Why wrap @ 80?

John Macdonald john-Z7w/En0MP3xWk0Htik3J/w at public.gmane.org
Fri Apr 23 22:45:02 UTC 2004


On Fri, Apr 23, 2004 at 06:00:09PM -0400, gabriel wrote:
> On April 23, 2004 05:49 pm, James Knott wrote:
> > Noah John Gellner wrote:
> > > Why do we wrap text at 80 columns on TLUG, and elsewhere on the Net for
> > > that matter? I know that if your mailreader doesn't autowrap it can be
> > > unpleasant to read, but don't the vast majority of readers wrap? The
> > > advantage of not wrapping is that fwds, replies, and follow-ups don't
> > > get funky formatting due to the hard carriage returns.
> > >
> > Well, that would go back to the 80 column IBM punch cards.  The 80
> > column "standard" was then applied to printers and terminals.
> 
> sorry to rain on everyone's history less (as interesting as it is... i had no 
> idea this issue went so far back)  i have to ask.  when then are we going to 
> abandon this very, very old obsolete standard?  i understand that there's 
> still a great many users on this list and elsewhere using mutt and pine etc. 
> but why don't those programs compensate for 80+ character lines?  wouldn't 
> this help the rest of us get out of this nearly century-old restriction?

Well, wrapping text is sometimes wrong.

For example, if you include code that happens to
go beyond 80 char lines, wrapping it makes the code
harder to read, and possibly makes it broken if it
is extracted from the message and used.  (It will
definitely be broken in that it has different content,
and a different number of lines, so that it is
changed from the original.  That could cause patch
failures, needless version number changes, numbered
line references to mean different things for different
readers of the "same" code; and it will occasionally
be even more broken when a line break gets inserted
into a string or other place where white space changes
are signigifcant to the working of the program.)

It is a pity that the clever new mail programs that
decided to provide paragraphs chose to use a layout
that was incompatible with the existing de facto
standard.

Instead of assuming that all mail contains paragraphs
that can be safely wrapped, and using a single line
of potentially unbounded length for a paragraph;
they could have simply continued to use convenient
length lines with a blank line to separate paragraphs,
adding a message header to say that paragraph wrapping
will not damage the body.  That would have worked
cleanly without making the mail program have to guess
whether wrapping a long line is the right thing to
do *this time*.

Instead, many years later, we still have two camps who
are annoyed at each other (and no resolution in sight).

-- 
--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml





More information about the Legacy mailing list