In <40B6BF2D(_dot_)5070100(_at_)ehsco(_dot_)com> "Eric A. Hall"
<ehall(_at_)ehsco(_dot_)com> writes:
On 5/27/2004 10:35 PM, wayne wrote:
I have provided data that shows that SPF is not a problem in practice.
You have provided nothing but handwaving.
You have provided data which demonstrates that there are already practical
occurances of overloaded TXT RRs, and that there will probably be more.
I found *zero* cases where SPF records were in a zone that needed to
fall back to TCP.
For the cases where there is already a lot of TXT data at the domain
level, much of can likely be removed. For example, I found one case
where someone used TXT to "comment" out a large part of their zone
file. The long rants about how the domain doesn't send email can
easily be replaced with something functional that says the same thing.
Your data is also out-dated, in that the SPF 'team' is already committing
to adopt XML syntaxes which will be significantly larger than the text
sizes you tested.
Remember that since the start around a year ago, the SPF group has
functioned very much like this working group. The SPF group doesn't
all agree any more than you and I always agree, even though we are
both part of MARID group.
That said, it is my understanding that the XML data will not be placed
in a TXT record at the domain level.
As to whether or not any of this is relevant, that's interpretation. I
think the argument speaks for itself -- if the next five designers are as
equally stubborn about using TXT as you are, then of course overlaod and
collision is unavoidable.
Actually, I gave evidence that if the next 5 was as carefully designed
as SPF records, there wouldn't be a problem.
What proof do you offer that they won't be
equally as stubborn in their demand to purposefully create confusion?
"purposefully create confusion"? whatever.
Anyway, I'm over the suggested 3 post-per-day limit, so I'll shut up
now.
-wayne