Hi, Bob.

BL>  BL> Paul's argument for qwk2pkt is that he "knows" what QWK is
BL>  BL> doing, so he just adds SOT/EOT on the end... and incidentally
BL>  BL> ignores my QWK Tearline to add one of his own.

BL>  FM> OK. Is your argument that you can always know, and so SOT/EOT
BL>  FM> is unnecessary?

BL>   There is no such thing as always. There are two unlikely situations
BL> where SOT and EOT would help.

BL>   SOT is useful in Netmail that is neither: international; from a point,
BL> nor to a point (no #1 kludge lines) where someone has written
"AREA: "
BL> at the beginning of their message. I don't know what the odds against
BL> this are.

Doesn't matter what the odds are, it *is* possible and is IMHO
sufficient justification for ^aSOT (at least to be inserted when
required). Paul has quoted a message from someone who wrote some
automatic reply software and, unthinkingly, put exactly that at the
beginning of the message!

BL> It is not the sort of thing you do accidentally, and if a
BL> dickhead wanted to use this method to post mail without an origin, he
BL> could do it better by editing a hex-dump of the packet itself, and
BL> remove his address from the header at the same time.

You certainly could do it accidentally, as the guy did above or by
typing it. I don't believe the SOT/EOT proposal has anything whatever to
do with deliberately corrupting a packet; obviously if you're going to
do that you can do anything you want. Including removing SOT/EOT.

BL>   Other than that, SOT is useless.

You said there were 2 reasons, what was the other?

BL>   And EOT is useless anyway, unless you have an email message with
BL> neither Tearline nor Origin line. But even then, EOT only tells
BL> you that these are missing. If they are missing you already know...
BL> unless the message has several Origin lines and Tearlines. EOT is
BL> only uselful is you have more than one Tearline (and Origin line) but
BL> the correct one on the end is missing. Assuming EOT is there. Which it
BL> won't be in 99% of cases. ROFL!

I think the argument against ^aEOT: is that only a message originator
can put it in reliably. A subsequent processor can't for the sorts of
reasons you mention. And if the originator can put it in, he can also
put in the tear and origin lines, serving the same purpose (even if
there are others in the "user text" earlier on).

BL>   But you gotta laugh.

I think the rationale is sound, based on the "don't let the transport
layer fuck with the content" argument. But given that we *have* FIDO,
we *have* a whole heap of existing software out there which *does* do
that, there's not much point. Specifically, there's no point with ^aEOT:
but ^SOT: should be used at least when it's needed.

Regards, FIM.

 * * Bigamist: One who makes the same mistake twice.
@EOT:

---