by:

i was 12 years old when i wrote my first lines of html, since then a lot of things changed on the internet.
some pioneers like netscape stopped, technologies like javascript are now more important than ever...

but when i look at html i don't see much progress, one issue since decades is the
handling of malformed documents another issue is the vague semantics in html.
both changed slightly with html5, but by far not enough.  

in my opinion html needs a fundamental new concept, data and presentation should be strictly separated.
all data must be machine readable, schema.org is for sure a move into the right direction, but:

Example of how a book is taged with schema.org

<div itemscope itemtype="http://schema.org/Book">
  <span itemprop="name">The Catcher in the Rye</span>?
  by <span itemprop="author">J.D. Salinger</span>
  Here is the book's <a itemprop="url" href="http://en.wikipedia.org/wiki/The_Catcher_in_the_Rye">Wikipedia page</a>.
</div>

Instead of trying to add additional meta descriptions to html, i would love to see a new model data language.

Example of Model Data

<book>
    <title>The Catcher in the Rye</title>
    <author>J.D. Salinger</author>
</book>


As well a Presentation framework

<book>
    {title} by {author}
    Here is the book's <a href="http://en.wikipedia.org/wiki/The_Catcher_in_the_Rye">Wikipedia page</a>.
</book>

I would remove all presentation tags from html (like h1, span, strong, font ..) and leave it to css how data is displayed.

Also i would use header, menu, video, book, product tags instead of divs for anything.


Of cause tagging data will increase the cost of each document (size), on the other hand we all will gain websites which are machine readable and therefore much better searchable. 
What do you think, what should be changed   ?
Happy Hacking
Andreas