Re:PDF: Still unfit for human consumption, 20 years later
by kindofastrawman on Monday August 10, 2020 @02:02PM
(#24112125)
> JS is allowed to fetch new files dynamically, and there's no reliable way to look at a piece of code and automatically figure out what it's going to fetch when you run it.
You've diverged from the context and are no longer doing an apples-to-apples comparison. The things you're describing are all opt-in and amount to having to deal with an adversarial input. There's nothing inherent to the medium that requires those things.
In other words, a person publishing a PDF is already abstaining from certain things. (Namely, the sorts of things you're bringing up that would make for a pathological case.) If the person who publishes a PDF does a straightforward translation into a web page, then you end up with something that doesn't exhibit any of the downsides you're discussing.
Re:PDF: Still unfit for human consumption, 20 years later
by anoncake on Monday August 10, 2020 @10:24PM
(#24116590)
No, but the medium allows these things. And that's a problem.
Re:PDF: Still unfit for human consumption, 20 years later
by oconnor663 on Monday August 10, 2020 @03:00PM
(#24112766)
Good point, and also relevant user name :)
Re:PDF: Still unfit for human consumption, 20 years later
by kevincox on Monday August 10, 2020 @05:07PM
(#24114002)
No, most browsers will save the resources as well and rewrite the HTML to reference them. You can have problems with dynamically loaded things but I have found that it works very well in practice. I have had maybe one page that was significantly broken saving from Firefox over the years.
There are tools that try to fetch those links and update the HTML to point to the local copy. But those tools can only go so far. JS is allowed to fetch new files dynamically, and there's no reliable way to look at a piece of code and automatically figure out what it's going to fetch when you run it.