ietf-mxcomp
[Top] [All Lists]

Re: A 30% solution ; a 40% solution

2004-05-13 09:44:29

On 5/12/04 at 7:36 PM -0700, Matthew Elvey wrote:

What will this proposal break?/require adapt? forwarders? mailing lists? greeting card sites?

It will break or require changes for forwarders and possibly greeting card sites. (I've got in mind changes that can be made for them that will work, but there is no question that they will either have to make changes or they will break.) Mailing lists (if they rewrite the MAIL FROM) should be fine. We've talked some about ways to address these problems. Let's start another thread if we want to go into it in detail.

Does it keep spammers from complying? If so, how? If not, what *does* it accomplish?

Let me answer this backwards: The main thing that doing the check with the MAIL FROM domain is to deter joe-jobs. Secondarily, it can be used (like the HELO/EHLO domain) for the hook into accreditation. So, insofar as it prevents joe-jobs, spammers can't comply. Insofar as the accreditation part goes, spammers can comply, but hopefully will eventually become un-accredited.

What does it impact, how many systems does it impact? MUA? MSA? MTA? MDA? DNS?

I believe the main impact is in MDA (which has to perform the function) and DNS (which has to provide the data). I don't think this impacts MUA, MSA, or MTA. But I'm not sure exactly what you're asking. (For instance, it might impact MUA if we want to get an "unverified" message into the UI, but I'm not sure if you're thinking of things like that.)

At first blush, it seems that the 30% solution  breaks a lot

Are forwarders and greeting card sites "a lot"?

and doesn't suggest any workarounds.

*It* doesn't. But then again, I was only thinking of proposing semantics for the solution, not laying out all of the impacts. I'll try and write up a separate "workaround" message.

pr
--
Pete Resnick <http://www.qualcomm.com/~presnick/>
QUALCOMM Incorporated - Direct phone: (858)651-4478, Fax: (858)651-1102


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