Subject: Re: Fwd: Re: Microsoft: Closed source is more secure
From: Lynn Winebarger <>
Date: Sun, 6 May 2001 08:06:44 -0500

On Sunday 06 May 2001 07:11, Simon Cozens wrote:
> On Sun, May 06, 2001 at 05:52:04AM -0500, Lynn Winebarger wrote:
> >     Can't speak directly to djbdns (that's why there's all those "if
> > qmail is any indicator"'s up there).  I wouldn't doubt that bind code
> > would be harder to read, though.
> I'll call the bluff. Have you read qmail, djbdns or bind code? I wouldn't
> like to think you were talking about things you have no experience of.

    I read a small amount of qmail code, none of djbdns or bind.  I haven't made
any claims about djbdns or bind (except as above), and only that the qmail source
is "odd" (compared to most source I've read).  That did assume it was written consistently
in the same style.
    Of course, the issue wasn't whether qmail had merits, it was how much time I was
going to spend on evaluating it for fitness for my purposes.  I certainly haven't claimed
qmail is unreadable or of poor quality.  I did spend a fair amount of time reading the
documentation included with the package itself, including the "notes" file.  In other
words, I attempted to read enough to estimate whether a full reading would be worth
the effort and fit within the time I had/have for it.
    I generally don't read large quantities of source code unless I've found some reason
I must modify it (and have no less time-consuming, reasonable alternative - e.g. other
software).  The major exception to this is when it's for educational purposes (for which

DJB's code might well qualify).  Or for purely recreational purposes (but that's not
my employer's time).
    Were you under the impression I was comparing the quality of bind and djbdns source
code (or qmail vs sendmail source for that matter)?   I wasn't and haven't.  At worst
I expressed
my faith in Ian's ability to tell which source was easier to comprehend - given that
the one he 
evaluated as more difficult is many times larger and older.