Re: [Asrg] Devilish: Forget about DNS

On 2/9/2004 10:21 AM, Hadmut Danisch sent forth electrons to convey:

> And, furthermore, analysis of all the comments I received for
>
>my RMX drafts and recent discussions showed that some domains
>would need to dynamically generate the records depending on the 
>query (see the hotmail example in 
>http://www.ietf.org/internet-drafts/draft-danisch-scaf-00.txt
>
>
>  
>
The hotmail example is bogus.  Here's why:
Huge mail service providers want people to accept mail from their users. 
In order to do this, they will have to maintain a good reputation.
They cannot do so and also provide free service to anyone and allow 
anyone to send mail "from" them.
How could they enforce anti-spam rules and allow virtually anyone to 
sign up to send email 'from' them?  Being a smarthost for their users 
makes it much easier for them to protect their reputation - e.g. they 
can better be proactive instead of reactive, use CAPTCHA, throttle, and 
can know for sure if their users sent spam.
Therefore, I don't see a need for dynamically generated records 
demonstrated.   Besides which, LMAP could allow a domain to specify a 
DNSRWL.  (A publicly queryable DNS server serving a whitelist of the IPs 
that are authorized to send from it, if dynamically generated records 
are needed for LMAP.)

>DNS is currently completely unable to provide dynamically generated
>replies. HTTP servers can easily do that (CGI).
>
>
>Again: I don't want to replace DNS in any way. I just don't want to 
>invent a new use for DNS which DNS can't do properly. That's all.
>  
>
Well-provisioned/current DNS can properly do what we want.  I think the 
switch from new (e.g. RMX) record types to the TXT record type is simply 
an effort to accommodate a tiny fraction of users not using good DNS 
software, in addition to the vast majority who would have been fine with 
a new record type.

IMO, DNS is a better fit to what we're looking to acomplish than HTTP.  
HTTP certainly has some advantages, as you've pointed out (e.g. https) 
but the disadvantages are large too.  IMO, we should stick to DNS 
(generally over UDP) using TXT.  It's far simpler and does the job.

>regards
>Hadmut
>  
>

>To: [email protected], [email protected], [email protected]
>
Crossposting BAD!  Don't do it, please!


Apropos IP:  Did you ever reply to the following?  Please do if you haven't:

esr wrote:
 > hd wrote:

>>The SPF draft violates the IETF's intellectual property rules
>> (http://www.ietf.org/IESG/Section10.txt).
>  
>
>
>I have read those rules and I do not see a violation, unless you
>are claiming proprietary rights over the concept of expressing
>LMAP information in a DNS zone file.  Are you doing so?
>
>  
>
Also eagerly waiting for Meng to publish his/her SPF as an I-D, as 
http://spf.pobox.com/rfcs.html perhaps falsely claims it has been. 
(It states that the next version will be 01, and links to the text of a 
00, which isn't on the ietf site.)
The "except that the right to produce derivative works is not granted." 
bit is unsettling.

I note it says :

>
>"12 Contributors and Acknowledgements
>
>  SPF is the child of RMX, by Hadmut Danisch, and DMP, by Gordon Fecyk.
>  It traces its ancestry farther back through "Repudiating Mail-From" by
>  Paul Vixie to a suggestion by Jim Miller in 1998.", etc.
>