04/22/2018 (Sun) 14:12:46
>>16153>Gee I didn't know HTML support is heavier than PDF
That is why browsers are very heavy today. Although it is mostly JS part, but HTML+CSS adds complexity.
PDF is slow moving standard that may be messy, but already established and working. HTML now just moving without real standard (it is living (tm) standard), when features adding every time without revisions.
HTML/CSS also very loose, it basically permits non-closed tags and doesn't forbid wrong markup. For example, it is wrong to put block element inside inline, but site wouldn't broke if you'll do this, because if parsers would be strict, most of web would stop working. There were proposition in early 2000s to implement really strict HTML - XHTML, when parser can throw error on wrong markup, but it didn't work because almost every site will be broken and no browser want to lose market share that much.
CSS is also pretty complex, especially on version 3 with all animations, transitions and other things. Of course you don't need this for book reader, but no one forbids author to put some shady hack in his book html. In reality, book reader need to embed high grade web engine or use small self-made parser, but be prepared to get inconsistent results. As far as I know, even serious devices now couldn't properly display every epub ebook.
And, for example, calibre converter generates horrible mess instead of html for epub.