@Section
   @Title { Spaces and braces }
   @Tag { spaces }
@Begin
@PP
Every symbol in Lout either consists entirely of letters ({@Code "@"}
symbols. @Index { symbols, makeup of }
is considered to be a letter) or entirely of punctuation characters.  Here
are some examples of each type:
@ID @OneRow @Tab
    @Fmta { @Col @I    @CC A ! @Col @I    @CC B }
    @Fmtb { @Col @Code @CC A ! @Col @Code @CC B }
{
@Rowa A { From letters } B { From punctuation }
@Rowb A { "@PP" } B { "{" }
@Rowb A { "margin" } B { "}" }
}
Now if two symbols made from letters are run together like this:
@ID {
@Code "@CurveBox@I  Hello" &8ct @I (wrong!)
}
Lout will take this to mean one word or symbol called {@Code "@CurveBox@I"},
which is wrong.  In the same way, a letter-type symbol cannot be run
together with a word.  However, punctuation-type symbols can be run together
with anything.  For example, in
@ID @Code "@CurveBox{ Hello @I { world }}."
Lout understands that @Code "@CurveBox" and @Code "{"
# }
are separate, and it also sorts out
# {{
@Code "}}." into two right brace symbols and a full stop.  It might
seem strange to treat punctuation and letters so differently,
but computer programming languages have done it like this for
many years, and it works well.  This is the first use for
spaces. @Index { spaces, significance of }
spaces:  to separate letter-type symbols from each other and from words.
@PP
To see the second use for spaces, consider two words side by side:
@ID @Code "Hello world"
We want this to produce Hello world, so a space between two words in
the input must mean a space between them in the result.  Apply the
golden rule (where you can put one object, you can put any object) and
you get this:  @I { a space between two objects in the input produces
a space between them in the result }.  For example,
@ID @Code "@CurveBox Hello @CurveBox world"
produces
@ID { @CurveBox Hello @CurveBox world }
The space between the two objects @Code "@CurveBox Hello" and
@Code "@CurveBox world" appears between them in the result; the other
two spaces do not separate objects so do not appear in the result.
@PP
Two objects may be separated by a number of spaces other than one.  If
they are separated by no spaces, they will appear immediately adjacent
in the result; if separated by two spaces, they will appear two spaces
apart; and so on.  In English it is correct to leave two spaces between
the end of one sentence and the beginning of the next, for example.  See
Section {@NumberOf white} for two alternative ways to interpret white
space in Lout.
@PP
Occasionally the two uses for spaces conflict.  For example, to produce
@ID { { @CurveBox Hello }{ @CurveBox world } }
we need to have no spaces between the two objects, but then @Code "Hello"
and the following @Code "@CurveBox" would be run together, which will
not work.  The solution is to use braces:
@ID @Code "{ @CurveBox Hello }{ @CurveBox world }"
None of the six spaces in this example lie between two objects.
@PP
However, the main use of braces is to inform Lout that the object
within them is to be kept together, so that any nearby symbols are to
apply to all of it.  For example, leaving the braces out of
@Code "@I { Hello world }" would mean that @Code "@I" applies only to
{@Code "Hello"}.
@PP
When an object-consuming symbol like @Code "@I" is followed by an
braces. @RawIndex { braces }
braces.in.lout @SubIndex { in Lout text }
object enclosed in braces, that is the object consumed.  For example,
@ID @Code "This is @I { absolutely necessary }, since otherwise ..."
produces
@ID { This is @I { absolutely necessary }, since otherwise ... }
with the object @Code "absolutely necessary" italicized, but not the
following comma.  If there are no braces, the object consumed is
everything up to the next object-separating space:
@ID @Code "This is @I necessary, since otherwise ..."
produces
@ID { This is @I necessary, since otherwise ... }
with an undesirable italic comma.  In practice, this means you can
avoid braces only when italicizing a single word with no punctuation
attached.
@PP
One common pitfall is to use unnecessary braces, like this:
@ID {
@Code "@I { @CurveBox { Hello world } }" &8ct @I (bad!)
}
Another is to think that all spaces produce space in the result, and so
write
@ID {
@Code "@I{@CurveBox{Hello world}}" &8ct @I (worse!)
}
Use braces only when necessary, and add extra spaces where they do not
separate objects, and your documents will be far easier to read while
you are working on them.  Don't be fooled by the argument that says it
doesn't matter because it doesn't affect the final printed result.
@End @Section