Yes, disagree.

1. The importance of a piece of scholarly work need not be immediately apparent to its contemporaries.

2. Being cited is a poor proxy for importance.

3. Work that is cited is rarely paraphrased or duplicated in a meaningful way.

4. Paraphrased citations are a poor proxy for canonical source; papers are often cited incompletely or sometimes outright inaccurately.

5. These are the fruits of people's labour. They spent days and months producing them. To lose them, especially when digital copies are so cheap, is an unnecessary disregard for said labour.

I disagree, I cite important work all the time and you definitely need the original paper. Paraphrasing isn't really done enough to negate the need for the original paper, otherwise what kind of lame plagiarized paper are you writing? Citations only tell you where to look, and if that article is gone then you're screwed.

What do you mean by "duplicated elsewhere"? Sitting on some scientist's hard drive doesn't count, it has to be discoverable. That's what the issue is: when journals die, how do we ensure that the papers are saved somewhere easily searchable?

We don't always know what is important at the time it's published, and there are many examples of this in the literature.

Meanwhile people are working hard to preserve every Commodore 64 game.

Part of the point of citing another article is that you don't then have to repeat all of it. So if B is cited by A, and B is no longer available, it's not really possible to read and understand A either. And likely A used information in B to justify some claim, which is now weaker.