Marketwatch's media columnist Jon Friedman calls WSJ.com "a monument to the success that a fee-based Web site can still enjoy in 2007."

But is that really the case? My Portfolio blog-mate Felix Salmon has been making the case for why a free WSJ.com is a no-brainer.

It's not a perfect comparison, but a good test case would be to look at how NYTimes.com fared before and after TimesSelect.

Looking at the site's traffic (as per comScore) tells us that the Times probably realized the $10 million $20 million TimesSelect was generating was not enough to makeup for the missing page views and ad dollars a completely free site could generate.

The blue line in the graph below shows the average monthly traffic for competing news sites (CNN Digital Network, Yahoo, MSNBC Digital Network, Gannett which includes USA Today, and Fox News Digital Network) and the orange line shows the traffic for NYTimes.com:

As you can see, NYTimes.com suffered a large drop in traffic after TimesSelect was introduced in September 2005. While its competitors also saw a dip, they managed to recover and grow traffic.

Page views at NYTimes.com reached their trough in July of 2006 and started to trend upward. While seasonal traffic patterns surely had something to do with that, three other developments probably helped out as well:

In March 2005, the company completed its acquisition of About.com. The site might not have had great content, but it knew how to show up in search results. Did it take 16 months for About's search engine optimization know-how to be incorporated into NYTimes.com? I don't know, but it's not a stretch.

In March 2006, the site launched it's first major (traffic-generating) blog, DealBook, since then they've added about 30 more.

And in April of 2006 NYTimes.com introduced Times Topics, it's version of Wikipedia, except filled by polished Times content. Do a search for "global warming" and you'll see the Times Topic page for it in the first set of search results. Other searches, however, aren't nearly as successful. But, now that much of NYT's archived content is free, traffic generated from these pages should spike.

Now, the interesting question is how much traffic did NYTimes.com miss out on by putting Maureen Dowd, Thomas Friedman, and Paul Krugman behind a pay wall? To get a sense of this, the graph below shows what NYTimes.com traffic would have been had it experienced the same monthly growth as its competitors. The orange line shows the actual traffic while the gray line shows the hypothetical numbers:

This translates into about 1.3 billion page views (for reference: MySpace gets 3 billion page views a day).

One argument for putting opinion content behind a pay wall was that ads weren't being sold on those pages at the time. That is no longer the case.

So is 1.3 billion worth page views $10 million? $20 million over two years? Not knowing anything about their inventory, I'd argue yes.

(Full disclosure: I used to work for The Times. None of the information for this post came from inside knowledge of the company.)