The second paragraph pretty much sums up where this article is headed;
"One of the hoped-for efficiencies of convergence was the practice of "author once, deliver many." This means trying to create one set of content that would be suitable for a large number of delivery formats-such as a design that would work both for broadcast or DVD as well as a compressed movie for the Web."
It then goes on to point out some of the factors that have failed to keep technical reality in step with developer and consumer desire, in particular the stupidity of the currently released video iPod, the "not quite ready for prime time" Flash player (although it's gotten a Hell of a lot better) and the still lagging uptake of high speed internet (and that's in the U.S.! Australia's pathetic excuse for broadband makes them look like South Korea by comparison).
A little shaky in parts, particularly where it moves from the video downside to the delivery downside, but it's a valid point. We live in a volatile media environment where everyone is trying to do their bit to be the "Universal medium" of the internet. While we live in this environment split between forward thinking digirati, luddite analogists and the vested interests in both camps, it has to be considered almost impossible to create video items that work flawlessly across all platforms. At this point in its development we live in a world of compromise.
But then, these challenges are sent to make us consider alternatives to the norm. Time to put the thinking cap back on. If anyone actually read this stuff I'd implore them to leave comments and ideas. At this point let's just chalk this up to "ideas and concepts that Leonard should revisit every six months or so."