Re: [DNSOP] Current DNS standards, drafts & charter

[Martin Hoffmann <[e-mail]>]

bert hubert wrote:
> 
> I've been looking at the amount of DNS out there, and I think we can
> do several things with them. I've also concluded that the mediocrity
> of DNS implementations outside of the well-known ones can not be
> fully blamed on "stupid programmers". The fact that we've offered the
> world 1000-2000 pages to read, with no guideline where to start, is
> also very likely to have contributed.

The thing that trips me up implementing DNS without intimate
knowledge of the assorted RFCs isn't the amount of them -- 2000
pages isn't all that bad if you come from SIP. Rather, it is that
important details are hidden all over the document base, often
in unexpected places. As an example, the SOA minumum field is
(re)defined in RFC 2308 which nominally talks about negative
caching, something I'd probably ignore if I implement a
simple authoritative server.

So, I'll step on that mine: What really would help new
implementers is a 1034bis.

Kind regards,
Martin