Wikipedia:A story from the early days of the web

On 31 August 2011, Jimbo Wales wrote this essay on his talk page diff, providing a rationale as to why paid-editing is banned. The full text:

This is just a story related to the discussion of paid editing, up above.

In the pre-Google days there was great competition between Yahoo and Altavista and several others. One of the questions out there was the question of algorithmic search versus human curated directories. Algorithmic search won out for the most part, in the long run, but my story is not about that.

Yahoo hired teams of editors to review websites and list them in their directory. I don't remember now exactly how many people this was, but I think it safe to say it was in the hundreds. These editors could find websites any old way, of course, and include whatever in their judgment was worth including, but there was also the possibility of submitting your site to Yahoo.

Yahoo at the time was incredibly powerful and so of course the submission queue was voluminous - to the point that it was nearly useless.

Someone at Yahoo then had the idea of "paid submission". It still exists today. It costs $299 annually(!) for most sites, $600 annually for "adult content and/or services".

Yahoo insisted that paying for expedited review of your website was not a compromise on the editorial neutrality of the directory. But the public perception was very strongly negative.

Once they took that step, the obvious incentive structure means that it's in Yahoo's interest to give a favorable review. Let's say I have a movie website, and I pay to get one page of it reviewed by Yahoo. If that page is rejected, I won't submit again. If my page about Clint Eastwood is accepted, and the amount of traffic I receive is worth it, I'll pay again and again. Maybe I'll submit 1000 pages, and pay $300,000. That's real money.

Acceptance of a really crappy page might be bad for Yahoo, of course, but notice that the cost/benefit analysis has shifted for Yahoo. They have a strong financial incentive to list my site as long as I don't do more than $300,000 per year in damage to the Yahoo brand.

Yahoo liked to insist that this wouldn't happen, but the public trust in Yahoo was diminished. Today, of course, Yahoo is no longer regarded as a dominant leader, and I think that shortsighted moves like this are a big part of the reason why. (That algorithmic search turned out to be the right answer in most cases is of course also a part of it.)

If you want to buy Google today, well the total market value of the stock is 174 billion. Yahoo, one tenth of that at 17.4 billion.

Now let's apply this line of thought to newspapers. We all know that newspapers make money from advertising, and that quality newspapers do take steps to isolate the editorial department from the advertising department. It's not perfect, but the system does mostly work.

Now imagine that the New York Times announced a program. For a $10,000 fee, you can pay them to send around a reporter to write a story about you. Imagine that it is claimed that this is no guarantee of the story actually being published. It still has to go through the normal processes and procedures, it is said. How would that impact the credibility of the newspaper?

My view is that it would be incredibly destructive. As per what I outlined above, simple financial incentives suggest that large companies would give it a try a few times, and if it resulted in favorable coverage they wouldn't have gotten, they'd do it again and again. And if it was a waste of money, they wouldn't do it again.

With advertising we worry about the indirect influence of the money on the editorial staff. That's problematic enough. But when connection between pay and getting coverage is made direct in this fashion - bleh.

Now imagine that you're a member of the general public and you read a story in the newspaper about Wikipedia. Two possible story lines. In one version, it's "Wikipedia announces paid submission program" - in a sudden change of heart and policy, Wikipedia has decided to allow a formal program whereby experienced Wikipedia editors are paid by PR companies to write articles for Wikipedia. Oh no, we insist, nothing changes about our editorial policies, of course not. People would rightly be deeply concerned about that. Suddenly people would read articles in Wikipedia and wonder - how tainted is this by the formal acceptance of Wikipedians being paid to write on behalf of companies? In the other version, it's "Wikipedia reiterates its stance against PR firms editing Wikipedia". The story is that Wikipedia editors have firmly rejected the concept of allowing people to come into Wikipedia as paid advocates to edit articles, due to the wrongness of the financial incentives, and the blurring of the passionate pursuit of the truth that has been a hallmark of the Wikipedia community.

For me, this has been, and continues to be, an absolute principle. Paid advocacy is banned from Wikipedia.

The objections that are often raised are not remotely compelling. Claiming that banning it only pushes it underground doesn't make sense, as there seems to be virtually no evidence for it. Most responsible PR firms understand that editing Wikipedia on behalf of clients is forbidden, and they have rules in place internally to prohibit it. Of course, this is a big place, and everything goes on to some extent - the goal is not to achieve perfection, but to have the right principles in place.

Another response to this objection is that it ignores that PR firms have a perfectly valid way to interact with Wikipedia, well-respected by the community, totally above-board and ethical. And that's to post to the talk page, declaring your conflict of interest, and asking people to take another look at something. That's the ethical approach, and it works. It completely prevents the question of whether or not what ends up in Wikipedia ended up that way because the Wikipedians themselves are corrupt.