Well, that was easy! Stack limit now working.

Makefile

@@ -44,7 +44,7 @@ test: $(TESTS) Makefile $(TARGET)
 
 .PHONY: clean
 clean:
-	$(RM) $(TARGET) $(OBJS) $(DEPS) $(SRC_DIRS)/*~ $(SRC_DIRS)/*/*~ $(TMP_DIR)/* *~ core
+	$(RM) $(TARGET) $(OBJS) $(DEPS) $(SRC_DIRS)/*~ $(SRC_DIRS)/*/*~ $(TMP_DIR)/* *~ core.*
 
 repl:
 	$(TARGET) -p 2> psse.log

docs/State-of-play.md

@@ -1,5 +1,16 @@
 # State of Play
 
+## 20260314
+
+When I put a debugger on it, the stack limit bug proved shallow. 
+
+I'm tempted to further exercise my debugging skills by having another go at 
+the bignum arithmetic problems.
+
+However, I've been rethinking the roadmap of the project, and written a long
+[blog post about it](https://www.journeyman.cc/blog/posts-output/2026-03-13-The-worlds-slowest-ever-rapid-prototype/). 
+This isn't a finalised decision yet, but it is something I'm thinking about.
+
 ## 20260311
 
 I've still been having trouble with runaway recursion — in `member`, but

lisp/documentation.lisp

@@ -12,7 +12,7 @@
             (item collection)
             "`(member item collection)`: Return `t` if this `item` is a member of this `collection`, else `nil`."
             (print (list "In member? item is " item "; collection is " collection))
-            ;; (println)
+            (println)
             (cond
               ((= 0 (count collection)) nil)
               ((= item (car collection)) t)

src/memory/stack.c

@@ -130,7 +130,7 @@ struct cons_pointer make_empty_frame( struct cons_pointer previous ) {
     uint32_t depth =
         ( nilp( previous ) ) ? 0 : ( get_stack_frame( previous ) )->depth + 1;
 
-    if ( stack_limit > 0 && stack_limit > depth ) {
+    if ( stack_limit == 0 || stack_limit > depth ) {
         result = in_make_empty_frame( previous, depth );
     } else {
         result =