ietf-mxcomp
[Top] [All Lists]

RE: DEPLOY: DNS Record Types

2004-08-26 11:02:59

On August 24, 2004, at 5:48 PM, Margaret Olson wrote (replying to my
request): 

1. Publishers MUST publish using the new SPF2 record type.
Change this to SHOULD

2. Publishers MAY also publish using TXT records.
Leave as is

3. Consumers MUST do lookups using the new SPF2 record type.
Change this to SHOULD

4. Consumers MAY also do lookups using TXT records.
Leave as is

5. Consumers MAY do both lookups (3 and 4) in parallel.
Leave as is

6. If consumers receive records from both lookups, they SHOULD use
    the SPF2 record and ignore the TXT record.
Change this to MUST, since we want the world to move to the SPF2
record

I'm OK with 1, 3, 5 and 6.  However, 2 and 4 need to at least be SHOULD
(they need to be as strong as the wording on 1 and 3).  My reasoning is
as follows:

You publish records primarily for the benefit of others. (I know there's
a secondary benefit to you, but that's not the point here.)  In order
for those consumers who can't see your SPF2 record to benefit, you'll
need to publish a TXT record.  Similarly, if you're a consumer, in order
to get the benefit from a publisher who can't publish an SPF2 record,
you'll need to query for the TXT record.

In spec language, the word MAY means truly optional.  But we don't want
truly optional here, we want people to understand that they should
publish and query for the TXT records in order to get this stuff off the
ground.

-- Jim Lyon



-- Jim Lyon
   mailto:JimLyon(_at_)Microsoft(_dot_)Com
   tel:+1-425-706-0867

Internet commerce will never really take off until you can buy something
online without getting spammed by the vendor.


-----Original Message-----
Here are my suggested changes, working from Jim Lyon's simplification. 
The goal is to both moving the world to the new resource record and to 
allow everyone to be compliant in the very short term.   I dislike the 
idea of having a spec with which most people in fact won't be able to 
comply. Coupled with the significant forgery/spam reduction incentive 
of implementing Sender ID the current wording ultimately works against 
the goals of the DNS folks by making a spec violation "standard 
practice".

1. Publishers MUST publish using the new SPF2 record type.
Change this to SHOULD

2. Publishers MAY also publish using TXT records.
Leave as is

3. Consumers MUST do lookups using the new SPF2 record type.
Change this to SHOULD

4. Consumers MAY also do lookups using TXT records.
Leave as is

5. Consumers MAY do both lookups (3 and 4) in parallel.
Leave as is

6. If consumers receive records from both lookups, they SHOULD use
    the SPF2 record and ignore the TXT record.
Change this to MUST, since we want the world to move to the SPF2 record

Margaret.



<Prev in Thread] Current Thread [Next in Thread>