questions from using lint

[Barry Shein]

>From: gwyn at BRL.ARPA (VLD/VMB)
>Re: "lint" not being as smart as Root Boy Jim
>
>That's really silly.  I write C code such that I expect absolutely
>NO warnings from "lint"...

Although I agree philosophically, this reminds me a lot of the days
when I prided myself on being the only person I knew who could write
PL/I code without even an informatory message (and then get to harangue
other programmers when they modified my code and left "I" messages..argh!)

I started to wonder after a while how much I was actually accomplishing
much more than simulating the damn error checker in my brain rather
than writing such wonderful code all the time (meaning, whether it really
proved anything that I could keep PLIOPT silent.)

The point is, that some of this is really to an extent more a display
of the prowess of the programmer's ability to simulate compilers and
checkers somewhat (not that that isn't useful, but I wonder if it's
not a little distracting sometimes.)

At any rate, it's something to strive for, I guess I just react to
anything that sounds a little dogmatic in either direction. I would
put it more like this: IF you worked for me, I may very well run
code you hand me through lint, ya better be ready to answer for
any messages, so you may as well run it through yourself.

I wish root-boy could be a little more specific, rather than dismissing
his claim as 'silly' I would be open to the idea that LINT needs some
more design work, 10 years ago we were more open to its foibles than
we should be today, us old-timers get too complacent.

A more interesting idea: Would it be reasonable to run a vendor's
code you just bought through LINT and call their warranty dept if
there was any serious bitching by lint? Just a thought.

	-Barry Shein, Boston University