On 10-01-13 22:55, Franck Martin wrote:
> What I (we?) don't want is that DMARC affects the reputation of the email
> as spammers will publish a DMARC record if it lowers their spamassassin
> score.

My idea was more along the lines:
- add some informational rules with near-zero scores that tell the
recipient about Dmarc policy existance and contents (see: DKIM_SIGNED)
- add some penalty rules when policy evaluation fails, or alignment fails.

While it does not improve deliverablity for poor-formatted senders (both
phish and legit), it would create a direct positive experience for
recipients: block more phish mail.

> 
> Also I think the RFC specifies that aggregate reports are an integral part
> of DMARC. You can't develop one without the other.

Noted.

> 
> The opendmarc milter/proxy way seems the way to go, spamassassin is
> installed this way on linux boxes, I don't see why it would not be as easy
> to install opendmarc.

I used Spamassassin just as an example, but let's stay on that route.
- It has already a huge install base, adding a plugin to something that
already exists is easier than integrating a new piece of software in
your mailstack.
- It has lots of ways to integrate other than a milter api.
- Not everyone runs a milter-capable MTA.

> 
> On 1/10/13 1:04 PM, "Tom Hendrikx" <[email protected]> wrote:
> 
>> Hi,
>>
>> I was pondering on using DMARC in a passive way: implementing a plugin
>> for spamassassin (or similar) that would only score/reject/quarantine
>> based on DMARC alignment and policy, and not handle all the complex
>> DMARC stuff (mainly database setup, ruf/rua sending).
>>
>> This would lower the implementation doorstep to test DMARC effectiveness
>> for interested receivers: such an implementation would need nothing but
>> access to DNS, making system administration (both setup and maintenance)
>> a breeze.
>>
>> I do see the risk that, while intended for evaluation, such a setup
>> would lower the incentive to switch to a full-blown setup such as the
>> existing milter solution later on. The specification document has some
>> notices pointing out the importance of the feedback loop (and I
>> understand and support them), which are blatantly ignored by the above.
>> But I still see an implementation benefit for small receiving sites,
>> whose reports would be less significant to senders (IMHO), as existing
>> large sites will also send them reports.
>>
>> My personal point of view: getting a DMARC sender (monitor) policy
>> running was quite easy for my personal domain, but I did not have time
>> to implement the receiving side yet, thus I don't get any 'benefits'
>> that prove to me (and f.i. my $workjob manager) that implementing DMARC
>> is actually useful. A lower doorstep would be nice.
>>
>> Note that I don't intend to cripple the feedback concept or DMARC in
>> general, but just interested in any opinions on this.
>>
>> Kind regards,
>>      Tom
>> _______________________________________________
>> dmarc-discuss mailing list
>> [email protected]
>> http://www.dmarc.org/mailman/listinfo/dmarc-discuss
>>
>> NOTE: Participating in this list means you agree to the DMARC Note Well
>> terms (http://www.dmarc.org/note_well.html)
> 
> 
> _______________________________________________
> dmarc-discuss mailing list
> [email protected]
> http://www.dmarc.org/mailman/listinfo/dmarc-discuss
> 
> NOTE: Participating in this list means you agree to the DMARC Note Well terms 
> (http://www.dmarc.org/note_well.html)
> 

_______________________________________________
dmarc-discuss mailing list
[email protected]
http://www.dmarc.org/mailman/listinfo/dmarc-discuss

NOTE: Participating in this list means you agree to the DMARC Note Well terms 
(http://www.dmarc.org/note_well.html)

Reply via email to