Re: [Lit.] Buffer overruns
From: infobahn (infobahn_at_btinternet.com)
Date: 02/01/05
- Next message: infobahn: "Re: [Lit.] Buffer overruns"
- Previous message: infobahn: "Re: [Lit.] Buffer overruns"
- In reply to: Larry Elmore: "Re: [Lit.] Buffer overruns"
- Next in thread: rpl: "Re: [Lit.] Buffer overruns"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: Tue, 1 Feb 2005 08:09:25 +0000 (UTC)
Larry Elmore wrote:
>
> infobahn wrote:
> > Paul Rubin wrote:
> >
> >>infobahn <infobahn@btinternet.com> writes:
> >>
> >>>I've tried hard to understand how you could possibly need ABC in
> >>>production,
> >>
> >>As Kernighan and Plauger put it in "software tools", enabling the
> >>safety checks for debugging and disabling them for production is like
> >>wearing a parachute on the ground but taking it off once you're in the
> >>air.
> >
> >
> > They're wrong. Yes, I know what I'm saying and I know who I'm saying
> > it about. But they're still wrong. The analogy is cute, but proof
> > by analogy is fraud (Stroustrup, if he wasn't quoting someone else).
> >
> > Here's another analogy for you, just as pertinent, just as relevant,
> > and just as fraudulent if offered as proof:
> >
> > Enabling the safety checks for debugging and disabling them for
> > production is like putting scaffolding up when you're building
> > a skyscraper and taking the scaffolding down again when you're
> > finished.
>
> No, more like building access hatches and maintenance tunnels and
> emergency evacuation paths into the skyscraper, and then when you're
> done, filling them all in with concrete.
See what I mean? You can pick your analogy to suit the point of
view you wish to espouse. So none of them are really any good.
The best an analogy can do is /illustrate/ a point of view. You
have graphically illustrated yours. :-)
But I view those access hatches and maintenance tunnels and so on
as analogous to normal error checking, normal and necessary bounds
checking by the program itself (PBC?), and so on. Nobody is
suggesting that we remove those.
- Next message: infobahn: "Re: [Lit.] Buffer overruns"
- Previous message: infobahn: "Re: [Lit.] Buffer overruns"
- In reply to: Larry Elmore: "Re: [Lit.] Buffer overruns"
- Next in thread: rpl: "Re: [Lit.] Buffer overruns"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|