> It's loose best effort markup with no good consensus on semantics. And PDF has good semantics? Are we still on the topic of how HTML is better than PDF, or…?

> It's loose best effort markup with no good consensus on semantics. And PDF has good semantics? Are we still on the topic of how HTML is better than PDF, or…? We're in the comments for a page that says that PDF tables are characters just floating in space, and people are saying most PDFs out there don't have semantic markup. Meanwhile HTML had semantics efforts for decades now, just choose your flavor. Blind people read HTML, you know. Do they read PDFs? > HTML with images is not good option for papers which have many equations. There's MathML for that, and IIRC other formats too. You could even have embedded TeX like Anki has. Use SVG for fallback. >> Have a static HTML version that's rendered the same in the future > We don't have that. Ooh, chicken-and-egg again? Freeze any of the versions from the past decade with the rendering standards, and you'll have it. But actually, it doesn't even matter, just like HTML 2.0 can be rendered fine on modern devices (aside from the different text size). Treat your paper as a paper instead of a webzine, don't use crazy layouts, just do “text, image, text” which you'll want anyway for the different displays—and your document will render fine in the future when it will be delivered straight to the retina, instead of making me scroll the PDF back and forth because no reflow.