because of Cowen linking it

That's how I found it, too. But I need the LessWrong karma and you don't. :D

What did you make of it?

If I followed the discussion of circulating versus fixed capital, and gross versus net increase, Ricardo is showing that (in modern jargon as opposed to Victorian jargon) if you set the elasticities correctly, you can make a new machine decrease total wages in spite of substitution effects. He seems to think about this in terms of the "carrying capacity" of the economy, ie the total population size, presumably because Victorian economists worked much closer to true Malthusian conditions than ours do. In other words it's a bit of a model, not necessarily related to any particular economic change that has ever actually happened. Possibly you could get the same result re-published today if you put it in modern jargon with some nice equations, but it would be one of those papers that basically say "If we set variable X to extreme value Y, what happens?" So it's probably not that important when discussing actual machinery, as Ricardo acknowledges; he's exploring the edges of the parameter space.