On Tue, Aug 11, 2009 at 10:12 PM, Kyle Wheeler<kyle-m...@memoryhole.net> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
>
> On Tuesday, August 11 at 09:34 PM, quoth Patrick Gen Paul:
>> Is there a well-respected mail etiquette, or RFC even, that I could
>> refer him to?
>
> Ah! Why did you say so? I thought you were trying to decide whether or
> not mutt was misbehaving (I wonder what Thunderbird does with this
> fellow's email).
Sorry, I should have mentioned that I was convinced from the word go that
there's nothing wrong with mutt's behavior, since no other email on this list or
any of the thirty I am subscribed to is thus affected.
In any case I would never think of filing a bug report for something
this frivolous.
Was thinking the fellow might start arguing and quite easily may know a
bit more about electronic mail than I do. Call it getting ready for the
onslaught :-)
Funny your mentioning T-Bird, btw, since my correspondent (?) actually
appears to use gnus.
> You may find RFC 822 useful. The relevant header definition is this
> one:
>
> authentic = "From" ":" mailbox ; Single author
> / ( "Sender" ":" mailbox ; Actual submittor
> "From" ":" 1#mailbox) ; Multiple Authors
>
> He's using the first option, which means his email address MUST
> conform to the "mailbox" description, which is as follows:
>
> mailbox = addr-spec ; simple address
> / phrase route-addr ; name & addr-spec
>
> He's chosen the latter form. His name eats up the "phrase" portion, so
> his email address must conform to a route-addr, defined thusly:
>
> route-addr = "<" [route] addr-spec ">"
>
> Since he's not including a route, the thing within the wockas (<>)
> MUST conform to the addr-spec definition, which is:
wockas..?
> addr-spec = local-part "@" domain
>
> The objectionable part of his address is the "local-part", which is
> required to be formed like so:
>
> local-part = word *("." word)
>
> His "local-part" is composed of two words, separated by a period. The
> FIRST word, though, is broken. A word is:
>
> word = atom / quoted-string
>
> The quoted-string definition doesn't apply here (since he isn't using
> quotes), and an "atom" is defined as:
>
> atom = 1*<any CHAR except specials, SPACE and CTLs>
>
> "specials", as you'll note in section 3.3 of RFC 822 are:
>
> specials = "(" / ")" / "<" / ">" / "@" ; Must be in quoted-
> / "," / ";" / ":" / "\" / <"> ; string, to use
> / "." / "[" / "]" ; within a word.
>
> In other words, he's using a colon in an atom, which is explicitly
> forbidden by the definition of email.
>
> Is that sufficiently explicit for you? :)
:-)
> (And, in case you get into an argument with him, addr-spec has not
> changed, even with more recent less-standardized RFCs)
.. I'm planning on politely asking him to desist - the simple fact that
he starts to argue would be quite sufficient to categorize him as killfile
fodder.
> I'm surprised that his email isn't instantly classified as spam.
> Usually it's just spammers that violate basic well-recognized rules
> like that, and it's fairly common to blacklist emails that are
> fundamentally malformed or disobey basic rules like that---especially
> in an age when it's equally common to see software come out with
> updates warning "security fix! malformed input caused cancer in users;
> who would have thought a stray colon would cause that kind of trouble!
> Oh the humanity, how were we poor developers to know that not everyone
> was trustworthy?!?" (or something similar).
That, actually is a much better option than killfiling him locally..!!
If he is not
amenable, I'll contact the list's whip and ask him to ban the guy until he
make amends.
With your permission, I will point my "report as spam" to your exposé
>> I'm giving him another 24 hours before I killfile him for good.
>
> Good luck with that.
Adding a procmail rule that directs his ensuing contributions to my "SPAM"
folder shouldn't be too hard.
Thank you,
Gen-Paul.
