diff --git a/builtins/nums.h b/builtins/nums.h
index 5b1b7c8..ce41c08 100644
--- a/builtins/nums.h
+++ b/builtins/nums.h
@@ -20,22 +20,7 @@ bool Num__isinf(double n);
 bool Num__finite(double n);
 bool Num__isnan(double n);
 double Num__nan(CORD tag);
-// Constants:
-#define C(name) const double Num__##name = M_##name;
-C(2_SQRTPI) C(E) C(PI_2) C(2_PI) C(1_PI) C(LN10) C(LN2) C(LOG2E) C(PI) C(PI_4) C(SQRT2) C(SQRT1_2)
-const double Num__INF = INFINITY, Num__TAU = 2.*M_PI;
-#undef C
 double Num__random(void);
-#define F(name) double (*Num__##name)(double n) = name;
-double (*Num__abs)(double) = fabs;
-F(acos) F(acosh) F(asin) F(asinh) F(atan) F(atanh) F(cbrt) F(ceil) F(cos) F(cosh) F(erf) F(erfc)
-F(exp) F(exp2) F(expm1) F(floor) F(j0) F(j1) F(log) F(log10) F(log1p) F(log2) F(logb)
-F(rint) F(round) F(significand) F(sin) F(sinh) F(sqrt)
-F(tan) F(tanh) F(tgamma) F(trunc) F(y0) F(y1)
-#undef F
-#define F(name) double (*Num__##name)(double x, double y) = name;
-F(atan2) F(copysign) F(fdim) F(hypot) F(nextafter) F(pow) F(remainder)
-#undef F
 extern const TypeInfo Num;
 
 CORD Num32__as_text(const float *f, bool colorize, const TypeInfo *type);
@@ -48,23 +33,8 @@ float Num32__mod(float num, float modulus);
 bool Num32__isinf(float n);
 bool Num32__finite(float n);
 bool Num32__isnan(float n);
-// Constants:
-#define C(name) const float Num32__##name = M_##name;
-C(2_SQRTPI) C(E) C(PI_2) C(2_PI) C(1_PI) C(LN10) C(LN2) C(LOG2E) C(PI) C(PI_4) C(SQRT2) C(SQRT1_2)
-const float Num32__INF = INFINITY, Num32__TAU = 2.*M_PI;
-#undef C
 float Num32__random(void);
 float Num32__nan(CORD tag);
-#define F(name) float (*Num32__##name)(float n) = name##f;
-float (*Num32__abs)(float) = fabsf;
-F(acos) F(acosh) F(asin) F(asinh) F(atan) F(atanh) F(cbrt) F(ceil) F(cos) F(cosh) F(erf) F(erfc)
-F(exp) F(exp2) F(expm1) F(floor) F(j0) F(j1) F(log) F(log10) F(log1p) F(log2) F(logb)
-F(rint) F(round) F(significand) F(sin) F(sinh) F(sqrt)
-F(tan) F(tanh) F(tgamma) F(trunc) F(y0) F(y1)
-#undef F
-#define F(name) float (*Num32__##name)(float x, float y) = name##f;
-F(atan2) F(copysign) F(fdim) F(hypot) F(nextafter) F(pow) F(remainder)
-#undef F
 extern const TypeInfo Num32;
 
 // vim: ts=4 sw=0 et cino=L2,l1,(0,W4,m1,\:0
diff --git a/compile.c b/compile.c
index a658624..0c7c050 100644
--- a/compile.c
+++ b/compile.c
@@ -1242,7 +1242,7 @@ module_code_t compile_file(ast_t *ast)
             env->code->typedefs, "\n",
             env->code->typecode, "\n",
             env->code->fndefs, "\n",
-            "void use$", module_name, "(void);\n"
+            "public void use(void);\n"
         ),
         .c_file=CORD_all(
             // CORD_asprintf("#line 0 %r\n", Text__quoted(ast->file->filename, false)),
@@ -1250,7 +1250,7 @@ module_code_t compile_file(ast_t *ast)
             env->code->funcs, "\n",
             env->code->typeinfos, "\n",
             "\n"
-            "void use$", module_name, "(void) {\n",
+            "public void use(void) {\n",
             env->code->main,
             "}\n"
         ),
diff --git a/environment.c b/environment.c
index be04f0a..55041f3 100644
--- a/environment.c
+++ b/environment.c
@@ -13,17 +13,6 @@ typedef struct {
     binding_t binding;
 } ns_entry_t;
 
-// static type_t *namespace_type(const char *name, table_t *ns)
-// {
-//     arg_t *fields = NULL;
-//     for (int64_t i = Table_length(ns); i >= 1; i--) {
-//         struct {const char *name; binding_t *binding; } *entry = Table_entry(ns, i);
-//         fields = new(arg_t, .next=fields, .name=entry->name, .type=entry->binding->type);
-//     }
-//     name = heap_strf("%s_namespace", name);
-//     return Type(StructType, .name=name, .fields=fields);
-// }
-
 env_t *new_compilation_unit(void)
 {
     env_t *env = new(env_t);
@@ -100,9 +89,9 @@ env_t *new_compilation_unit(void)
             {"min", "Int8__min", "Int8"},
             {"max", "Int8__max", "Int8"},
         )},
-#define C(name) {#name, "Num__"#name, "Num"}
-#define F(name) {#name, "Num__"#name, "func(n:Num)->Num"}
-#define F2(name) {#name, "Num__"#name, "func(x:Num, y:Num)->Num"}
+#define C(name) {#name, "M_"#name, "Num"}
+#define F(name) {#name, #name, "func(n:Num)->Num"}
+#define F2(name) {#name, #name, "func(x:Num, y:Num)->Num"}
         {"Num", Type(NumType, .bits=64), "Num_t", "Num", $TypedArray(ns_entry_t,
             {"near", "Num__near", "func(x:Num, y:Num, ratio=1e-9, min_epsilon=1e-9)->Bool"},
             {"format", "Num__format", "func(n:Num, precision=0)->Text"},
@@ -112,9 +101,12 @@ env_t *new_compilation_unit(void)
             {"isfinite", "Num__isfinite", "func(n:Num)->Bool"},
             {"isnan", "Num__isnan", "func(n:Num)->Bool"},
             C(2_SQRTPI), C(E), C(PI_2), C(2_PI), C(1_PI), C(LN10), C(LN2), C(LOG2E),
-            C(PI), C(PI_4), C(SQRT2), C(SQRT1_2), C(INF), C(TAU),
+            C(PI), C(PI_4), C(SQRT2), C(SQRT1_2),
+            {"INF", "INFINITY", "Num"},
+            {"TAU", "(2.*M_PI)", "Num"},
             {"random", "Num__random", "func()->Num"},
-            F(abs), F(acos), F(acosh), F(asin), F(asinh), F(atan), F(atanh), F(cbrt), F(ceil), F(cos), F(cosh), F(erf), F(erfc),
+            {"abs", "fabs", "func(n:Num)->Num"},
+            F(acos), F(acosh), F(asin), F(asinh), F(atan), F(atanh), F(cbrt), F(ceil), F(cos), F(cosh), F(erf), F(erfc),
             F(exp), F(exp2), F(expm1), F(floor), F(j0), F(j1), F(log), F(log10), F(log1p), F(log2), F(logb),
             F(rint), F(round), F(significand), F(sin), F(sinh), F(sqrt),
             F(tan), F(tanh), F(tgamma), F(trunc), F(y0), F(y1),
@@ -123,9 +115,9 @@ env_t *new_compilation_unit(void)
 #undef F2
 #undef F
 #undef C
-#define C(name) {#name, "Num32__"#name, "Num32"}
-#define F(name) {#name, "Num32__"#name, "func(n:Num32)->Num32"}
-#define F2(name) {#name, "Num32__"#name, "func(x:Num32, y:Num32)->Num32"}
+#define C(name) {#name, "(Num32_t)(M_"#name")", "Num32"}
+#define F(name) {#name, #name"f", "func(n:Num32)->Num32"}
+#define F2(name) {#name, #name"f", "func(x:Num32, y:Num32)->Num32"}
         {"Num32", Type(NumType, .bits=32), "Num32_t", "Num32", $TypedArray(ns_entry_t,
             {"near", "Num32__near", "func(x:Num32, y:Num32, ratio=1e-9f32, min_epsilon=1e-9f32)->Bool"},
             {"format", "Num32__format", "func(n:Num32, precision=0)->Text"},
@@ -135,9 +127,12 @@ env_t *new_compilation_unit(void)
             {"isfinite", "Num32__isfinite", "func(n:Num32)->Bool"},
             {"isnan", "Num32__isnan", "func(n:Num32)->Bool"},
             C(2_SQRTPI), C(E), C(PI_2), C(2_PI), C(1_PI), C(LN10), C(LN2), C(LOG2E),
-            C(PI), C(PI_4), C(SQRT2), C(SQRT1_2), C(INF), C(TAU),
+            C(PI), C(PI_4), C(SQRT2), C(SQRT1_2),
+            {"INF", "(Num32_t)(INFINITY)", "Num32"},
+            {"TAU", "(Num32_t)(2.f*M_PI)", "Num32"},
             {"random", "Num32__random", "func()->Num32"},
-            F(abs), F(acos), F(acosh), F(asin), F(asinh), F(atan), F(atanh), F(cbrt), F(ceil), F(cos), F(cosh), F(erf), F(erfc),
+            {"abs", "fabsf", "func(n:Num32)->Num32"},
+            F(acos), F(acosh), F(asin), F(asinh), F(atan), F(atanh), F(cbrt), F(ceil), F(cos), F(cosh), F(erf), F(erfc),
             F(exp), F(exp2), F(expm1), F(floor), F(j0), F(j1), F(log), F(log10), F(log1p), F(log2), F(logb),
             F(rint), F(round), F(significand), F(sin), F(sinh), F(sqrt),
             F(tan), F(tanh), F(tgamma), F(trunc), F(y0), F(y1),
diff --git a/tomo.c b/tomo.c
index c2332e6..1e55c85 100644
--- a/tomo.c
+++ b/tomo.c
@@ -126,7 +126,7 @@ int main(int argc, char *argv[])
             "(void)argv;\n"
             "GC_INIT();\n"
             "detect_color();\n"
-            "use$", module.module_name, "();\n"
+            "use();\n"
             "return 0;\n"
             "}\n"
         );