3 files changed, 11 insertions(+), 11 deletions(-)

M 01-duskcc/04-wordsshell.md
M 01-duskcc/09-dusktillc.md
M 01-duskcc/10-beast.md

@@ 80,7 80,7 @@ process is:
 3. Stop the loop when a whitespace is encountered.
 
 You have a working loop! … for a certain definition of “working”. Being
-directly wired to they keyboard, without any line editing mechanism, makes the
+directly wired to the keyboard, without any line editing mechanism, makes the
 system very difficult to use. But for now, this will have to do because it will
 be a while before we add a line buffer.
 

@@ 1,4 1,4 @@
-# [Tumble Forth](/): The Unbearable Immediateness of Compiling
+# [Tumble Forth](/): From Dusk Till C
 
 We now know the [secret ingredients that make Forths alive][prev], but we won’t
 go all the way to a full Forth because it would be much less interesting than


@@ 66,7 66,7 @@ prompt.
 
 For this story arc, we’ll make the filesystem root our workspace. We’ll write
 our work in files at the root filesystem and then load them in Dusk. Let’s test
-that this work by creating a file called `myasm.fs` in the `fs/` subfolder in
+that this works by creating a file called `myasm.fs` in the `fs/` subfolder in
 Dusk’s source code, with this content:
 
     ." Hello Dusk!\n"


@@ 240,7 240,7 @@ First of all, let’s have useful constants:
 
 We’ll use these constants as arguments to our API. Now, the tricky part is to
 elegantly express the “use `EAX` as a direct destination and use `ESI` as an
-indirect source`”` command to the `add` instruction. Dusk’s i386 assembler has
+indirect source” command to the `add` instruction. Dusk’s i386 assembler has
 such an API, but the scope of this subject is too wide for this story arc.
 We’ll defer this to another story arc and go with a brutally simple, albeit
 ugly, API: specifying the form in the word name.


@@ 296,7 296,7 @@ compiler, when comes the time of generating the code.
 
 We won't go further, assembler-wise, in this story arc. If you wish to explore
 this subject further, I recommend you a [good tool to help you visualize x86
-encoding][x86enc].
+encoding][x86enc]. Then, you can of course take a look at Dusk's i386 assembler.
 
 ## Up next
 

@@ 63,7 63,7 @@ To avoid making words too heavy, we don't bother commenting the stack status at
 each step. However, Dusk has the habit of documenting stack status at particular
 points:
 
-1. Right after the beginning of the loop
+1. Right after the beginning of a loop.
 2. Sometimes at the end of a loop. Especially in cases where there is more than
    one exit point for the loop, for example in `begin..while..repeat`, in which
    exit points can have different stack status!


@@ 105,7 105,7 @@ struct, is provided automatically by structbinds, so for us, the effective word
 signature is `( -- c )`.
 
 The `:getc` word yields the character at current stream position and then
-advance this position by 1. If the end of the stream has been reached, -1 is
+advances this position by 1. If the end of the stream has been reached, -1 is
 returned. Therefore, we can reimplement (badly), the `:spit` method above thus:
 
     : myspit ( -- )


@@ 119,9 119,9 @@ you can do `0 file :seek`.
 
 ## Tokenizer API
 
-What we'll want to build is a word that, when called, yields either the next
-token, as a string, from `file`. If we've reached the end of the file, we yield
-zero. Let's call this word `nextt ( -- str-or-0 )` (for "next token").
+What we'll want to build is a word that, when called, yields the next token, as
+a string, from `file`. If we've reached the end of the file, we yield zero.
+Let's call this word `nextt ( -- str-or-0 )` (for "next token").
 
 Let's start roughly and yield our stream line by line, leveraging
 `IO :readline`:


@@ 147,7 147,7 @@ whitespace? Let's start a new `tok.fs` unit with this utility:
 
     : isWS? ( c -- f ) SPC <= ;
 
-`SPC` is a system constant for $20. In signatures, `f` means "flag" (0 or 1).
+`SPC` is a system constant for `$20`. In signatures, `f` means "flag" (0 or 1).
 
 Then, all we need is a buffer and a loop: