We're talking about mostly-text-pages which have no sensible reason to rely on JavaScript to render, not rich webapps. In these cases I consider the text to be a meaningful representation for the purposes of indexing or archive.
I don't agree with this:
> Search engines and archival tools can handle all this very well today.
I think most developers have been in a situation where they had to explain to users that a moon on a stick isn't easy just because Google can do it. It may be commonplace, but it's not easy or cheap.
The difference between real HTML and JS execution is orders of magnitude of time, expense, maintenance, and stuff to break. That really can't be underestimated. It's the difference between a 500 millisecond call to curl and a whole headless browser and a 5 second wait.
I hope you don't think I'm putting this too strongly, but this raised bar is a force in the monopolisation and de-diversification of the web.
I don't agree with this:
> Search engines and archival tools can handle all this very well today.
I think most developers have been in a situation where they had to explain to users that a moon on a stick isn't easy just because Google can do it. It may be commonplace, but it's not easy or cheap.
The difference between real HTML and JS execution is orders of magnitude of time, expense, maintenance, and stuff to break. That really can't be underestimated. It's the difference between a 500 millisecond call to curl and a whole headless browser and a 5 second wait.
I hope you don't think I'm putting this too strongly, but this raised bar is a force in the monopolisation and de-diversification of the web.