-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Randy MacDonald <[EMAIL PROTECTED]> writes:

> (with apologies for the drift towards chat...)
> Hello Bill;
>
> Is this posting potentially significant? I just saw a citation here:
>
>    http://tech.groups.yahoo.com/group/extremeprogramming/message/134835

Randy,

You raise a good point.  I know J is different, and attempts to force J
into the mold of another language seem to go awry, but I wonder if
there's a standard mode of thinking in J that seems most helpful.

Here are two of my habits:

  The XP idea (IIRC) of creating and maintaining test cases seems
  worthwhile here, too; I increasingly put examples in the code for
  scriptdoc which I can use as test cases, although I rarely write
  full-fledged regression tests.

  Depending upon how much J and how much of the problem I've got in my
  head when I start work, I may write small tacit verbs and combine them
  in a bottoms-up sense; when I've got something that works, I may fix a
  composition of those small verbs to get one still manageable verb that
  does what I want and avoids name proliferation.  

Feedback and other comments are welcome.  (Perhaps this is tending
towards J Programming a bit now?)

Bill
- -- 
Bill Harris                      http://facilitatedsystems.com/weblog/
Facilitated Systems                              Everett, WA 98208 USA
http://facilitatedsystems.com/                  phone: +1 425 337-5541
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (MingW32)
Comment: For more information, see http://www.gnupg.org

iD8DBQFG1LrC3J3HaQTDvd8RArfnAJwOKUV2K/KmbIombVZhzvj0V8GtcgCeKXL5
Rfi3zLLrPp6GQ04l9qdcK3U=
=MeW9
-----END PGP SIGNATURE-----

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to