pdf and the semantic web

I would like to know how applicable could the PDF format be within the
context of the Semantic web? The PDF format is closed; annotating PDFs, as
in tagging not the file but the information within the file, is not possible
by means different from those provided by ADOBE. For instance, if I wanted
to tag a word, or an image within, inside, a PDF I would have to do it with
my acrobat reader -the latest version; But if I wanted to facilitate such
operation via WEB I could only do it if and only if I had the XSLT so I
could transform the PDF into XML. This limitation is, IMHO, a huge one
within the context of the semantic web where we should be able to define
links and use them. Furthermore, being forced to have a third party
application just for displaying a file that should be displayed directly by
the browser is not a nice feature. If PDF was open it could be rendered by
the browser.  Aren't closed formats such as PDF viable within the context of
the SW? After all the PDF was a solution within the context of portability
and exchange of information; the main problem it was solving was a simple
one "I want my document to look on display and once printed,  the same
everywhere" and "I want people to be able to read my documents without
loosing the format of the document and without having to consider the OS".
Isn't the PDF obsolete within this context?

-- 
Alexander Garcia
http://www.usefilm.com/photographer/75943.html