• From: "Jeremy H. Griffith" <jeremy@o...>
• To: <xml-dev@l...>
• Date: Fri, 20 Jan 2012 17:13:31 -0500

On Fri, 20 Jan 2012 07:02:36 -0500, "David Lee" <dlee@c...> wrote:

>* Was there a natural evolution from tagging little bits of text to the
>point where it was felt that all of it should be tagged ?  Or perhaps there
>were cross-purposes from different camps ?

Yes, there was.  The first markup scheme I worked with
was for typesetting using TTS encoding (6 bit), on
Mergenthaler typesetters like the 202 in 1970.  The
medium was 6-level punched paper tape.

With only 64 characters, you had 26 letters, 10 digits,
a small assortment of symbols (period, comma, etc.),
and a few codes that changed state: shift, unshift, 
supershift.  And for Merg, there was one escape code,
bell, that introduced control sequences.  

So what you had was text with control sequences 
interspersed, such as font changes, size changes, 
column width, left and right indents, and alignment.

There were no start/end tags; you remained in the
new state until you changed it explicitly.  So you
had to start your tapes with a set of defaults, so
your job didn't get the properties of whatever had
been run just before it.

I was familiar with this because I wrote the first
text editor program that ran on micros that produced
the coding needed for the 202 (and later, the L300).
Until then, you used minicomputer front-end systems,
which were quite pricey.  ;-)

Early word-processor encoding was similar.  IIRC,
the concept of matching start/end tags arrived 
with SGML...

-- Jeremy H. Griffith, at Omni Systems Inc.
  <jeremy@o...>  http://www.omsys.com/

• References:
  • Text Markup Part II
    • From: "David Lee" <dlee@c...>