diff --git a/02-stack-machine/README.md b/02-stack-machine/README.md
new file mode 100644
index 0000000..1441276
--- /dev/null
+++ b/02-stack-machine/README.md
@@ -0,0 +1,23 @@
+Style #2
+==============================
+
+Constraints:
+
+- Existence of an all-important data stack. All operations
+  (conditionals, arithmetic, etc.) are done over data on the stack
+
+- Existence of a heap for storing data that's needed for later
+  operations. The heap data can be associated with names
+  (i.e. variables). As said above, all operations are done over
+  data on the stack, so any heap data that needs to be operated upon
+  needs to be moved first to the stack and eventually back to the heap
+
+- Abstraction in the form of user-defined "procedures" (i.e. names
+  bound to a set of instructions), which may be called something else
+  entirely
+
+Possible names:
+
+- Go-Forth (as in the Forth programming language)
+- Stack machine
+
diff --git a/02-stack-machine/tf-02.py b/02-stack-machine/tf-02.py
new file mode 100644
index 0000000..5032d23
--- /dev/null
+++ b/02-stack-machine/tf-02.py
@@ -0,0 +1,116 @@
+#!/usr/bin/env python
+
+import sys, re, operator, string
+
+#
+# The all-important data stack
+#
+stack = []
+
+#
+# The new "words" of our program
+#
+def read_file():
+    """
+    Takes a path to a file and returns the entire
+    contents of the file as a string.
+    Path to file expected to be on the stack
+    """
+    path_to_file = stack.pop()
+    f = open(path_to_file)
+    # Push the result onto the stack
+    stack.append([f.read()])
+    f.close()
+
+def filter_chars():
+    """
+    Takes a string and returns a copy with all nonalphanumeric 
+    chars replaced by white space. The data is assumed to be on the stack.
+    """
+    str_data = stack.pop()
+    # This is not in style. RE is too high-level, but using it
+    # for doing this fast and short. 
+    stack.append(re.compile('[\W_]+'))
+    pattern = stack.pop()
+    # Push the result onto the stack
+    stack.append([pattern.sub(' ', str_data[0]).lower()])
+
+def scan():
+    """
+    Takes a string and scans for words, returning
+    a list of words. The data is assumed to be on the stack.
+    """
+    str_data = stack.pop()
+    # Push the result onto the stack
+    # Again, split() is too high-level for this style, but using it
+    # for doing this fast and short. Left as exercise.
+    stack.append(str_data[0].split())
+
+def remove_stop_words():
+    """ 
+    Takes a list of words and returns a copy with all stop 
+    words removed. The data is assumed to be on the stack.
+    """
+    word_list = stack.pop()
+    f = open('../stop_words.txt')
+    stack.append([f.read().split(',')])
+    f.close()
+    # add single-letter words
+    stack[0][0].extend(list(string.ascii_lowercase))
+    stop_words = stack.pop()[0]
+    # Again, this is too high-level for this style, but using it
+    # for doing this fast and short. Left as exercise.
+    stack.append([w for w in word_list if not w in stop_words])
+
+def frequencies():
+    """
+    Takes a list of words and returns a dictionary associating
+    words with frequencies of occurrence. The word list is assumed
+    to be on the stack.
+    """
+    word_list = stack.pop()
+    word_freqs = {}
+    i = len(word_list)
+    # A little flavour of the real Forth style here...
+    for wi in range(0, len(word_list)):
+        stack.append(word_list[wi]) # Push the word, stack[0]
+        # ... but the following line is not in style, because the naive implementation 
+        # would be too slow, or we'd need to implement faster, hash-based search
+        if stack[0] in word_freqs:
+            stack.append((word_freqs[stack[0]], word_freqs[stack[0]])) # (w, f) in stack[1]
+            stack[1] = (stack[0], stack[1][1] + 1) # Swap the tuple the stack with a new one
+            word_freqs[stack[-1][0]] = stack[-1][1]  # Load the updated freq back onto the heap
+        else:
+            stack.append((stack[0], 1)) # Push the tuple (w, 1)
+            word_freqs[stack[-1][0]] = stack[-1][1] # Load it back to the heap
+        stack.pop() # Pop (w, f)
+        stack.pop() # Pop word
+
+    # Push the result onto the stack
+    stack.append(word_freqs)
+
+def sort():
+    """
+    Takes a dictionary of words and their frequencies
+    and returns a list of pairs where the entries are
+    sorted by frequency 
+    """
+    word_freq = stack.pop()
+    # Not in style, left as exercise
+    return sorted(word_freq.iteritems(), key=operator.itemgetter(1), reverse=True)
+
+
+#
+# The main function
+#
+stack.append(sys.argv[1])
+read_file()
+filter_chars()
+scan()
+remove_stop_words()
+frequencies()
+word_freqs = sort()
+
+for tf in word_freqs[0:25]:
+    print tf[0], ' - ', tf[1]
+