On 7/19/2011 7:18 PM, John W Kennedy wrote:
On Jul 19, 2011, at 9:20 PM, Peter Constable wrote:
So you want to be able to discuss NBSP (say) in plain text. You can already do
that; in fact, you have multiple ways that everybody here will have no
difficulty understanding:
"NBSP"
"no-break space"
"U+00A0"
Creating a different character for SYMBOL FOR NBSP doesn't make communication
here any easier; in fact, it would lead to confusion as to whether you are, in
fact, meaning to refer to NBSP or to SYMBOL FOR NBSP.
But it's futile to argue that. People in the real world have been using such
conventions going back at least to the early 1960s, and, the last I heard,
Unicode is supposed to be used to encode the characters that people use.
If "people in the real world" are using symbolic notation, it's eligible
to be considered for encoding. *After* it's been established that such
usage exists on a scale to make it worthy of standardization. On some
level, it's futile to argue that, because that's so much of a long
established process. Whenever someone makes an assertion of the
existence, in the real world, of a particular notational convention, the
very next question will be "can you document this?". That's how the game
is played - and even in cursory discussion, people want to know whether
proof and documentation might be forthcoming, or whether it's just an
educated guess that some convention could exist.
Peter made a statement that could be interpreted in a way that he might
not have intended:
In contrast, nobody -- not even in the context of this discussion list --
needs to be able (e.g.) to send an email that contains in plain text a
character that depicts in a visible manner a character like NBSP or CGJ.
This could be read as a prescriptive policy, where somehow the encoding
committees define what users "need". I think it's meant in a slightly
different way: there are (several) established conventions for referring
to these characters without using symbols, so the absence of a symbolic
notation does not prevent discussion. That's clearly the case. On top of
that, these conventions define the most common usage.
Now, if, at some future time, it became common practice for people to
use a particular symbolic representation, then users who wish to follow
that new (and by assumption) common convention would "need" to use
symbols - and at that point, there might be a case to ask for
characters. Peter left that part unstated, I believe, because he's not
seen any indications that such convention has established itself, or is
about to establish itself, and hence that part is moot.
Whenever new notational conventions are developed, there's a period of
time where other means than standard characters have to be used to
express that. It could be images, private use characters etc. That goes
for mathematics as well as for other notations. The only regular
exception in Unicode are currency symbols - because once they are
introduced, they are guaranteed to be in very widespread use.
A./
