On Fri, 23 Feb 2001, Peter Cavender wrote:
> It sounds like some supposed Linux supporters, rather than being
> flag carriers for new and better software, are merely geriatric Unix
> sysadmins who want to ride on the coattails of the movement, and refuse to
> aquaint themselves with new software, spreading FUD when they are
> threatened.
I could be wrong, here, but it strikes me that we are discussing two
different issues:
1) Whether djbdns is a better program than BIND, and
2) Whether djbdns' license and creator are arrogant.
It would appear, from a cursory glance, that the answer to both items is
"yes." The Generic Linux User might wish to embrace #1, but is put off
rather badly by #2: we like our coders to be self-deprecating, and we like
them to GPL their code. While this certainly doesn't change the code,
itself, it does change the perception of the code(er) a great deal.
Mind you, I'm speaking broadly, and there are always exceptions -- lots of
people will fiercely defend djb's right to say and license what and how he
wishes -- and, frankly, that's his perogative. It does, however, make
sure that his code won't be found on, say, the Debian release, etc. It
truly strikes me as odd that, given the putative superiority, he doesn't
just GPL it, and watch it become a de-facto standard, with him standing
over it much as Linus does the kernel. Hell: if he truly cared about
security, instead of ego, that's what he *should* do -- it would allow
folks much more ready access to a more secure alternative. Sure, it would
allow others to alter it to suit their needs, but he could give his "seal
of approval" explicitly to his release and those that he felt had earned
it, and let the rest do what they will. The GPL allows for a certain form
of darwinism that just doesn't exist when programs can't go head-to-head,
and his current licensing paradigm prevents this from occuring.
In the meantime, though, none of this appears to be about to change, and
my beating the poor ol' dead horse just ain't gonna make it any better.
*sigh*
$2e-2
-Ken
