At 8:21 PM -0500 5/30/03, Joe Laffey imposed structure on a stream of electrons, yielding:
On Fri, 30 May 2003, Bill Cole wrote:

 At 12:34 PM -0500 5/29/03, Joe Laffey  imposed structure on a stream
 of electrons, yielding:
 >
 >Get a unix box. (I like www.netbsd.org it will run on almost anything,
 >including old Macs, even 68k macs run dns fine)

Or get an OS X machine.

Of course.



>Then use rbldns http://cr.yp.to/djbdns.html


 I also would strongly recommend against running djbdns or anything
 else written by Dan Bernstein. Aside from the overwhelming evidence
 that the man is a raving lunatic who answers support queries and
 feature requests with personal attacks, if you are concerned about
 the legal status of software you run you will have an interesting
 time figuring out just exactly what your license to any of his
 products really is. I know a lot of people who manage to deal with
 the unclear licensing and the effective total lack of support and
 even the paranoid delusional conspiracy theories (as you read through
 the djbdns site, keep that phrase in mind when Dan starts discussing
 BIND) but I would not recommend trying.


Heh. I just migrated to djbdns. The license makes sense to me.

What license?


I'm absolutely serious. If you can point me at a license for any piece of djbdns I'd appreciate the help. There is nothing I can find in the distribution or on the website.

The product
is far simpler and easier in my mind than BIND. It also has an excellent
security record, unlike BIND, which is one of the most notorious apps for
being insecure...

History certainly shows a lot of BIND security problems, largely because BIND has always been the testbed for new DNS technologies, as well as simply because it has a history embedded in code.



BIND also has a terrible caching strategy, where is purges the most recent
queries first, making the cache worthelss when it is full. I moved to
djbdns for performance, and I have realized it in faster queries, and less
problems under load.

I honestly am not familiar with that caching issue, although I am familiar with some of the debate between DJB and people he perceives as enemies and am not impressed with the truth level of that discourse. It seems to me that what I do know of the BIND caching strategy is that it honors TTL (which puts it ahead of most Windows systems) so it would never really reach a point where the cache really become stale no matter how it is managed I know that different people seem to get different performance results when trying to compare the way BIND and djbdns perform, and I'm not interested enough to try to figure out why.


DJB is opinionated, but that doesn't affect the quality of the software he
writes.

That is only true if you append "in ways that most people care about." The use of DJB-ware pretty well eliminates any capacity to migrate to IPv6 as a direct result of Bernstein's childish and paranoid disagreements with the ways IPv6 standards ended up.


 Open Source authors can hardly be expected to provide full support
for free. Sure I try to with my OSS projects, but mine are very small in
comparison.

DJB-ware does not qualify as 'Open Source' by the usual definitions. I have never been actually able to find a license (there is none in the djbdns package and none on the djbdns website) so it's rather hard to pass judgement on the license. DJB has made legal threats against people distributing modified versions of his software, so it is unclear where he draws the line on his ownership of it.
--
Bill Cole
[EMAIL PROTECTED]



############################################################# This message is sent to you because you are subscribed to the mailing list <[EMAIL PROTECTED]>. To unsubscribe, E-mail to: <[EMAIL PROTECTED]> To switch to the DIGEST mode, E-mail to <[EMAIL PROTECTED]> To switch to the INDEX mode, E-mail to <[EMAIL PROTECTED]> Send administrative queries to <[EMAIL PROTECTED]>



Reply via email to