> Sadly, online comments aren't independent statistical samples from your entire user base.

That idea cuts both ways. Sadly, also because they are not independent, the chance that those that comment and participate are probably also the ones that drive the community and make for nice discussion, which those other 98% who just browse end up reading. What else is Reddit if not links to contents followed by comments.

Those 2% stop, the site goes to crap.

You're making a huge, possibly fatal, assumption about quality remaining constant.

Quality of submissions or comments?

I was part of another forum-based company that went from zero advertising to full display advertising with banners and adsense (the ads brought in another $200k to $500k per month). Some of the vocal members (read: always bitchy about everything, even mild color or margin changes) complained at first, then went back to their normal bitching about everything else in the world.

All changes hurt at first, but we can't not change everything forever.

Quality of the site as a whole, both subs and comments.

Again, you're assuming that the presence of additional advertising will have no effect on the content. Perhaps your example webboard always sucked, you know?

One additional way to look at it is to gauge whether your webboard is more popular and/or more highly trafficked than its related subreddits. This last point doesn't necessarily speak to content quality, but it does relate to the relationship and motivations of its community.

A tenet of advertising is to maintain a blindspot for self-reflectivity, to assume that everything would be the same whether they were present or not. Leaving aside the economics of production, for a simple example of TV shows, a lack of advertising would enable longer shows and thus more content, which is what people presumably watch these shows for. A negligible number of people watch TV only for the advertising content, and I think we can agree that a station that is only comprised of advertising would not be very popular. QVC will never get the ratings that "Duck Dynasty" does, much less one composed entirely of 30sec spots.

So you have one decent example where such a change didnt really alter the quality of the site, yet that doesnt magically apply to every scenario.

You're making a huge, possibly fatal assumption that Seiji views human beings as anything other than clicking machines attached to credit cards, there to be exploited.

As Kant said: "view people always as a means, never as an end."

Just for posterity: that's not my view, but that is the exact view of visionary sv asshole ceos (note: not of the good ceos, just the jerk ceos).

In their eyes, you only exist as "traffic" for them to exploit. You are not an individual. You have no life. You are a finger attached to a pointing device attached to a credit card.

Their goal in life is to divert Internet traffic to their own sites by any means necessary to increase their own traffic numbers (views, hits, sessions, clicks, impressions, ...) thereby increasing their own perceived worth so they can spout their vanity asshole metrics off to boards/investors/audiences.

[Note: I may be bitter about jerks exploiting the world raking in more success/money/exposure than non-exploiting builders.]

We who shall be fleeced salute you.

http://en.wikipedia.org/wiki/Ave_Caesar_morituri_te_salutant

That's missing a number of points, both historical and behavioral. Digg.com died when they tried to increase ads and there was a mass exodus from Digg to Reddit. Additionally, even if only those 2% leave, you're still screwed because those 2% produce all the content/share all the links that the other 98% want.

Rule of thumb: if you're business is selling your users to advertisers, don't piss off your users, because then you'll have no product.

> What actually happens

You have Digg as an example of what happens.