Last post on this. But worth noting:
I’m certainly not one to come to the DMN’s rescue, but you’re misusing the Alexa data in this context. Alexa is not measuring pageviews. It is measuring how many out of every million pageviews (of the subset they track) go to a given site. So pre-2004, there was a lot less on the internet. Now there is more there–YouTube, blogs, you name it. Our moms and grandmas have gone online. So the DMN’s pageviews have probably gone up over that period, but not proportionally to the overall traffic tracked by Alexa. You’ll notice many sites that have been around a long time have similar dips. In this instance, Alexa isn’t useful for mapping the trend.
See that the Chronicle has a similar pattern.