Who are the "spacists"?

Marko Rauhamaa marko at pacujo.net
Mon Mar 20 14:58:03 EDT 2017

Steve D'Aprano <steve+python at pearwood.info>:

> On Tue, 21 Mar 2017 12:30 am, Marko Rauhamaa wrote:
>> Unicode was 25 years late to the game
> I don't understand that comment. Are you suggesting that the oldest
> convention wins? The Unix 8-column convention is older than Unicode, so it
> wins?

The 8-column convention is older, alive and well on Unix so it wins on

> Well... in that case, there are tab conventions that pre-date Unix, e.g.
> those used by COBOL.

COBOL is not highly relevant in the Unix world.

> But wait... before there was COBOL, there were *typewriters*.

Typewriters and teletypes are indeed behind the original intended
(pre-Unix) semantics of things like BS, HT, CR and LF. However, they
acquired modified meanings in Unix. For example, CR on input is
converted to an LF, and LF on output is converted to a CRLF.

> So there you go. Unix was 50+ years late to the game.

Unix is normative for Unix.

I seem to recall, though, that the same 8-column interpretation was
there with CP/M and MS-DOS. The practice must be older:

   In practice, settable tab stops were rather quickly replaced with
   fixed tab stops, de facto standardized at every multiple of 8
   characters horizontally, and every 6 lines vertically (typically one
   inch vertically). A printing program could easily send the necessary
   spaces or line feeds to move to any position wanted on a form, and
   this was far more reliable than the modal and non-standard methods of
   setting tab stops. Tab characters simply became a form of data

   <URL: https://en.wikipedia.org/wiki/Tab_key>


More information about the Python-list mailing list