Re: Sendmail white paper, SRS, and forwarding

On Sun, 2004-11-21 at 17:11 -0500, Chuck Mead wrote:
> David Woodhouse wrote:
> > On Sun, 2004-11-21 at 16:47 -0500, Chuck Mead wrote:
> >
> > > But if SPF is doing what I want why does that mean it's broken? I am 
> > > implementing for my own domain which I have every right to control.
> >
> > That's nice dear. And you also have every right to tell the list servers
> > they may not use your domain in the From: header when you've sent mail
> > to the list. And they have every right to laugh at you for that too.
> But the list servers pass my mail just fine.
That's not the point. I have a _right_ to declare that the list servers
may not send mail bearing a From: header with my name and address in it.
You have the right to ignore me when I make such a silly request.

As you say, you do indeed also have the right to declare that mail hosts
to which you send mail may not send that mail on, intact, to its
ultimate destination.

You also have the right to wear your underpants on your head if you so
desire. You have the right to actually send mail from 
csm(_at_)moongroup(_dot_)com
from a host which is _not_ included in your SPF record. You have the
right to run a mail server which rejects all 'MAIL FROM:<>' and hence is
utterly broken and doesn't accept bounces.

You have all of these rights. But that is not the point. I was talking
about what makes _sense_, not what is within your rights. Do you have
trouble understanding that concept?

> > And you've digressed even further from the point, which was that this
> > brokenness is nothing new and not at all related to SES.
> But your one and only point is to tear down and denigrate SPF by saying 
> it's broken. 
Is English not your first language, or are you otherwise hard of
understanding? I was quite clearly responding to Roger's assertion that
SRS breaks SES. I was pointing out that it's not really SRS which breaks
SPF, it's just part of the normal brokenness of SPF which breaks _all_
mail setups, and it's nothing to do with SES.

My point is that this breakage has nothing to do with SES. That's all.

> I say it's not because it does what I want it to do and it 
> works the way I want it to. If it does that for me must I still consider 
> it broken? According to you apparently so. I simply wish to point out 
> that there are many aside from me who disagree with your point of view.
It was Roger who first used the term 'breaks' in this context, at least
in this thread. I was just pointing out that the breakage is nothing
new.

-- 
dwmw2