[Mailman-Users] Carriage Return in Archives

[J C Lawrence]

On Tue, 19 Nov 2002 08:36:00 -0500
John DeCarlo <jdecarlo at mitre.org> wrote:
> J C Lawrence wrote:
>> On Fri, 15 Nov 2002 14:22:25 -0500 John DeCarlo wrote:

>> 1) Not everybody uses a windowing interface.

> Why would this be important?  I can turn on word wrap on a VT100.
> Automatic word wrap has been available in every kind of interface for
> decades.

And in the case of email, wisely not used.

>> 2) Not all text is prose

> So?

Mostly that's a comment on the semantics of reflowing being intensely
content sensitive.

> I guess I should have been more precise.  Any decent MUA over the past
> 10-20 years should not insert extra characters in the user's message.

The definition of "extra" is very subjective.  My tendency there would
be to state that the sending MUA should send the message *exactly* as
the author composed it, and that the receiving MUA should do its best to
present it in the same form.

> If the user wants to put in a line break after every word, or only
> after every 1000 word paragraph, that is the user's decision.

Absolutely.

> If the user wants to have a fancy looking poem with words centered,
> that should be a user decision.

Quite, and then the receiving MUA displays what was sent, as it was
sent.  Should the receiving MUA be configured to alter that default
presentation, then, again, that is the receiver's option, and he earns
any trouble and problem he thus incurs.

>> 3) Not all text can be reflowed without losing data.

> Exactly.  You want it reflowed when the MUA creates the email message.

No, I don't give a rat's arse about reflowing one way or the other.  I
want the author to compose the message as he sees fit, the sending MUA
to respect that, and the receiving MUA to default to honouring that
sender's decisions.

> I disagree.  Let the receiver decide how to view it.  Then the
> receiver can turn on word wrap at 30 characters or 72 or 185 or not
> every turn it on at all.

Text email predates (effectively) the use of markup languages, and thus
depends on tight coupling between sender and receiver as to format and
presentation.  Its a function of tacit agreements.  Richtext, HTML, etc,
don't suffer that particular limitation (but come with others).  You get
to pick.

>> Nope.  This argument was fought and quite properly lost almost 20
>> years ago.

> No.

We disagree.  I argued it at the time (and lost).

--
J C Lawrence
---------(*)                Satan, oscillate my metallic sonatas.
claw at kanga.nu               He lived as a devil, eh?
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.