflict and/or brainwashing the all knowing experts of all topical domains are not going to be able to see past their own biases and brainwashing, and I doubt people can create rule sets or software which does a good job of avoiding that. Thus anything with significant reach and a bottoms up approach is typically going to be biased toward conventional wisdom, perhaps offset with a few outside fanatical voices.

Marketing works (and will never go away) because humans have inherent flaws, limited attention spans, and the market for something to believe in is infinite. But any structure that becomes authoritative is going to need to fire some of its top users if it is to stay relevant.

Larry Sanger, a co-founder of Wikipedia, recently announced Citizendium, which is sorta going to be like Wikipedia, but it will also have content verified by topical experts. I think I was the first donor to the project, and I would love to see it take off.

But I wonder if authority is the enemy of any social project. You want the authority because you get the distribution, but after you start to gain it you get gamed to bits and people start letting your content and software represent a large part of their identities or worldviews and it all goes to crap fast.

One of the more important reasons to try to grow out slowly and not force it too much is that you get to react many times before you get big. People who get rich fast often get poor fast. Sites that have their authority grow beyond their programming skills will have their flaws exposed so heavily that it presents a great marketing opportunity for others aiming to enter the same market.

Topical Expert Available, Only 12 Cents a Day

PeterD flames ChaCha. And can you fault him? What is up with a search engine that takes forever to answer? How good can their topical experts be at $5 an hour? How can you respect a topical expert who sits at your beck and call to earn only $5 an hour? And with an earning cap at $20 an hour? If you chose to use ChaCha hopefully your questions are not related to business, entrepreneurialism, capitalism, marketing, or finance.

The biggest reasons that ChaCha will fail though are not just inefficiency, the low expert payout, and having to wait for results. I think that the model causes other (and worse) side effects.

Right now if I search and buy something bad I am likely to feel it was my fault for being a sucker...like I misused search. If a paid guide leads me astray and I take their advice then I feel they are the ones at fault. So ChaCha shifts the blame from me to the engine.

Another big problem with the pause in the search process is that wants / desires / impulse purchases are going to be far less appealing if I have more time to think about and refine my thoughts (and have to share that thought process with others), rather than just being able to say it was an impulse purchase.

One of the biggest errors I have done (likely wasting at least a couple hundred thousand dollars) was answering tens of thousands of questions via emails and doing nothing with that content. When you have the ability to recycle content or make it valuable for many people rather than just one or two the cost can be greatly subsidized by many people over time. An inefficient and ineffective model becomes practical once you can use time and small distributed demand as an advantage. Ask Dave Taylor is a great example of a smart question answering site. Over time that model pays him far more than ChaCha ever could, plus it helps him build a brand and relationships that editors at a search engine would not be able to build.

Put another way, the money is in the archives, stupid. And that is why Google is so hungry to expand their archive any way they can, even if they do not make direct revenue off it right away. This is probably far more important than most people think it is.

What is the Difference...

Sep 2nd

Between slapping ads on cloaked content, search results, other's content, link blogging, and thin content meta journalism sites?

Which, if any, of these will be viable for years to come? Especially as all of those markets flood, and automated content generation becomes more and more useful?

Self Reinforcing Market Positions

Aug 16th

"I am the Center of the Universe" - random A list blogger.

An innocent fraud is a lie, but it's a lie that's more white than black. It's a lie that makes most everyone happy. It suits the purposes of the powerful because it masks the full extent of their power, and it suits the purposes of the powerless because it masks the full extent of their powerlessness.

Most of the people blogging about making money probably do not make that much money. Most people selling how to (insert your topic here) advice also fall in the same category. And they do it off the backs of people who link at them hoping to one day do the same. But in reality most people fail because it usually takes quite a bit of innovation, time, effort, risk, personality, or passion to break out of the mold, and many self-reinforcing institutions and social norms make it hard to succeed.

I think I have been learning enough about social networks, sociology, psychology, marketing, business models, authority structures, etc. that if SEO ever somehow lost its direct value that I would still be able to do well, but imagine the day that a field you studied for years was rendered useless. Would you instantly be able to change your model or pick another field? Would you keep pushing your ideas even after you knew they hurt more people than they helped (like the old LinksToYou link farm did)? Where do you draw the line? Or imagine that if many people you wanted to help never gained anything out of it other than the ability to help you grow more authoritative while they paid you with their time, attention, trust, link equity, and perhaps cash too.

Of course there is the hope that those things are not true, but the value and quality of advice you get from people (as well as how accessible their ideas are) is not just a function of what they know, but also market timing. Anyone who is doing well on the web right now was born it to some amazingly lucky timing to have found the web while it is still so nascent. Most people and/or business models that get to the top of a social structure have some idea how it works, and would never want to admit that their structure is overrated or their field has died. All that comfort, all that self reinforcing market position would be erased.

Why does Matt Cutts warn people about certain types of links? Google's authority is based on links representing relevancy. Without relevant links search has no ad based business model.

As the blogophere has become more rigidly hierarchical, not by design but as a natural consequence of hyperlinking patterns, filtering algorithms, aggregation engines, and subscription and syndication technologies, not to mention human nature, it has turned into a grand system of patronage operated - with the best of intentions, mind you - by a tiny, self-perpetuating elite.

Much like traditional media there are certain biases to blogging and web publishing.

  • Old sites get more exposure than new ones.

  • Controversy spreads fast.
  • Lists and types of bite sized content that offer immediate reward to an attention and time scare audience typically spread further than content which requires more attention. The attention deficit most of us live with is going to constrain the types of ideas that are profitable.
  • Better tracking and targeting, more social networks and meme trackers, cheaper and more efficient distribution, more feedback loops, and ad targeting engines that block certain words or categories are making it easier for the average publisher to know how profitable writing about an idea is before they even type the first key.
  • If my SEO for Firefox extension was SEO for IE7 it would have got about 12 links instead of a couple thousand. Is that group think linking legitimate?

It feels weird sometimes when you come across some of the self reinforcing patterns in action...like when you predict an idea will spread for a specific reason then see it happen, or see a high ranked article from someone talking about a topic they clearly demonstrate they know nothing about, or something spreading quick as correct when it was factually incorrect garbage the day it was published.

And that is another part of the reason it is so easy to rely on your established authority. The fear of being called out (some economics students hated my post on central banks) when trying to learn something new. And thinking of all the time and effort required to get back into another self reinforcing market position.

But relevancy is a personal thing. The market for something to believe in is infinite. Those who can get in early and evangelize their field will likely profit from it long after their techniques are rendered useless or their field has died. And if you are associated with an important market then your distribution and the self reinforcing nature of search will allow you to heard in other markets as well.

Google Vertical Search Canibalizing Google's Organic SERPs

Aug 14th

I searched to see if the movie An Inconvenient Truth was playing in a local theater. Google not only showed the Movie OneBox result, and offer a movie search feature, but they also rank the Google Video trailer in their search results and are caching the movies result page. Loren recently posted an in depth article showing how much Google is doing to add interactivity to and exposure for Google Video.

As Google adds features and consumer generated media to Google hosted vertical content pages many review sites and thin sites in high margin verticals will lose a good portion of their value, link equity, and traffic. A big thing that places Google ahead of most review sites is that they will not only collect and structure their own feedback, but their knowledge of language and the web graph makes it easy to access some of the best review information on other sites.

In a couple clicks I can go from reading feedback on Google to reading aggregated feedback snippets from other sites to reading some of the other best reviews on the web. For example, it takes little effort to see the official site, the contempt some sectors show the film, a more objective review, and a speech which inspired the creation of the film.

Bite ¿Byte? Sized Content

Aug 11th

Recently Google allowed you to link to an exact minute and second of video. They also give each page of a book its own URL.

Pages






    Email Address
    Pick a Username
    Yes, please send me "7 Days to SEO Success" mini-course (a $57 value) for free.

    Learn More

    We value your privacy. We will not rent or sell your email address.