1 files changed, 179 insertions, 0 deletions
diff --git a/documentation/specification/Syntax.md b/documentation/specification/Syntax.md
new file mode 100644
index 000000000..aa107d9cf
--- /dev/null
+++ b/documentation/specification/Syntax.md
@@ -0,0 +1,179 @@
+# Bit
+
+The "bit" is the Lux equivalent to the bool/boolean in other programming languages.
+
+There are 2 options:
+
+* `#0`, which is equivalent to `false`, in other programming languages.
+* `#1`, which is equivalent to `true`, in other programming languages.
+
+# Nat(ural)
+
+These are unsigned, 64-bit integers (that is to say: the positive numbers, and zero).
+
+They look like consecutive digits, potentially separated by commas.
+
+For example:
+
+* `0`
+* `123,456,789`
+* `123456789`
+* `1,23,456,789,`
+
+The only caveat is that the number must not start with a comma.
+
+# Int(eger)
+
+These are signed, 64-bit integers.
+
+They look like consecutive digits, potentially separated by commas, with a mandatory plus or minus sign at the beginning.
+
+For example:
+
+* `+0` or `-0`
+* `+123,456,789`
+* `-123456789`
+* `+1,23,456,789,`
+
+The only caveat is that the number must not start with a comma (after the sign).
+
+# Rev(olution)
+
+These are unsigned, 64-bit fractions.
+
+They look like consecutive digits, potentially separated by commas, with a mandatory period/dot at the beginning.
+
+They represent numbers between 0 and 1 (without including 1).
+
+For example:
+
+* `.0`
+* `.123,456,789`
+* `.123456789`
+* `.1,23,456,789,`
+
+The only caveat is that the number must not start with a comma (after the period/dot).
+
+# Frac(tion)
+
+These are signed, 64-bit floating-point numbers.
+
+They are what other languages call double-precision floating-point numbers.
+
+They look like an int(eger), immediately followed by a rev(olution).
+
+For example:
+
+* `+0.0`
+* `-123.456,789`
+* `+123.456789`
+* `-1.23,456,789,`
+
+The only caveat is that the number must not start with a comma (after the period/dot).
+
+# Text
+
+Text is the equivalent to _string_ in other programming languages.
+
+It is a delimited chunk of text, syntactically enclosed in double-quotes.
+
+Unlike in other programming languages, Lux text does not have escaping syntax for writing difficult-to-express characters within text.
+
+For example:
+
+* `"this is a simple example"`
+* `"here, the \n will not be replaced by a new-line, since there is no escape syntax"`
+* `"the backslash (i.e. \) stands for itself and serves no special function"`
+
+# Identifier
+
+Identifiers (called symbols in other lisp languages) are used as _names_ for functions, variables, etc.
+
+They look like sequences of characters with the following rules.
+
+* They cannot contain parentheses, brackets, braces, whitespace or double-quotes (i.e. `()`, or `[]`, or `{}`, or ` `, or `"`).
+* They cannot contain periods/dots or hash characters (i.e. `.`, or `#`).
+* They cannot begin with a digit, but may have digits throughout the rest of the identifier.
+* They can be written as 2 parts (separated by a single period/dot), or as a single part (which may begin with either 1 or 2 period/dots).
+
+For example:
+
+* `this-is.an-identifier`
+* `and_so_is_this_one`
+* `.ThisOneToo`
+* `..This4thOne,Also`
+
+# Tag
+
+Tags look similar to identifiers, and obey the same syntactic rules, with 1 additional rule: they must all begin with a single hash character.
+
+For example:
+
+* `#this-is.a-tag`
+* `#and_so_is_this_one`
+* `#.ThisOneToo`
+* `#..This4thOne,Also`
+
+# Form
+
+Forms are composite syntactic structures made up of a (possibly empty) sequence of syntax tokens, delimited by parentheses.
+
+Any valid syntax token may be inside a form, including other forms.
+
+For example:
+
+* `(this #is "a" valid.form)`
+* `(+ 123 456)`
+* `(def: my-constant +3.14)`
+
+# Tuple
+
+Tuples are composite syntactic structures made up of a (possibly empty) sequence of syntax tokens, delimited by brackets.
+
+Any valid syntax token may be inside a tuple, including other tuples.
+
+For example:
+
+* `[this #is "a" valid.form]`
+* `[+ 123 456]`
+* `[def: my-constant +3.14]`
+
+# Record
+
+Records are composite syntactic structures made up of a (possibly empty) sequence of syntax tokens **pairs**, delimited by braces.
+
+Any valid syntax token may be inside a record, including other records.
+
+For example:
+
+* `{this #is "a" valid.form}`
+* `{+ [123 456]}`
+* `{def: {my-constant +3.14}}`
+
+Because pairs of syntax tokens are expected, _the number of elements in a record must always be an **even** number_.
+
+# White space
+
+The only white space that the Lux parser recognizes as serving a function within the language are the normal space character (which separates syntax tokens within a line), and the new-line (which separates lines).
+
+Other white space characters serve no purpose within the language, and may even be used as part of identifiers and tags (although that is not advisable).
+
+The only exception to this rule is the carriage-return, which can show-up next to new-lines in text generated in some operating systems.
+
+The Lux parser can recognize the carriage-return, and processes it along its associated new-line, in order to accomodate files which contain it.
+
+# Comment
+
+Comments in Lux are single-line.
+
+They start with **2** hash signs, and they continue until the end of the line.
+
+For example:
+
+```
+## This is a comment.
+### So is this.
+
+#but-this-is-a-tag.so-don't-do-this!!!
+```
+