| HTML is perhaps a text presentation system. Typesetting
| implies that one has more control than HTML allows.
True. One of the points for PDF is that it allows a lot
of typesetting control, but as it is a *lobotomized*
PostScript, that is expected. OTH, that is precisely what
many commercial users of the WWW want in order
to competitively differentiate their pages. PDF is essentially
a PostScript viewer with some objects added for hypertext.
HTML is a better language for reuse and longer life
cycle documentation. It isn't optimum, but it wasn't
built to be all things to all systems. It is improving.
Fully-capable SGML systems are better for
high end, complex applications, but HTML is likely
to remain the staple of the WWW for text applications
in which formatting isn't the dominant requirement
simply because it is easy to learn and apply. Not
everyone wants or needs to design data description
languages and schemata.
| Can anyone give a quick description of the text encodings
| possible in Adobe's PDF? That seems to be another
| system gaining widespread acceptance.
The PDF spec is published by Adobe through
Addison-Wesley (ISBN 0-201-62628-4). As part of a
PDF font resource (a dictionary), there is an optional
field that specifies the character encoding of the font.
Its value may be an encoding resource or the name of a
predefined encoding. If no value is specified, the font's
built in encoding is used. The predefined encodings
as of the pub I have (june 1993) lists MacRomanEncoding,
MacExpertEncoding, WinAnsiEncoding, StandardEncoding
(???) and PDFDocEncoding. More could have been added.
PDF has been the center of some controversy not because
of its competition with HTML, but because of a project
at the National Institute for Standards and Technology
(NIST) to provide a Federal Information Processing Standard
(FIPS) for it, and the perception by some that NIST
seeks to endorse PDF over other notations
for hypermedia thereby using Federal policy to enforce
an unfair competitive advantage for Adobe. This is denied
by NIST and Adobe, and, IMO, is probably not the case.
I do not endorse PDF or condemn HTML. I note the
market pressures on both which should be considered
by the VRML designers for planning long and short
range enhancements to the language specification.
Len Bullard