curly-quotes

[Jonathan Lang]

Given perl6's use of unicode as a basis, could we get "curly quotes",
both single and double, to do the same things that straight quotes do?
That is: "text" does the same thing as "text", and 'text' does the
same thing as 'text'. Other than "looks neat", why do this? Because
curly-quotes come in matching sets, like parentheses and brackets do;
this lets you nest them.

(This seems so simple and obvious that I'll be surprised if someone
hasn't already proposed this; however, I don't recall seeing it
anywhere.)

--
Jonathan "Dataweaver" Lang

[Larry Wall]

On Sun, Apr 02, 2006 at 02:32:03AM -0800, Jonathan Lang wrote:
: Given perl6's use of unicode as a basis, could we get "curly quotes",

: both single and double, to do the same things that straight quotes do?

Depends on what "same thing" means, I suppose. Looks like you want
them directional, which is not exactly the same thing. An argument
could also be made for simply treating them all as the same character.
That being said...

: That is: "text" does the same thing as "text", and 'text' does the

: same thing as 'text'. Other than "looks neat", why do this? Because
: curly-quotes come in matching sets, like parentheses and brackets do;
: this lets you nest them.

Hmm, yes. That would be nice. But then, any number of bracketing
characters would be nice to steal from non-Latin-1 Unicode, and we're
trying to restrain ourselves for the moment. Because...

: (This seems so simple and obvious that I'll be surprised if someone

: hasn't already proposed this; however, I don't recall seeing it
: anywhere.)

Either your mailer or some MTA along the way seems to have silently and
oh-so-helpfully converted your pretty curly quotes to ASCII quotes.
And therein lies the (hopefully short-term) problem with getting
beyond Latin-1. Not that we can't support the higher characters, but
that the current infrastructure will *silently* introduce bugs when you
send such code around. It's even a little bit risky sending Latin-1,
but at least that tends to produce complete gobbldygook when it fails.

So the best we could do right now with the quotes would be to allow them
but to require you not to use them with semantics different from plain
ASCII double quotes.

Makes me wonder whether we can set up some automated way to at least
detect such accidental de-unicodification. Even just knowing the maximum
intended codepoint in the message might help detect most such errors.
Or even just knowing that *any* characters above 255 were intended...

Larry