<< Previous Message Main Index Next Message >>
<< Previous Message in Thread This Month Next Message in Thread >>
Date   : Sun, 09 Jun 2013 16:16:44 +0100
From   : jgh@... (J.G.Harston)
Subject: Acorn newline sequence

Stephen Harris wrote:
>> Does anybody have any ideas why Acorn ended up with their newline
>> sequence of <10><13> instead of <13><10> as used everywhere else in 
>> the world?
>
> "everywhere" ?  Unix uses <13> and had done since the early 70s.

I specifically said "newline", not "end-of-line". Acorn's end-of-line 
sequence is a <10><13> newline. Unix's end-of-line sequence is a single 
<10>. Apple's end-of-line sequence is a single <13>. CP/M's and DOS's 
end-of-line sequence is a <13><10> newline.

> maybe Acorn picked it just for perversity ;-)

Or accident. :) Somebody coded OSASCI, OSNEWL and OSWRCH in an early 
System ROM and it's stayed that way around ever since.

Though, looking at the System/Atom MOS entries, it runs:

   OSRDCH     JMP (RDCHV)   ; Read a character
   OSECHO     JSR  OSRDCH   ; Read character and echo
   OSASCI     CMP  @#0D     ; Write ASCII sequence
              BNE  OSWRCH
   OSCRLF     LDA  @#0A     ; Write LF-CR Newline sequence
              JSR  OSWRCH
   OSWRCR     LDA  @#0D     ; Write CR
   OSWRCH     JMP (WRCVEC)  ; Write a character

So, having the newline sequence the other way around would mean that 
OSECHO couldn't just run on into OSASCI like that.

> It does, however, match close to what happens on a manual typewriter;
> a tap of lever does a linefeed; a push of the lever does a linefeed 
> and
> then starts to return the carriage to the start point.

With my manual typewriter when you push the newline lever across from 
the right to the left it moves the carriage across, and then at the 
maximum extent it engages the cogs to move the drum up one line. So, 
<cr><lf> (<13><10>).

-- 
J.G.Harston - jgh@...      - mdfs.net
<< Previous Message Main Index Next Message >>
<< Previous Message in Thread This Month Next Message in Thread >>