[l2h] Suggesting SVG-format for expression images

[Ross Moore]

> On Wed, 11 Jul 2001, Ross Moore wrote:
> 
> > To implement it, we need availability of tools to both read and write
> > the format. 
> > 
> > Does Netpbm have a  pnmtosvg  tool ?
> 
> Converting bitmaps directly to SVG would likely not be advantageous - the
> original author had suggested that SVG's strong point is scalability, so a

Yes, of course. Being an Adobe format it is most likely based on PostScript
or PDF. For LaTeX2HTML, that means that the methods currently used with
the bitmapped graphics (within the pstoimg) script, will need to be replaced
by something completely different.
It may be sufficient to simply alter BoundingBox numbers, but it may require
more than this when there is supposed to be an opaque colored background.

Effects like Rotation and Reflection/Flips and Scaling will certainly need
more detailed edits.


> raster to vector tool like autotrace
> (http://sourceforge.net/projects/autotrace/)  might be of use. You could

Given that PostScript output is generated first, there should be no need
for any  raster to vector  conversion.
So the real question should be, 
  "Is there an SVG device for GhostScript ?"
If so, in which version was it first available ?


> also move a step back and use pstoedit (a "shareware" plugin for SVG
> output is available for pstoedit) to render directly from the PostScript
> version.

Indeed, that would be the correct strategy.

 
> > > Web-browser plug-ins is presently available for Windows and MacOS. 
> > > Unfortunately not for *nix:
> > 
> > Hmm; that's definitely a drawback.
> > It cuts down the audience for your web-pages.
> > Hopefully it's just a temporary thing for the major browsers.
> 
> There is a Java interpretor available via the Apache people, which runs
> outside a browser. (I haven't used this. I have used Amaya to view SVG,
> but it is not satisfactory as a browser.)

Unfortunately, this doesn't help at all for inline mathematics.  


> Ian.
> PS Personally, I would be reluctant to recommend SVG at this time,
> however, I have to admit that I have effectively no experience with it, so
> you'll have to take my comments with a large grain of salt.

Sure. 
I see no point in implementing it for LaTeX2HTML, except as an academic exercise,
until it has a large acceptance within Web browsers on all platforms.
But if someone wishes to do that exercise, then by all means do it,
and I'll help in whatever way is reasonable.


All the best,

	Ross Moore


> ----------------------------------------------------------------------------
> Ian MacPhedran,    Engineering Computer Centre,   2B13 Engineering Building,
> University of Saskatchewan,  57 Campus Drive,  Saskatoon SK  S7N 5A9, CANADA
> Phone: (306)966-4832 Fax: (306)966-5205  Email: Ian_MacPhedran@engr.USask.CA
>