Re: Fwd: Re: Lojban word processor for Windows?

[David Brookshire Conner <nellardo@concentric.net>]

Robert J. Chassell writes:
 > From: "Robert J. Chassell" <bob@rattlesnake.com>
 > 
 > David Brookshire Conner <nellardo@concentric.net> wrote:
[....]
 >    And typography, well, I'm a structured markup fiend. Mixing
 >    typography with word processing seems misguided and encourages lots
 >    of visually ugly documents...
 > 
 > I am confused here.  Suppose you are writing on cyrillic, Tibetan, 
 > and latin: do you use structured markup for the different fonts?  I
 > don't think so.  I suspect you use markup for whether your Tibetan or
 > Korean `Watch Out!' should be emphasized or not.

You don't use the structured markup to denote what *font* you are
using but you might very well use it to denote what *language* you are
using.  Whether or not you do this depends on exactly what kind of
doucment you are writing.

Suppose you are writing a novel that includes fluently multilingual
characters.  The *structure* may have more to do with who says what
than it does with what language someone is speaking.

Now suppose you are writing a textbook for learning a foreign
language. Here, clearly, marking the language can be quite important,
whether or not the languages in the book use the same glyphs or
not. For example, writing a textbook on Russian, sections will
describe Cyrillic, including the characters, but the language will be
English (well, it will if *I* write it :-)

 > Surely, structured markup is orthogonal to what glyphs used for
 > straight text?

Yes of course - I wasn't suggesting that.

Glyphs are not fonts. Unicode does not describe a font. Unicode
describes characters which have stereotypical appearances (the
glyph).  You need a font to render something - that's typography, not
word-processing.

Hmmm. I suppose I'm getting definitional here, so here's how I'm using 
things:

character - value in some sort of script

glyph - the archetypical appearance of a character; alt. the
particular appearance of a character represented by a particular font.

Word processing - rearranging characters (usually in groups, i.e.,
words :-)

(Structured) markup - notating the logical structure of a string of
characters.

Formatting - Mapping markup to particular renderings.

Typography - The subset of formatting concerned with fonts and
placement of glyphs (in sense 2 above).

Font - a set of graphical symbols with a mapping from symbol to
character. The map need not be complete, but is usually a function
(i.e., one graphical symbol is associated with one character. One
character may have many representations in the font).

 >    ... from taking a book from outline to camera-ready form...
 > 
 > Gosh, a voice from out of history. :-)  `Camera-ready' is only one kind
 > of output format.

Of course.

 > For the past couple of decades people I know have read
 > manuals both online and printed: books go from outline to *two* forms,
 > one them `camera-ready', the other `display' ready.  

Right - of course. This wasn't a manual. This was a textbook. Six
years ago (when I wrote it), Addison Wesley wasn't about to consider
distributing a textbook on CD ROM. Three years ago (when I was working 
on the revision), they were, and had that project continued, I would
have produced both camera ready and display ready copy, most likely
from one SGML source.

 > Oh, I know that.  The problem is and has been for some years strictly
 > legal:  the Lucid/X Emacs people are unable to obtain the kinds of
 > disclaimers/assignments that the lawyers I deal with require for wide
 > spread, safe distribution.

Ah, of course. Lawyers, gotta love em.


[...]

 > Most programmers I know rightfully hate these sorts of legal concern;
 > or else they pay little attention on account they lack experience and
 > street smarts.

I hope I'm the former, not the latter :-)

Brook

---------
A computer's attention span is as long as it's power cord.

---------
Fancy. Myth. Magic.
http://www.concentric.net/~nellardo/