Mark Watson  <markw@markwatson.com> wrote:
+---------------
| I used to also do this occasionally in Lisp (e.g., in the 1980s, I
| prototyped all 12 of the SAIC ANSim neural network paradigms in Lisp,
| then rewrote everything in C for Windows 1.0).
| 
| Perhaps I just have odd work habits, but I was wondering if other people
| do the same sort of thing?
+---------------

Yup. Back when I was working for SGI, I routinely prototyped fragments
of Unix networking driver or firmware code in Scheme [note: I hadn't
quite yet converted to using CL back then] and then when the algorithm
was solid, hand-transliterated the code to C.  Example: A routine for
the firmware of an ATM card that used Bresenham's algorithm to populate
a cell-scheduling table -- the resulting C code even used a higher-order
function that accepted a (quasi-)closure as an argument! [I say "quasi-
closure", 'cuz although C doesn't have "real" closures you can usually
fake it well enough with the well-known hack of a callback function
pointer and an opaque cookie that's passed back to the callback.]


-Rob

-----
Rob Warnock, PP-ASEL-IA		<rpw3@rpw3.org>
627 26th Avenue			<URL:http://rpw3.org/>
San Mateo, CA 94403		(650)572-2607