[l2h] Anything "new"?

[Bruce Miller]

Robin Fairbairns wrote:
> Jean-Pierre Chretien <Jean-Pierre.Chretien at cert.fr> wrote:
> 
>> In addition, pdf pollutes the web with bunches of non-hypertext
>> stuff.
> 
> what???  the pdf i generate has as much hypertext in it as any html i
>  generate.

Well, you're both sorta right...
You can certainly have hyperlinks within and
to the outside from pdf, but from outside,
only to the whole pdf, and not to points
within it.  
[It would seem that pdf could define fragment
anchors, in principle, but afaik it isn't yet
possible.]

> there's a valid use for pdf on the web, as much as there is for html.
>  the sole objection i know of is that it's a proprietary standard
> which changes in inscrutable ways, and at a tiresomely fast rate.

Oh, there are a couple of other objections:
Pdf integrates rather poorly (depending on what
you're after) into web browsing.  At best, you
step into another, nested, browser with a distinct
notion of "page", and separate navigation mechanisms.
[Again, it is conceivable that a pdf viewer
sufficiently integrated with a web browser could
mask a lot of this]

And, pdf documents have a rigid layout, they
don't reflow on resizing or reshaping.  This is,
of course, either a strength or weakness, depending
on what you're after.

To a certain extent, these points relate to pdf
being largely self-contained.  And these features
are either advantages or disadvantages for different
applications.

>> If you want to publish to be read on the web, use html.
> 
> or pdf, imo.

Again, you're both right.  There is valid uses of
pdf on the web, and there are valid reasons to
avoid it --- depending on what you're trying to do.

>> If you want to publish to be downloaded, printed and read on paper,
>>  publish in pdf.
>> 
>> With unique LaTeX (LyX?) source, you may in addition do both.
> _______________________________________________ latex2html mailing
> list latex2html at tug.org http://tug.org/mailman/listinfo/latex2html


-- 
bruce.miller at nist.gov
http://math.nist.gov/~BMiller/