[pptp-server] PPTP masquerade && MS non-compliance

[Allan Clark]

Jamin Collins wrote:
> 
> Allan Clark [mailto:allanc at caldera.com] wrote:
> > We were back somewhere around "multiple connections between two boxes,
> > one of which is masquerading for the actual PPTP clients" --
> > and whether this is a good thing for poptop and the masquerading folks
> > to consider.
> >
> > There's a heavy precident in the industry for "that's what the spec
> > says, but we're emulating the non-compliance of those guys".
> > I think we should consider supporting this, **AND** draft an
> > ammendment or revision to the RFC.
> 
> I'm not so sure that is a good route to begin walking down.  Just because
> there is a precedent to not follow the RFC because X other company doesn't
> isn't a good way to do things.  This is equivelant to the whole peer
> pressure question "if everyone else jumped off a bridge would you?".
> 
> We need to consider not just whether someone else did something, but whether
> it is the right thing to do.  For me, it's simple, it's not the right thing
> to do.


You don't explain the reason for your judgment.


When making this "simply not right" judgment, are you working on Logic
("It's not right because <reason 1>  <reason 2>  <reason 3>") or on
faithful compliance ("It's not right because the RFC says so").

Blind faith on a spec that seems to have necessitated non-compliance
should be reviewed.  The RFC is a guideline which should be followed as
best possible.  Even Military services change rules; the RFC might
indeed be wrong.  We've shown a point where it could be incorrect.  RFCs
often fail to concretely describe protocols until a few iterations and
clarifications.

This is why I suggest "fix and amend": if it's wrong, build a reference
source product, and amend the RFC to concretely describe the better
way.  

Allan