[ On Sat, September 13, 1997 at 15:11:15 (+0200), Alan Barrett wrote: ]
> Subject: Re: Reverse IP lookups
>
> If you penalise the user, then presumably the user will do one or more of:
> (A) complain to the vendor of the broken product, (B) switch to a
> non-broken product, or (C) complain to you. If you can prove (by
> reference to the specs) that the product is broken, then hopefully that
> will encourage the user to do A or B rather than to persist with C. I
> claim that that all the above is useful.
I concur 100%.
> In my opinion, the "be liberal in what you accept and conservative in what
> you send" robustness principle should be interpreted as a guide for what
> to do when the specs are ambiguous. If the specs are not ambiguous, I
> believe that refusing to talk to non-conforming implementations is
> usually useful.
Indeed.
However what should we do when the robustness principle gets encoded
into the specifications in such a way that it results in an apparent
contradiction of goals (eg. the SMTP greeting name verification fiasco
where the sender "MUST" send the correct name but the receiver "MUST"
ignore an obviously incorrect name and accept the connection anyway)?
As I've stated before I don't see any problem with a given site
implementing local policy that breaks such rules one way or another.
Yes this will result in a possibly increased level of chaos.
I think I've also stated that I believe these contradictions result in
an invalid specification that needs to be fixed (i.e. I'd like to see
the level of chaos caused by such confustion to be kept to a minimum).
Unfortunately the IETF DRUMS group seems to be repeating these mistakes
even though they're aware of the issues. (Yes I will be participating
in DRUMS as I have time in the future).
--
Greg A. Woods
+1 416 443-1734 VE3TCP <gwoods@???> <robohack!woods>
Planix, Inc. <woods@???>; Secrets of the Weird <woods@???>
--
* This is sent by the exim-users mailing list. To unsubscribe send a
mail with subject "unsubscribe" to exim-users-request@???
* Exim information can be found at http://www.exim.org/
