Metcalf argues that the increasingly popularity of watching television by the season, rather than by the episode, is changing the very nature of what we want from television. (And it's a circular process in which the changing nature of television encourages and rewards such binge viewing.)
Once upon a time, TV was an episodic medium, in which characters and their circumstances were not allowed to change; now, a season of a series (or a series as a whole) is viewed as a complete work which is primarily devoted not to telling a separate story each week, but a single overarching story that may last for several years. There may be contained stories in each episode to keep the casual viewer involved, but the show is meant to absorbed as a whole; it is a "DVD novel," as opposed to a series of short stories.
There are obvious elements to that change -- series that have ending dates planned in advance, and stories that come to a firm ending -- but Metcalf also sees this change in some less obvious ways -- the rise of the unsympathetic lead, the increased blending of genres, the shorter seasons that dominate cable, a visual style and vocabulary borrowed from comics and graphic literature.
These aren't all new changes, or at least not as new as we might think. Metcalf goes as far back as The Fugitive for the roots of the series-long arc in American television (will Richard Kimble ever find that one-armed man?). And he traces much of the change to the influence of British television, in which seasons have usually been shorter -- as few as 6 or 7 episodes -- and self-contained programs of only a few hours have always been part of the TV landscape. (We went through a brief craze for those in the US; we called them "miniseries.")
And at the root of all of this, Metcalf places Dennis Potter's 7-hour series The Singing Detective, which featured a distinctively unpleasant protagonist, and which mixed forms like mad -- reality with hallucination, noir with musical numbers, realism with fantasy; it was a formative influence on Steven Bochco, whose Hill Street Blues and L.A. Law began the slow serialization of American television.
Metcalf's writing isn't overly scholarly, and for the most part, his arguments are clear and easy to follow. The biggest exception, for me, is in his chapters on comedy, where he argues that we've shifted away from situation comedy to "attitudinal comedy;" it was never quite clear to me what distinction he was trying to draw.
And there are a handful of annoying factual errors that any good fact checker should have caught. The L.A. Law character is named Michael Kuzak, not Kusack; All in the Family began in 1971, not 1968. Such mistakes always make a bit nervous about how many other errors are lurking that I'm not informed enough to catch.
Still, those details aren't at the heart of Metcalf's premise, which is fascinating, and which he presents in convincing fashion.