Updated Whitespace FAQ (markdown)

Whitespace-FAQ.md

@@ -6,7 +6,7 @@ Virtually all programming languages use whitespace as part of their syntax. Many
 
 Most statements have only one line, and are followed by a line break. With languages whose lines are delimited by semicolons, the programmer pays a tax every line, just so that multiline statements can be used. But as presented below, Nimrod lets you use multiline statements most of the time w/o requiring this semicolon tax.
 
-Really long lines that must be broken are bad style, but often unavoidable. In those cases, Nimrod provides many implicit continuation . The rule of thumb is that after a `,`, unary or binary operator and `(` `[` `{` you can safely put a line break and continue the statement in the following line. If none of those things occurs naturally, you can always add a parenthesis. Two characters in those exceedingly rare situations, instead of one character every line.
+Really long lines that must be broken are bad style, but often unavoidable. In those cases, Nimrod provides many implicit continuation . The rule of thumb is that after a `,`, unary or binary operators and `(` `[` `{` you can safely put a line break and continue the statement in the following line. If none of those things occurs naturally, you can always add a parenthesis. Two characters in those exceedingly rare situations, instead of one character every line.
 
 After a continuation, the code can be positioned quite freely. The only rule is that the continuing line must be indented at least one level above the first line. 
 
@@ -18,6 +18,8 @@ someReallyLongProc(withMany, commands,
                    that, may, be, broken,
                    in many, lines)
 
+var c = long().method.()chains().
+  ofFunctionsWithlooongNames()
 
 var x = 2 +
         2
@@ -30,9 +32,11 @@ const DeBruijnNumbersTable: array[32, int8] =
 
 ```
 
+Nimrod also has semicolons. They are redundant as simple statement terminators, but useful if you want to fit multiple statements in one line. Statement lists can also be used  in some places where you would need to define a block using indentation
+
 # Indentation
 
-In 'curly braces languages', indentation works like comments. You will put it to make the code understandable, but its meaning is not enforced by the compiler/interpreter and thus can get out of sync with the real meaning, defined by the braces, or lack of it. This can lead to dangerously ambiguous code (for humans to understand, of course). For example (from ["Python: Myths about Indentation")](http://www.secnetix.de/~olli/Python/block_indentation.hawk):
+In 'curly braces languages', indentation works like comments. You will put it to make the code understandable, but its meaning is not enforced by the compiler/interpreter and thus can get out of sync with the real meaning, defined by the braces, or lack of it. This can lead to dangerously ambiguous code (for humans to understand, of course). For example (from ["Python: Myths about Indentation"](http://www.secnetix.de/~olli/Python/block_indentation.hawk)):
 
 >```C
 if (some condition)
@@ -45,14 +49,14 @@ else
 
 ### Tabs vs Spaces
 
-This is a non-issue in Nimrod, as only spaces are accepted as an indentation character --- a syntax error will be thrown if the compiler detects tabs being used for indentation (though this can be worked around by using a (link here w/ text "syntax filter"). However, languages that allow both of them to be mixed are dangerous. Some languages, like python 2 by default (fixed on python 3), tries to convert a tab to a certain number of spaces, and thus an indentation error introduced by mixing tabs and spaces will only be detected as a run-time failure. Other languages, that ignores any type of indentation, will generate code that behaves differently than it looks at first sight when using an editor with different tab-stops configuration. This can hide/introduce bugs, like in the previous C/C++ example. 
+This is a non-issue in Nimrod, as only spaces are accepted as an indentation character --- a syntax error will be thrown if the compiler detects tabs being used for indentation (though this can be worked around by using a [syntax filter](http://nimrod-lang.org/filters.html)). However, languages that allow both of them to be mixed are dangerous. Some languages, like python 2 by default (fixed on python 3), tries to convert a tab to a certain number of spaces, and thus an indentation error introduced by mixing tabs and spaces will only be detected as a run-time failure. Other languages, that ignores any type of indentation, will generate code that behaves differently than it looks at first sight when using an editor with different tab-stops configuration. This can hide/introduce bugs, like in the previous C/C++ example. 
 
 
 ### Sharing code
 
-Use dedicated source code paste sites and tags, that usually also gives you syntax highlighting and maybe even versioning and forking support. Having to read code on the internet w/o indentation sucks.
+Use dedicated source code paste sites and tags, that usually also gives you syntax highlighting and maybe even versioning and forking support. Having to read code on the internet without indentation sucks.
 
-\#{ and #} are also an option, with a script to reindent it.
+`#{` and `#}` are also an option, with a script to reindent it.
 
 ### Copy pasting
 
@@ -64,4 +68,6 @@ Many editors have an option to visibly show indentation markers, helping to visu
 
 # Strong spaces
 
-A experimental feature, currently only available through a source code filter, is strong spaces. The relative spacing between operators and operands overrides their standard precedence, like adding parenthesis would.
+A experimental feature, currently only available through a source code filter, is strong spaces. The relative spacing between operators and operands overrides their standard precedence, like adding parenthesis would.
+
+Only 0, 1, 2, 4 or 8 spaces are allowed and the compiler enforces that infix operators have the same amount of spaces before and after them. This rules does not apply when a newline follows after the operator, then only the preceding spaces are considered.