Not in the sense that it's primarily defined by what it is not --- but rather, in the sense that increasingly the self-definition of what the press does is both negative and critical.
Consider, for example, the assertion that "it's not news when a plane lands safely." The underlying assumption here is that it is news when a plane fails to land safely. But now, consider --- was it news when the Shuttle landed safely, just a few weeks ago?
On the other hand, it is news when a suicide bomber blows himself up in Baghdad --- but not, apparently, when it happens at an OU game.
There is a theoretical way in which news could be defined in terms of information theory. I'll spare you the possible geek-fest for the moment and simply define "information" as "the size of the surprise". In other words, the more surprising it is, the more newsworthy it ought to be --- and for good reason, because something is surprising when it doesn't match what you expect.
But now consider some recent examples:
- as I noted, it's not much news when a suicide bomber blows up in Oklahoma, but it is in Baghdad --- yet we know Baghdad is a war zone, but don't think of Oklahoma that way;
- it's news that the power problem is serious in Baghdad, but not when the power generation capacity exceeds what Saddam's government ever provided;
- it's news when a mayor swears at the President on national TV about the President's supposed failings (and yes, that is a surprise, and news), but it's not news when the failings turn out to be the Mayor's --- even though that would seem to be a surprise too.
Why would this be? It's easy to just say it's a biased mainstream press, and I think there is certainly some aspect of that --- especially in the Bernie Goldberg sense that comes from an automatic comparison with what the people around you are saying. But then where does that bias come from? And how does that explain, for example, the Oklahoma game coverage?
I conjecture that the "mainstream press" model for what is news is not "information content", but rather the degree to which it corresponds to a preconceived model, which is essentially critical. In other words, it's "news" if it allows the press to criticize someone, preferably someone in power, and even more preferably someone with whom the reporter or editor doesn't agree.
This model, on first thought at least, seems very predictive: the news content of New Orleans was not the screwups themselves, but the criticism --- the places where things went well, as in Mississippi, aren't news, and the facts on the ground in New Orleans aren't news. Bombings in Baghdad allow criticism of the war --- but the more surprising progress in Iraq's economy, in civil order, in political organization, and in the declining capabilities of the revanchist and counter-revolutionary insurgency is not.
Jizm on a blue dress isn't news, at least at first --- but critical coverage of the spiking of the story is, at least for Drudge.
This is just a conjecture so far, and I can think of some possible counter-arguments, but it seems to me to stand up. If it does, I think it brings up an essential question.
Is criticism as criticism really the ideal or central function of the press?
 In fact, this is literally the way information is defined mathematically. Strictly, the definition should be "log base 2 of the size of the surprise" so that the result is in "binary decisions", or "bits". This might have been the key mathematical insight of the 20th century; it has led to some amazing things, which would be worth an essay of their own someday.