[Ach] PDF2HTML

[Alexander Wuerstlein]

On 14-05-20 15:53, Tobias Pape <Das.Linux at gmx.de> wrote:
> On 20.05.2014, at 15:36, Aaron Zauner <azet at azet.org> wrote:
> 
> > So,.. this would look like that:
> > http://www.mediafire.com/view/cdl67eb2d82hhva/applied-crypto-hardening.pdf
> > 
> > not too bad IMHO
> 
> Not too bad, yes.
> I have a bad gut feeling for the overall layout, however.
> PDFs made for booklike things (the way it is currently done for the ACH pdf)
> does not fit well with typical html conventions.
>   That said, this is one of the best pdf->html conversions I have seen.

Thats considered PDF->HTML? I thought that was some kind of ebook. There
are imho several problems with this format:
- Javascript and Cookies are required to view it. Since turning those to
  default-off is one of the better security advices concerning browsers
  I would consider this a huge problem.
- Internal hyperlinks do not work: Entries in the ToC are not clickable,
  etc.
- C&P does not work any better than in PDF: Cipher-Strings still contain
  spurious line-breaks. So this kind of HTML conversion buys us nothing.
- The layout does not reflow if I resize the window, which negates one
  of the more important features of HTML: output device independence.
- The not-really-HTML-but-Javascript-magic mess means that search
  engines won't be able to index that
- It also means that links into that document are impossible
- Search inside the document does not work, except for the part of the
  document that is currently displayed.

HTML conversions of that kind, while admittedly good-looking are imho
useless, PDF.js in Firefox or the PDF viewer in Chrome provide the same
functionality with less of a hassle for us with even better visual and
UI quality and features.



Ciao,

Alexander Wuerstlein.