On Mon, 20 Jan 2003, William Chops Westfield wrote:

*>For the same reasons that you don't need to worry so much about
*>efficiency on large mircocomputers, you also don't need to worry QUITE
*>so much about maintainability and readability on SMALL microcomputers.

*>I mean, how obscure can you get, being moderately careful, in less than
*>2K instructions?

Hehehe. I used to write spagetti assembly for 16C54 based on this
assumption. I had not managed to fill the chip (512 code words) before I
started losing track of things in the source between editing sessions
(days apart). Don't do it. I had to learn the hard way.

I also found that I wanted to reuse 80% of what I wrote 'just for that
project' 1-2 years later. Don't do spagetti code, it is a real real
pain.

Imho being moderately careful here is to be fascist about hungarian
notation for symbols, document each code change with 1 line of comment
and keep an outline of why you did things the way you did them in an
associated text file. Also note down any problems that happened while
testing or integrating it. Eg. if integration of module A with module B
breaks something write it down BOTH in module A's comments and in B's,
especially if you can't solve the problem right away.

Last, I think that version numbers are the best thing invented by mankind.
Keep a version number on everyting (like each module and subroutine) and
increment it every time you feel like it, but in any case with each and
every code change, at the latest at the end of a session. If you do not,
later you will have 3 copies of the code (say a backup, the previous
version or the version the text editor wants to recover after a crash or
irregular shutdown, and the last edited version), and you will start
wondering which one was the one that worked right.

Peter

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu